HOC VOLUME1 Chapter14

14 · Itineraries and Geographical Maps in the Early and
Late Roman Empires

o. A. w. DILKE
Whereas Greek knowledge about the theory of map- to reality. Since no ancient handbook of Roman roads,
making-as well as the map image of the known world- illustrated or not, has survived, this is difficult to ascer-
tended to accumulate and to build on previous writers, tain. The bureaucrats' maps attached to the Notitia Dign-
Roman efforts in both small- and large-scale mapping itatum, a directory of officeholders and administrators,
tended to be diluted over time. The record is also ex- follows textual rather than topographical order, some-
tremely fragmentary and drawn out over a period of times with confused results. Even the maps in the Corpus
some five hundred years. In this chapter, we discuss Agrimensorum tended to deteriorate with repeated
itineraries and small-scale geographical maps in both the copying, particularly in their jumbled nomenclature.
early and the late empire and conclude with a review of The late empire is thus often dimissed as of little con-
the use of maps in the Roman period as a whole. sequente by historians of cartography. It is given short
The relative decline in the theoretical aspects of Ro- shrift in the standard authorities; only one or two sur-
man cartography cannot be disputed. Although emp- viving maps such as the Dura Europos shield or the
erors such as Hadrian and Caracalla (M. Aurelius Sev- Peutinger map are described. 2 However, once we start
erus Antoninus) were great philhellenes (and Hadrian's to evaluate a broader spectrum of evidence it becomes
principate-A.D. 117-38-coincided with much of Ptol- clear that the idea of the map was not only kept alive
emy's working life), they do not seem to have encouraged in western and northern Europe, but also transmitted to
Roman scholars to build on the foundations of Greek the eastern empire after the foundation of Constanti-
geography and astronomy. Latin writers such as Pom- nople on the site of Byzantium in A.D. 330. Set against
ponius Mela and the elder Pliny did little to modify an apparent lack of scientific progress in mapmaking,
Hellenistic concepts of the inhabited world, and in com- sound evidence for the widespread use of geographical
parison with Greek writers such as Hipparchus or maps is revealed in literary sources, found in images on
Strabo, the status of maps within their work is relatively coins or incorporated into mosaics, and even seen in the
ambiguous. decoration of lamps. As much as in Hellenistic Greece,
During the late empire, scholars became even further such maps continued to have meaning in Roman society.
removed from the sources of Greek geographical culture
and from the cartographical knowledge it had contained. ITINERARIES AND THE PEUTINGER MAP
It is true that mathematics continued to flourish at Al-
exandria-where Pappus, in the early fourth century, It is widely accepted that measured itineraries are of
not only commented on Ptolemy's Almagest and Plani- fundamental importance in the construction and devel-
sphaerium, but also wrote a Chorography of the opment of geographical maps and marine charts. This
Oikoumene (now lost) based on Ptolemy's Geo- was true of many societies that developed such maps,3
graphyl-but this did not lead to a revision of the maps. and the Roman world is no exception, though a clear
Following the Antonine and Severan dynasties (A.D. 1. Ivor Bulmer-Thomas, "Pappus of Alexandria," in Dictionary of
138-235) there was a period of rapidly changing em- Scientific Biography, 16 vols., ed. Charles Coulston Gillispie (New
perors, and, apart from legal writings, the arts and the York: Charles Scribner's Sons, 1970-80), 10:293-304.
2. Armando Cortesao, History of Portuguese Cartography, 2 vols.
sciences-including the knowledge that related to (Coimbra: Junta de Investiga~6es do Ultramar-Lisboa, 1969-71),
maps-cannot be said to have flourished much. If we 1:148-50; Leo Bagrow, History of Cartography, rev. and en!. R. A.
take accuracy in geographical mapping as a yardstick, Skelton, trans. D. L. Paisey (Cambridge: Harvard University Press;
standards were declining when compared with the high London: C. A. Watts, 1964), 37-38; Gerald R. Crone, Maps and
point of Greek influence. For example, while the Peu- Their Makers: An Introduction to the History of Cartography, 5th
ed. (Folkestone: Dawson; Hamden, Conn.: Archon Books, 1978), 3-
tinger map was of fourth-century origins but indebted 4.
to a first-century A.D. map, the earlier map may have 3. P. D. A. Harvey, The History of Topographical Maps: Symbols,
positioned towns and roads with a closer resemblance Pictures and Surveys (London: Thames and Hudson, 1980), 135-52.
234
Itineraries and Geographical Maps in the Early and Late Roman Empires 235
M £ D
Constantina.
• Tlberlas
r'Sebaslye (Samaria)
eMadabe
Jerusalem
eluslum .
1 Cairo (Babylon)
o 100 200 300 400 miles

I I II II I 'I II
o 100 200 300 400 500 600 km
FIG. 14.1. PRINCIPAL PLACES ASSOCIATED WITH ITIN-

ERARIES AND GEOGRAPHICAL MAPS FROM THE
ROMAN EMPIRE. See also figure 12.1.
distinction must be drawn between written itIneraries vinciarum Antonini Augusti and lmperatoris Antonini
and itinerary maps. In the Roman period the former Augusti itinerarium maritimum. These titles make it
were more common, being used for both military and clear that the journeys mentioned were, in origin at least,
civil purposes-and together with the portable sundial either planned for or completed by an emperor of the
providing a principal aid to the well-informed traveler. Antonine dynasty; and there is general agreement that
The earliest surviving Roman itineraries are the Vicarello this emperor was Caracalla. Since the longest single jour-
goblets, which give a list of stages from Cadiz to Rome ney is overland from Rome to Egypt via the Bosporus,
via the Po valley, with the mileage between successive it seems only reasonable to link this with such a journey
stages. 4 undertaken by Caracalla in A.D. 214-15. 6 A long pres-
The best-preserved examples are the Antonine itin- tigious journey by an emperor would require careful
erary, the Bordeaux itinerary, and the Ravenna cos- planning by civil servants, with provision for supplies,
mography that is associated with the Byzantine empire. changing of horses, and so on, at appropriate staging
The first two of these are simply lists of places along posts. Every contingency had to be foreseen, and local
routes, giving the distances between them, but because
of their close relationship to geographical mapping, all 4. Jacques Heurgon, "La date des gobelets de Vicarello," Revue des
three will be examined alongside the Peutinger map- Etudes Anciennes 54 (1952): 39-50; Raymond Chevallier, Les voies
which remains as the sole surviving example of the itin- romaines (Paris: Armand Colin, 1972),46-49, or for an English trans-
lation, Roman Roads, trans. N. H. Field (London: Batsford, 1978),
erary map from the Roman period, unless we assign the
47-50. O. A. W. Dilke, Greek and Roman Maps (London: Thames
Dura Europos shield to this category. and Hudson, 1985),122-24.
5. Otto Cuntz, ed., Itineraria Romana (Leipzig: Teubner, 1929),
vol. 1, Itineraria Antonini Augusti et Burdigalense; Konrad Miller,
THE ANTONINE ITINERARY
Itineraria Romana (Stuttgart: Strecker und Schroder, 1916), LV H.
and regional sections; Dilke, Greek and Roman Maps, 125-28 (note
The Antonine itinerary, the most important of the an-
4).
cient list-type itineraries to be preserved (as opposed to 6. D. van Berchem, "L'annone militaire dans I'empire romain au
the map or "painted itinerary"), is in two parts: land III' siecle," Bulletin de fa Societe Nationafe des Antiquaires de France
and sea. s The full titles of these are ltinerarium pro- 80 (1937): 117-202.
236 Cartography in Ancient Europe and the Mediterranean
representatives complained that they often had to make miles, sometimes in leagues, sometimes in both (1 1/2 Ro-
such provision at points where in fact the emperor never man miles = 1 leuga).12 The other is that there are in
stopped at all.? Nevertheless, the existence in the An- some cases considerable differences of mileage between
tonine itinerary of forms of place-names later than Cara- the same two places according to which journey is fol-
calla's reign, such as Diocletianopolis for Pella and Her- lowed. More research is needed that will not only study
aclea for Perinthus (Marmara Eregli), suggests that the recorded distances on the modern map, but take
routes were reused, with or without amendment, over account of archaeological and epigraphic evidence, to-
a long period (fig. 14.1). An example of addition is in gether with such geographical factors as alterations in
Sicily, where between Catana (Catania) and Agrigentum sea level or in the course of riverbeds. 13
(Agrigento) two routes are given, the second including
the phrase mansionibus nunc institutis (by the staging ITINERARIES FROM THE LATE EMPIRE
posts now set up). The date of the final version of the
Antonine itinerary may have been between A.D. 280 and The vast extent of the empire, with its expansion of the
290. bureaucracy, encouraged the production of many itiner-
The organization that planned such journeys was the aries, which, since roads with milestones continued to
cursus publicus, 8 set up by Augustus for transporting be kept up, provided acceptable accuracy.14 As barbar-
officials and their families and for carrying official mail. ians pressed in from north and east, military require-
Hence the cursus publicus had its own lists, and in some ments became more important then ever. Vegetius, the
cases straightforward journeys may well have been cop- civil servant whose military manual dates from about
ied directly from these. But the Antonine itinerary cannot A.D. 383-95 but draws on much older material, writes
simply have been a version of those lists, because of the of the ideal general:
numerous omissions, duplications, and extremely
In the first place, a commander should have itineraries
roundabout routes. Thus the Peloponnese, Crete, and of all the war zones very fully written out, so that he
Cyprus are unrepresented, and considerable parts of may thoroughly acquaint himself with the intervening
Gaul, the Balkans, and Asia Minor are thinly covered. terrain, as regards not only distance but standard of
A good example of a circuitous route is the second jour-
7. Dio Cassius Roman History 78.9.3 and 78.9.6-7; see vol. 9 of
ney in Britain, iter II, which reaches Richborough from Dio's Roman History., 9 vols.., trans. Earnest Cary, Loeb Classical
Birrens via Carlisle, York, Chester, and London. 9 Such Library (Cambridge: Harvard University Press; London: William
a route must have been tailor-made for a particular jour- Heinemann, 1914-27).
ney, stopping at the legionary fortresses of York and 8. References in Konrat Ziegler and Walther Sontheimer., eds., Der
kleine Pauly, 5 vols. (Stuttgart: Alfred Druckenmuller., 1964-75), s.v.
Chester among other places.
"cursus publicus," and in Oxford Classical Dictionary, 2d ed., s.v.
The method in the Antonine itinerary was to list the "postal service (Roman)."
starting and finishing points of each journey and the 9. A. L. F. Rivet and Colin Smith, The Place-Names of Roman
total distance in Roman miles (in Gaul, leagues, as men- Britain (Princeton: Princeton University Press, 1979), 157-60, fig. 12.
tioned below). Then the individual stages were listed, 10. Rivet and Smith, Place-Names, 148-82 (note 9).
with the mileage for each. The totals sometimes corre- 11. Warwick Rodwell., "Milestones, Civic Territories and the An-
tonine Itinerary," Britannia 6 (1975): 76-101.
spond to the added individual mileages, sometimes do 12. Gallic leagues were officially recognized by Septimius Severus
not; in the latter case especially, one or more of the about A.D. 202.
figures may well be corrupt. 13. Francis J. Carmody, La Gaule des itineraires romains (Berkeley:
The Antonine itinerary begins at Tangier and covers Carmody, 1977), contents himself with criticizing Miller's early iden-
most of the provinces of the empire rather unsystemat- tification of Vetera, Germania Inferior, with Rheinberg and himself
proposing Alpen. But in fact there has been no dispute for many years
ically. The British section, last before the sea routes, is that Vetera was quite near the previous station, Colonia Traiana (Xan-
self-contained and consists of fifteen journeys, some co- ten). There may have been two successive sites, one three kilometers
inciding in the same or the opposite direction. 1o Except southeast of Xanten, on the Furstenberg north of Birten, the other on
in cases where they are clearly corrupt, the mileages are land that in Roman times was on the opposite bank of the Rhine,
fairly reliable. It has been shown, however, that distances east-northeast of Vetera I. See H. von Petrikovits, "Vetera," in Paulys
Realencyclopadie der classischen Altertumswissenschaft, ed. August
from a settlement sometimes start from the center, some- Pauly, Georg Wissowa, et al. (Stuttgart: J. B. Metzler, 1894-), 2d ser.,
times from the outskirts. 11 Since Colchester is in one 8 (1958): cols. 1801-34, esp. 1813-14.
place called Camulodunum, in another Colonia (it was 14. The emperor's name on a milestone often indicates substantial
one of the four colonies of Roman Britain), one may repair or improvement to a road rather than a new one. For milestones
suspect that the routes were not all contemporary. see the series Itinera Romana, ed. Gerald Sicilia Verbrugghe, Ingemar
Konig, and Gerold Walser, 3 vols. (Bern: Kiimmerly und Frey, 1967-
An interpretation of the Antonine itinerary routes in 76); Jeffrey P. Sedgley, The Roman Milestones of Britain: Their
northern Gaul encounters two difficulties. One is that Petrography and Probable Origin, British Archaeological Reports no.
distances in Gaul are sometimes reckoned in Roman 18 (Oxford: British Archaeological Reports, 1975).
roads, and may study reliable descriptions of short- Synesius of Cyrene, about 370-413. Among his sources
cuts, deviations, mountains, and rivers. In fact, we he mentions the Geography of "the most divine and wise
are assured that the more careful commanders had, Ptolemy," whose coordinates he clearly edited. 22 He ac-
for provinces in which there was an emergency, itiner- cepts the size of the earth according to Ptolemy, not
aries that were not merely annotated but even drawn
Eratosthenes' measurement of the circumference of the
out in color [picta], so that the commander who was
earth. The surviving parts of his Periplus maris exteri23
setting out could choose his route not only with a
mental map but with a constructed map to examine. is cover the southern coasts of Asia (which he may well
have illustrated by a map based on Ptolemy's coordi-
Itineraries were also used by pilgrims and by soldiers nates) and the coasts of the less familiar parts of Europe;
rejoining their legions; they were expected to take good some, such as the Iberian Peninsula, are covered in
care of them, not to leave the route, and to stop at the
mansiones (staging posts) indicated. 16
15. Vegetius De re militari (Military Institutions of the Romans)
The official recognition of the Christian church in A.D. 3.6, author's translation. No annotated itineraries are extant, but from
313 affected cartography as it did other branches of Vegetius's previous words they must have commented among other
science; one direct result was that pilgrimages to Chris- things on quality of road surface. The words "there was an emergency"
tian shrines created a new use for geographical itiner- render the past tense gerebatur, which, however, is a conjecture for
the present geritur by Carl Lang in his edition of Vegetius, Epitoma
aries. The principal itinerary of the late empire is that rei militaris, 2d ed. (Leipzig: Teubner, 1885), 75. Pascal Arnaud (crit-
of the Bordeaux-Jerusalem pilgrimage, A.D. 333, of icism at a seminar 12 December 1983 on "La Tabula Peutingeriana
which the best manuscript is the Pithoeanus, now Par. et Ie Corpus Agrimensorum" at Centre Jean Berard, Institut Fran<;ais
Lat. 4808, of the ninth century (no maps).1? Distances de Naples) takes picta differently as referring to itineraries with paint-
are recorded in leagues (2.22 km) as far as Toulouse, ings.
16. Saint Ambrose (bishop of Milan) Expositio in psalmum 118
then in Roman miles (1.48 km). Another itinerary at- 5.2; see Opera, pars quinta (V): Expositio psalmi CXVIII, ed. Michael
tached to this records a journey from the Holy Land to Petschenig in Corpus Scriptorum Ecclesiasticorum Latinorum, vol. 62
Chalcedon (Kadikoy), in Asia Minor opposite Constan- (Vienna: F. Tempsky; Leipzig: G. Freytag, 1913), 82-83.
tinople, and back via Nicomedia (Izmit), Ancyra (An- 17. Cuntz, Itineraria Romana (note 5); P. Geyer and Otto Cuntz,
kara), Tarsus, and Tyre. A third goes from Heraclea "Itinerarium Burdigalense," in Itineraria et alia geographica, in Corpus
Christianorum, Series Latina, vols. 175 and 176 (1965), 175 :XVIII-
Pontica (Eregli) via Macedonia, Albania, and the east 26; Henri Leclercq, "Itineraires," in Dictionnaire d'archeologie chre-
coast of Italy. In addition to this major document, there tienne et de liturgie, 15 vols., ed. Fernand Cabrol and Henri Leclercq
are fragments of itineraries from monumental inscrip- (Paris: Letouzey et Ane, 1907-53), 7.2 (1927): cols. 1841-1922, esp.
tions of various periods from several territories under 1853-58; Aubrey Stewart, trans., Itinerary from Bordeaux to Jeru-
the Roman Empire. 18 salem: "The Bordeaux Pilgrim," Palestine Pilgrims Text Society, vol.
1, no. 2 (London: Palestine Exploration Fund, 1896), 15 H., abstracted
A second category of written itineraries relates to jour- by George Kish, ed., A Source Book in Geography (Cambridge: Har-
neys made by sea. It has already been noted that part vard University Press, 1978), 156-58.
of the Antonine itinerary consisted of an itinerarium 18. Wilhelm Kubitschek, "Itinerarien," in Realencyclopiidie 9
maritimum, and in view of theories about an association (1916): cols. 2308-63, esp. 2314 ff. (note 13); Miller, Itineraria Ro-
between these periploi and the development of portolan mana, LIV ff. (note 5); Leclercq, "Itineraires," cols. 1841 H. (note 17);
Annalina Levi and Mario Levi, Itineraria picta: Contributo allo studio
charts,19 such itineraries have been widely discussed in della Tabula Peutingeriana (Rome: Erma di Bretschneider, 1967), 27-
the literature of the history of cartography. An anony- 28, n. 29. Two of these are on pilasters, one on a column, others on
mous and incomplete Greek periplus of about the third stone or terra-cotta tablets. An inscription from Solin, Yugoslavia, the
or fourth century A.D. is known as the Stadiasmus maris town near Diocletian's palace at Split, lists four roads leading from
magni. 2o It records distances in stades between harbors there. Of three Gallic inscriptions, one at Autun lists part of the road
from there to Rome; one from Valence refers to the road to Vienna
and watering facilities around most of the eastern Med- (Vienne, France); and one in Luxembourg covers part of the area
iterranean, covering the North African coast as far west between there and Mainz. A columbarium fragment at Vigna Codini
as Utica. Thus a fair amount of detail is given in the is thought to refer to the main road from Cilicia or Syria to Rome.
entry for even such a small area as Djerba Island, Tun- 19. See chapter 19 in this volume, "Portolan Charts from the Late
isia. Rhodes is particularly well covered, with sea dis- Thirteenth Century to 1500," pp. 371-436. The latest writer on peri-
ploi, Pietro Janni, La mappa e il periplo: Cartografia antica e spazio
tances to twenty-seven harbors of the eastern Mediter- odologico, Universita di Macerata, Pubblicazioni della Facolta di Let-
ranean and Aegean. But some areas, such as the Levant, tere e Filosofia 19 (Rome: Bretschneider, 1984), rightly stresses the
are very poorly covered. Muller questionably attributed lack of maps in periploi.
the whole work to a much later period. 21 20. See Karl Muller, ed., Geographi Graeci minores, 2 vols. and
It is likely that writers in some ports specialized in the tabulae (Paris: Firmin-Didot, 1855-56), 1:427-514.
21. Muller, Geographi Graeci minores, l:cxxiii-cxxviii (note 20).
production of these aids to the mariner, and such was 22. Marcianus Periplus maris exteri in Muller, Geographi Graeci
Marcianus of Heraclea Pontica, a Greek writer of peri- minores, 1:515-62, quotation on 516 (note 20), author's translation.
ploi, who is thought to have been a contemporary of 23. See note 22 above.
greater detail than others. For southern Asia he gives The proportions of the Peutinger map are such that
fuller measurements as far as Gedrosia (Pakistan west distances east-west are represented at a much larger scale
of the Indus), then in less detail. than distances north-south; for example, Rome looks as
though it were nearer to Carthage than Naples is to
THE PEUTINGER MAP
Pompeii. The archetype may well have been on a papyrus
roll, designed for carrying round in a capsa (roll box).
The road map known as the Peutinger map,24 Codex As such, its width would be severely limited, whereas
Vindobonensis 324, was originally a long, narrow its length would not. In the extant map a north-south
parchment roll, 6.75 meters long but only 34 centimeters road tends to appear at only a slightly different angle
wide. It is in the Nationalbibliothek, Vienna, and has from an east-west one, and distances are calculated not
been divided into sections for preservation. Its date of by the map's scale but by adding up the mileages of
transcription is twelfth or early thirteenth century, but successive staging posts.
it has long been recognized as a copy of an ancient map. The date of the archetype is likely to have been be-
In his will of 1508, the humanist Konrad Celtes of tween A.D. 335 and 366. Such dating is suggested by the
Vienna left to Konrad Peutinger (in whose hands it had three personifications placed on Rome, Constantinople
been since the previous year) what he called Itinerarium (labeled Constantinopolis, not Byzantium), and Anti-
Antonini. This was not justified as a title: it is indeed a och; and it fits in well enough with biblical references
road map, but not connected with the Antonine emper- on the map. Sometime after the foundation of Constan-
ors and different from the Antonine itinerary described tinople in A.D. 330 as a new Rome on the site of By-
above. It was first published in 1598 by Markus Welser, zantium, Antioch was recognized as the important bas-
a relative of the Peutingers, and since 1618 it has gen- tion against the Parthians. But the suggestion that this
erally been known as the Tabula Peutingeriana or trans- fourth-century archetype was based on a much earlier
lations of that phrase. 25 The original roll at the time of map would account for the inclusion of Herculaneum,
its transcription in the early Middle Ages was of eleven Oplontis, and Pompeii, which had been destroyed in the
sheets, but as such it was incomplete, since much of eruption of Vesuvius in A.D. 79 and not rebuilt, except
Britain, Spain, and the western part of North Africa were for parts of Pompeii. It is also perhaps easier, on this
already missing at the time of copying; there may also supposition, to see why certain roads are omitted, such
have been an introductory sheet forming part of an ear- as the major routes through the Parthian empire men-
lier prototype version. It was evidently not, as was once tioned in the Mansiones Parthicae (Parthian stations) of
thought, the work of the Dominican monk Konrad of Isidorus of Charax. This work is believed to have been
Colmar, who in 1265 quite independently produced a compiled in the late first century A.D. 28
mappamundi that he says he copied onto twelve parch-
ment pages; the paleography suggests an earlier date. 24. Ekkehard Weber, ed., Tabula Peutingeriana: Codex Vindobo-
The second sheet of the Peutinger map was treated as if nensis 324 (Graz: Akademische Druck- und Verlagsanstalt, 1976);
it had been the first, with spellings of truncated names Konrad Miller, Die Peutingersche Tafel (Stuttgart: F. A. Brockhaus,
containing false initial capitals (for example, Ridumo 1962); Levi and Levi, Itineraria picta (note 18); Luciano Bosio, La
Tabula Peutingeriana: Una descrizione pittorica del mondo antico, I
for what was originally Moriduno). Hence a total of
Monumenti dell'Arte Classica, vol. 2 (Rimini: Maggioli, 1983). Dilke,
twelve sheets extant at the time of copying can be ac- Greek and Roman Maps, 113-20 (note 4).
counted for only by assuming that, when the copyist 25. Tabula is one Latin word for a map, but forma is more common.
mentioned this number, he was including a title sheet. 26 Inexplicably, the word tabula has been translated "table" rather than
The Peutinger map was primarily drawn to show main "picture" or "map" in popular usage. It is now time to call it the
"Peutinger map" to avoid any misconception that the original image
roads, totaling some 70,000 Roman miles (104,000 km), was somehow carved on a table or was like a statistical table. The
and to depict features such as staging posts, spas, dis- alternative naming of the Peutinger map as the "world map of Cas-
tances between stages, large rivers, and forests (repre- torius" has met with very little support. Castorius, a geographical
sented as groups of trees). It is not a military map, though writer of the fourth century A.D., is several times mentioned as a source
it could have been used for military purposes, but the in the Ravenna cosmography; but there is no evidence to link him
directly with the Peutinger map.
words of Vegetius quoted above (pp. 236-37) give an 26. During this century preservation has been a major problem,
indication of its possible function. They suggest that, particularly since the green coloring used on the parchment has re-
whether or not the term itinerarium pictum (painted sulted in deterioration of the sea portions of the map. Photographing
itinerary), was in current use, it is a convenient phrase maps through the glass covers used for preservation has produced
for this unique map.27 The distances are normally re- inaccurate colors in some reproductions.
27. Hence the title of Levi and Levi, Itineraria picta (note 18). For
corded in Roman miles, but for Gaul they are in leagues, an alternative suggestion, see end of note 15.
for Persian lands in parasangs, and for India evidently 28. For a discussion of the date and the significance of Isidorus of
in Indian miles. Charax, see Sheldon Arthur Nodelman, "A Preliminary History of
Around the personification of Rome-a female figure were distinguished by no more than a name (fig. 14.4).
on a throne holding a globe, a spear, and a shield-are Attempts to differentiate between types of settlements
twelve main roads, each with its name attached, a prac- on the map and to establish criteria for the attribution
tice not adopted elsewhere (plate 5). The Tiber is cor- of signs have not been entirely successful. Certain im-
rectly shown with 90 percent of the city on its left bank. portant cities are shown with walls: Aquileia, Ravenna,
But owing to the personification the city surround is Thessalonica (Salonika), Nicaea (Iznik), Nicomedia, and
formally shown as a circle, enlarged in proportion to the Ancyra. But why should the triple-gable sign appear only
very narrow width of the Italian peninsula. The Via at Forum Iulii (Frejus), Augusta Taurinorum (Turin),
Triumphalis is indicated as leading to a church of Saint Luca (Lucca), Narona (on the Neretva River), and Tomis
Peter; the words ad scm [sanctum] Petrum are given in (Constanta) ?34 It is interesting to see that, just as there
large minuscules on the medieval copy. Ostia is shown is one personification in the West and two in the East,
with a harbor occupying about one-third of a circle, in so two cities of the second rank, symbolically given
a fashion similar to that of miniatures in the early man- walls, are in the West and four in the East. Important
uscripts of Virgil's Aeneid. 29 Constantinople is repre- cities like Carthage, Ephesus, and Alexandria are not
sented by a helmeted female figure seated on a throne shown with a distinctive sign.
and holding in her left hand a spear and a shield (fig. The road network is thought to have been based (at
14.2). Nearby is a high column (rather than a light- least within the empire) on information held by the cur-
house)3o surmounted by the statue of a warrior, pre- sus publicus, responsible for organizing the official trans-
sumably Constantine the Great. Antioch has a similar port system set up by AugustuS. 35 This system, extended
female personification, perhaps originating in a statue under the late empire to troop movements, relied very
of the Tyche (fortune) of the city, together with arches largely on staging posts at more or less regular intervals;
of an aqueduct or possibly of a bridge. Nearby is the couriers traveled an average of fifty Roman miles (74
park of Daphne, dedicated to Apollo and other gods and km) a day.
famous for its natural beauty and as a leisure center (fig. The part of the British section of the Peutinger map
14.3). Even though the temple of Apollo was burned that survives is so fragmentary that it covers only a
down in 362, there were many other temples, so that limited area of the southeast, not even including London,
this is not necessarily a guide to the dating. It has been and an even smaller area around Exeter. 36 Colchester,
claimed that in A.D. 365-66 all three personified cities surprisingly, is given no cartographic sign. The most
were important, since the pretender Procopius had his northerly place extant in Britain appears as "Ad Taum";
seat of power in Constantinople, Valentinian I in Rome,
Characene," Berytus 13 (1960): 83-121, esp. 107-8, and Fergus Mil-
and his brother Valens in Antioch. 31 But in fact, although
lar, "Emperors, Frontiers and Foreign Relations, 31 B.C. to A.D. 378,"
Valens set out for Antioch, he was diverted to fight Britannia 13 (1982): 1-23, esp. 16.
Procopius and he cannot be correctly associated with 29. For example, Rome, Biblioteca Apostolica Vaticana, Vat. Lat.
that last-named city.32 3225, of about A.D. 420; the importance of its miniatures for the
Throughout the map, mountains are marked in pale history of cartography has been recognized.
brown and principal rivers in green. Names of countries 30. The latter is the interpretation of Levi and Levi, Itineraria picta,
153-54 (note 18).
and of some tribes are recorded. Apart from the per- 31. Levi and Levi, Itineraria picta, 65 H. (note 18).
sonifications, cartographic signs include representations 32. Glanville Downey, A History ofAntioch in Syria: From Seleucus
of harbors, altars, granaries, spas, and settlements. A to the Arab Conquest (Princeton: Princeton University Press, 1961),
unique sign is that for a tunnel @ , used for the Crypta 399-400.
Neapolitana, near Pozzuoli. Harbors, if indicated, are 33. The closest parallel to the use of this convention is in a work
of doubtful authenticity, the so-called Bellori picture, found on the
given the arcuate shape mentioned in connection with Esquiline in 1668 but now lost. Which settlement was intended to be
Ostia. The sign for a spa is an ideogram of a roughly represented on that veduta prospettiva is uncertain; it may have been
square building with an internal courtyard, often with Pozzuoli. See P. S. Bartoli's drawing in Giovanni Pietro Bellori, Ichno-
a gabled tower at each end of the near side. There are graphia veteris Romae (Rome: Chalcographia R.C.A., 1764), 1. Ichno-
fifty-two such buildings represented, of which twenty- graphia is a word Vitruvius used for the drawing of a ground plan in
De architectura 1.2.2; see On Architecture, 2 vols., trans. Frank Gran-
eight are at places specifically called Aquae; in some ger, Loeb Classical Library (Cambridge: Harvard University Press;
other cases there is reason to think that a place so de- London: William Heinemann, 1931-34).
noted had prominent baths. 33 There are also in the Peu- 34. Levi and Levi, Itineraria picta, 92-93 (note 18), compare this
tinger map places with cartographic signs for granaries, with an incised glass beaker depicting seaside houses with triple gables
in the Bay of Naples (New York, Metropolitan Museum): R. W. Smith,
denoted as rectangular roofed buildings. One such is
"The Significance of Roman Glass," Metropolitan Museum Bulletin
Centumcellae (Civitavecchia), which had a corn-im- 8 (1949): 56.
porting harbor of some size. Variants of a two-gabled 35. See above, p. 236 and n. 8.
building were used to depict some settlements, but most 36. Rivet and Smith, Place-Names, 149-50 (note 9).
FIG. 14.2. THE PEUTINGER MAP: WESTERN ASIA ~.izeof the original: 33 x 56.3 em. By permission of the
MINOR AND EGYPT. The elongated deformation of the map Osterreichische Nationalbibliothek, Vienna (Codex Vindo-
is shown by the ribbonlike representation of (top to bottom) bonensis 324, segment VIII).
the Gulf of Azov and the Black, Aegean, Mediterranean, and
Red seas. Constantinople is named as such and not as Byzan-
tium, which confirms the pre-fifth-century date of the arche-
type. Its personification is in the form of a female warrior,
enthroned with a shield and spear. Close by are a column and
a statue, presumably of Constantine the Great.
but it is very far removed from the river Tay. This name,
however, really consists of the ends of [Ven]ta
[lcenor]um (Caistor Saint Edmund, Norwich), and the
~
only unusual feature is ad, which may have belonged to
an adjacent name.
~
One of the important features of the map is that it
records so many small places. This can be well illustrated
by a name in Italy otherwise recorded only (in corrupt
form) in the Ravenna cosmography. On the Gulf of
Naples, marked as being six Roman miles from Her-
culaneum and three miles each from Pompeii and Stabiae
(Castellammare di Stabia), is shown a large building with
the name Oplont(i)s. Until recently scholars could not
place this name, like a number of others. But since 1964
a large palace, which probably belonged to Nero's em-
press Poppaea, has been excavated at Torre Annunziata, FIG. 14.4. TOWN SIGNS ON THE PEUTINGER MAP. Most
and it seems to authenticate the detail on the map.3? Or towns are shown only by a name, but the more prominent are
represented by signs.
37. Alfonso de Franciscis, "La villa romana di Oplontis," in Neue After Annalina Levi and Mario Levi, Itineraria picta: Contri-
Forschungen in Pompeji und den anderen vom Vesuvausbruch 79 n. buto allo studio della Tabula Peutingeriana (Rome: Erma di
Chr. verschutteten Stiidten, ed. Bernard Andreae and Helmut Kyrieleis Bretschneider, 1967), 197-201.
FIG. 14.3. THE PEUTINGER MAP: THE EASTERN MEDI- ~.izeof the original: 33 x 61.4 em. By permission of the
TERRANEAN. The north-south axis of Syria and the Holy Osterreichische Nationalbibliothek, Vienna (Codex Vindo-
Land is here shown parallel to Asia Minor and Cyprus. The bonensis 324, segment IX).
prominent city to the right of Cyprus is Antioch, the third of
the cities on the map personified as a seated figure with spear
and shield.
again, a much earlier discovery near Aquileia in 1830 wandered for forty years guided by Moses), and there
appears to correspond to an entry on the Peutinger map. are other biblical references. There is also an area in
A large bathing establishment, mentioned also by the central Asia labeled Hie Alexander responsum aeeepit
elder Pliny, was discovered on the lower reaches of the usq[ue] quo Alexander (Here Alexander was given the
river Isonzo. 38 This is probably the place given the car- oracular reply: "How far, Alexander?"). Perhaps these
tographic sign for a spa, with the words Fonte Timavi ampler descriptions, whether Christian or pagan, were
(spring of the river Timavus). Its fresh waters by the sea added on otherwise empty space about the fifth or sixth
were regarded as an unusual phenomenon and obviously century A.D. In several areas research is in progress com-
worth mapping. bining fieldwork with study of the Peutinger map and
Owing to the shape of the map, the Nile could not of the history of place-names. One such is the area be-
be represented as a long river if it were made to flow tween the Gulf of Aqaba and Damascus. 39 A question
northward throughout its course. Instead it is made to that emerges is the extent to which we can argue from
rise in the mountains of Cyrenaica and to flow "east- silence: Does the absence of an important road on the
ward" to a point just above the delta. The delta itself is
shown in less compressed form from south to north than (Recklinghausen: Aurel Bongers, 1975), 9-38; Carlo Melandrino,
most parts of the Peutinger map (see fig. 14.2). The Oplontis (Naples: Loffredo, 1977).
distributaries of the Nile are shown to have many is- 38. Luciano Bosio, La "Tabula Peutingeriana": Una carta stradale
lands, three of them marked with temples of Serapis, romana del IV secolo (Florence: 3M Italia and Nuova Italia, 1972),
16 (text accompanying filmstrip).
three with temples of Isis, while the roads are somewhat
39. Dr. D. L. Kennedy of the University of Sheffield is researching
discontinuous. On the Sinai desert we find the words on behalf of the Aerial Photographic Archive for Archaeology in the
desertum ubi quadraginta annis erraverunt filii Israelis Middle East, and Professor G. W. Bowersock gave a talk on the subject
dueente Moyse (the desert where the children of Israel at the University of London in 1980.
Peutinger map suggest that the mapmaker, perhaps of gustus's division into eleven regions), sometimes alpha-
the fourth century A.D., was relying for less familiar areas betical, though as often in antiquity only the first letter
on an earlier map, of the first or second century, made is necessarily in strict order.43 Only book 2, mainly con-
before such a road was built? cerned with the universe (which Pliny calls a constantly
revolving perfect sphere),44 and books 3-6, covering the
THE LATIN GEOGRAPHICAL MANUSCRIPTS geography of the oikoumene, contain materials of po-
AND THEIR MAPS tential cartographic interest. 45
Many of the geographical manuscripts of Roman origin 40. See Pomponius Mela, De chorographia, ed. Gunnar Ranstrand,
are of less cartographic interest than their Greek coun- Studia Graeca et Latina Gothoburgensia 28 (Gothenburg: Acta Uni-
terparts. Whatever the reasons for this, no continuous versitatis Gothoburgensis; distribution Stockholm: Almqvist och Wik-
tradition of writing in Latin about these subjects took sell, 1971), and a newer edition, Pomponius Mela, De chorographia
libri tres, ed. Piergiorgio Parroni, Storia e Letteratura, vol. 160 (Rome:
root. It is often difficult to say if a Roman author com- Edizioni di Storia e Letteratura, 1984). Also see Nicolaus Sallmann,
posed with a map in front of him or indeed whether a "De Pomponio Mela et Plinio Maiore in Africa describenda discre-
map was drawn at all to illustrate a particular text. And pantibus," in Africa et Roma: Acta omnium gentium ac nationum
in other Latin manuscripts with maps, such as the No- Conventus Latinis litteris linguaeque fovendis, ed. G. Farenga Ussani
titia Dignitatum, the maps suggest that the compilers (Rome: Erma di Bretschneider, 1979), 164-73.
41. Pliny, Naturalis historia, 6 vols. in 3, ed. D. (S. D. F.) Detlefsen
either did not have access to such standard maps as those (Berlin: Weidmann, 1866-82; reprinted Hildesheim: Georg Olms,
of Agrippa or Ptolemy or lacked the cartographic know- 1982); Pliny, Natural History, 10 vols., trans. H. Rackham et aI., Loeb
ledge to exploit these sources properly. Classical Library (Cambridge: Harvard University Press; London: Wil-
liam Heinemann, 1940-63). For a French translation with notes of
book 5, part 1, sections 1-46 (on North Africa), see Pliny, Histoire
LATIN GEOGRAPHICAL WRITERS IN THE EARLY EMPIRE naturelle, ed. and trans. Jehan Desanges (Paris: Belles Lettres, 1980);
of book 6, part 2 (on central and east Asia), see Pliny, Histoire na-
This tendency to ignore maps, even when Greek influ- turelle: L'Asie centrale et orientale:J L:JInde, ed. Jacques Andre and
ence was at its height, is probably shown in the work Jean Filliozat (Paris: Belles Lettres, 1980); and of book 7, Pliny, His-
of Pomponius Mela (fl. A.D. 37-42), one of the few Latin toire naturelle:J livre VII, ed. and trans. Robert Schilling (Paris: Belles
geographical writers from the early empire whose text Lettres, 1977).
42. These include the Trophee des Alpes at La Turbie, quite wrongly
has come down to us. Mela was born in southern Spain.
located in the Loeb translation by Rackham, Natural History 2:100,
His Chorographia, in three books, written under Gaius note g (note 41).
or Claudius, is a brief world geography, but there is no 43. The eleven regions are conveniently listed in the Oxford Classical
evidence that it ever contained maps.40 Mela's world is Dictionary, 2d ed., s.v. "Italy," sec. 6.
surrounded by seas and divided into two hemispheres, 44. Pliny Natural History 2.64.160 (note 41). After the section
dealing with earth and water (2.66.166), he discusses, in no particular
Asia in the eastern, Europe and Africa in the western.
order, maritime exploration: the Sea of Azov (Palus Maeotis), he says,
From north to south, as in Eratosthenes' poem Hermes certainly exists, but is it a gulf of the ocean or an overflow from it?
and Virgil's Georgics, it is divided into five zones, two Africa had been circumnavigated by Hanno and several others
cold, two temperate, and one hot. (2.67.167 ff., esp. 169). He then discusses sundials and hours of day-
In much the same way, there is relatively little of ex- light throughout the known world (2.77.186 ff.). There is also a sec-
tion on earthquakes (2.81.191) and changes in the coastline; thus at
plicitly cartographic interest in the geographical com-
the harbors of Ambracia and Piraeus the sea receded by ten and five
pendium of the elder Pliny (A.D. 23/24-79), a native of miles respectively (2.87.201); no date is given, and in the case of
Como, who held important offices under Vespasian but Piraeus, at least, the measurement is inconsistent with known topo-
who as admiral of the fleet at Misenum perished in the graphy. A number of islands or mountains are known to have appeared
eruption of Vesuvius. His Natural History in thirty- or disappeared, and towns have disappeared; of these he quotes ex-
seven books was completed in A.D. 77. 41 It is an ency- amples (2.89.202 ff.). Later in the second book he records sea depths,
underground rivers, and other phenomena (2.105.224 ff.). The book
clopedia based on one hundred major and a large num- concludes with some overall measurements: India to Gades is 8,568
ber of subsidiary Greek and Latin authors. For each or 9,818 miles according to Artemidorus and Isidorus (of Charax),
section he lists the major writers he has followed (often respectively (2.112.242 ff.).
quite closely, as we can tell from extant works), and 45. The material is organized thus:
Book 3 Western and central Europe bordering on
much research has been devoted to these sources. Pliny's
the Mediterranean and the northern Atlantic.
information includes both useful up-to-date material and Book 4 Greece and adjacent areas; the Black Sea and
old travelers' tales. Of Latin writers, he mostly quotes adjacent European areas; northern Europe.
Cornelius Nepos (author of-besides biographies-a Book 5 Africa bordering on the Atlantic and Mediterranean,
lost geographical work) and Agrippa. He incorporates including all of Egypt; Asia bordering on the Mediterranean
and Aegean.
information from inscriptions, statistical data, and lists
Book 6 The Black Sea and adjacent Asiatic areas;
of tribes and places in provinces. 42 These lists are some- other parts of Asia; Ethiopia and the upper Nile valley.
times in geographical order (for Italy he follows Au- As is evident from this summary, the Black Sea itself is treated twice.
From these books it is clear that Pliny was a user of mission. Indeed, although there are other manuscripts
maps rather than a contributor to the theory of their with maps, only three main writers-Avienius, Macro-
construction or compilation. Unfortunately, in giving his bius, and Julius Honorius-can be noted under the pres-
sources he does not distinguish between maps and writ- ent heading.53 Only the second of these can definitely
ten geographies. Thus we have to infer his use of maps be said to have written a work containing maps. This
by indirect means such as the cited measurements be- poverty in the Roman tradition may partly reflect the
tween places or his descriptions of countries and cities fact that some of the more important texts containing
in terms of their shapes. With respect to the latter, for cartographic knowledge continued to be written in
example, Italy is not as today described as boot-shaped Greek rather than in Latin. This is shown by the work
but is said to be like an elongated oak leaf, bending to of Agathemerus, whose brief prose manual summarizes
the left at the top and "ending in the shape of an Am- Greek mapmaking up to the first century B.C. 54 Rufus
azon's shield,,46 (following Strabo and Eratosthenes); Festus Avienius,55 a senator from Volsinii in Etruria,
the Peloponnese has the shape of a plane tree leaf; and wrote two geographical works in Latin verse about A.D.
so on. 47 380-400. The first is a general work in hexameters,
There are two passages where the context may be Descriptio orbis terrae, intended as a revision of the
considered fully cartographic. The first is where he crit- Greek work of Dionysius Periegetes. The second, Ora
icizes the length and breadth of Baetica, southern Spain, maritima, contains only 703 iambs but is thought to be
as given in Agrippa's map and approved by AugustuS. 48 incomplete.56
The second is where he says: "There are a number of The commentary by Ambrosius Theodosius Macro-
segments of the earth which we Romans have called bius (fl. A.D. 399-422) on Cicero's Somnium Scipionis
circles, while the Greeks have called them parallels. ,,49 (The dream of Scipio, incorporated in his Republic) con-
He then gives what we may define as seven climata (he tains cosmology and Macrobius's impression of the ap-
does not use the word) from south to north, with lengths pearance of the world. 57 Whereas he accepts Eratos-
of gnomon shadows, numbers of hours in longest days, thenes' sphericity and measurement, he turns to Crates
and principal countries or cities. These range from south-
ern India and the two provinces of Mauretania (the latter 46. Pliny Natural History 3.5.43 (note 41). Shaped like a crescent
but with two curves on the inner side and a promontory between
are in fact at a very different latitude from southern
them.
India) to the north of the Black Sea and Aquitania. Pliny 47. See chapter 10, "Greek Cartography in the Early Roman
attributes these to earlier Greek theory, but he also gives World," esp. pp. 174-75.
three additional zones to the north, which he says later 48. Pliny Natural History 3.1.16-3.2.17 (note 41); see also above,
Greek geographers have added. pp.207-8.
49. Pliny Natural History 6.39.211-20, esp. 212 (note 41), author's
Other evidence is not quite so conclusive; it may de-
translation.
pend on our interpretation of posuere (they have placed). 50. Pliny Natural History 4.21.113-4.22.14 (note 41), author's
Pliny may well be thinking of placing on a map, but we translation.
cannot be sure. Thus in one passage he writes: "Then 51. Pliny Natural History 6.25.95 (note 41), author's translation.
there projects into the sea a promontory with a vast 52. Pliny Natural History 6.32.143 (note 41).
53. There are also several extant geographical manuscripts in Latin
horn, which some have called Artabrum [from Cape
without maps. For example, Vibius Sequester, De fluminibus~fontibus~
Roca to Lisbon] . . . . The distance from here to the lacubus~ nemoribus~ paludibus~ montibus~ gentibus per litteras libellus,
Pyrenees is given by quite a number as 1,250 miles, and ed. Remus Gelsomino (Leipzig: Teubner, 1967), is an alphabetical
they record there a non-existent tribe of Artabres: they gazetteer listing, under separate headings, rivers, springs, lakes, groves,
have placed [posuere] in this area, by a change of let- marshes, mountains, and peoples. Similarly, two accounts are provided
of the boundaries of provinces and of countries outside the empire,
tering, the Arrotrebae whom I mentioned before the
based on Agrippa's map or a revision: Dimensuratio provinciarum
Celtic Promontory [Finisterre].,,50 Similarly in another and Divisio orbis terrarum, both in Geographi Latini minores, ed.
passage, "Others have placed [posuere] the Gedrusi and Alexander Riese (Heilbronn, 1878; reprinted Hildesheim: Georg
Sires over a stretch of 138 miles, then the fish-eating Olms, 1964), 9-14 and 15-23 respectively.
Oritae, who do not speak the Indian language, for 200 54. Agathemerus Geographica antiqua, ed. Jacobus Gronovius (Lei-
den: J. Luchtmans, 1700); and Geographiae informatio, in Miiller,
miles.,,51 We may also feel that Pliny is more likely to
Geographi Graeci minores 2:471-87 (note 20).
have been looking at a map than a book when he com- 55. Avienius, not Avienus, is the correct form: Alan Cameron, "Ma-
pares the Arabian peninsula to Italy not merely for being crobius, Avienus, and Avianus," Classical Quarterly, n.s., 17 (1967):
surrounded by two seas but for having what he calls the 385-99, esp. 392 ff.
. . 52 56. See Avienius, Rufi Festi Avieni Carmina, ed. Alfred Holder
same orIentatIon.
(Innsbruck, 1887; reprinted Hildesheim: Georg Olms, 1965), chap. 3
and chap. 4, which contains both works. For further references on
LATIN GEOGRAPHICAL WRITERS IN THE LATE EMPIRE
the Ora Maritima, see chapter 9, p. 150 and n. 16.
During the late empire, Latin geographical writing seems 57. Macrobius, Commentary on the Dream of Scipio, ed. and trans.
to have been confined to relatively few channels of trans- W. H. Stahl (New York: Columbia University Press, 1952).
of Mallos for the concept of ocean and land masses: in the southeast (plate 6).67 But we know from the metri-
"Separating us from the people of the southern hemi- cal epitaph of a governor that Corinium (Cirencester)
sphere, Ocean flows along the whole extent of the equa- was the capital of Britannia Prima; and it is likely that
tor; again, as its streams branch out at the extremities Maxima Caesariensis had the most important settle-
of both regions, it forms two islands on the upper face ment, London, as its capital, yet this province is placed
of the earth and two on the underside.,,58 Later he adds: on the map somewhere near Lincoln. Probably, if the
"The accompanying diagram will lay everything before civil servant who compiled this list or its official coun-
our eyes" and "from our diagram we shall also under- terpart had gone to the maps department, he could have
stand Cicero's statement that our quarter is narrow at been put right by the comes formarum (director of
the top and broad at the sides.,,59 Macrobius's map of maps), who was under the Rome city prefect and who
the world was circular (2.5.13), with north at the top provided the only record from the Roman world of an
and with cold, temperate, and hot regions, though per- official working for what must have been a civil service
haps its equatorial ocean was narrower in latitude than maps and plans department. 68
is shown in manuscripts and printed editions. The man- The comes Italiae had as part of his command the
uscript maps accompanying the commentary had a region of Italy near the Alps, but this is illustrated in
strong influence on zone maps of the Middle Ages. the Notitia Dignitatum only by a walled hilltop settle-
The Cosmographia of Julius Honorius is an inaccurate ment. Isauria, western Cilicia, is likewise mountainous;
compilation of about the fifth century; only excerpts it is given something approaching a perspective map,
survive. 60 His list starting Seres oppidum, Theriodes op- with south at the top. Mount Taurus, complete with a
pidum . totally confuses settlements, tribes, and wild animal's hindquarters, is in the center, and Tarsus
even rivers, calling them all towns. 61 He selects a few and the Mediterranean are in the background. Meso-
places in northern Italy, partly in topographical order, potamia has the Tigris and Euphrates correctly placed,
partly not, before suddenly switching to Dalmatia. One and also Carrhae, but there is a great deal of confusion.
of the most incorrect entries may be rendered: "The river Thus Amida (Diyarbakir) and Constantina are each in-
Chrysorroas rises in the plains of Syria and flows through 58. Macrobius Commentary 2.9.5 (note 57); see also Kish, Source
Syria, Antioch, and Palestine and the remaining cities of Book in Geography, 140-42 (note 17).
Syria. Its mouth is in the Aegean, where the island of 59. Macrobius Commentary 2.9.7-8 (note 57).
Cyprus is. It runs for 830 miles.,,62 60. Julius Honorius Excerpta eius sphaerae vel continentia; see lulii
Honorii Cosmographia, in Riese, Geographi Latini minores, 24-55
(note 53).
THE NOTITIA DIGNITATUM
61. Julius Honorius Excerpta 6 H. (note 60).
The full title of this work may be translated "directory 62. Author's translation of Julius Honorius Excerpta 10 (note 60).
of officeholders and administration, both civil and mil- 63. Otto Seeck, ed., Notitia Dignitatum (Berlin: Weidmann, 1876;
itary.,,63 There are primary manuscripts at Cambridge, reprinted Frankfort: Minerva, 1962); R. Goodburn and P. Barthol-
Frankfort, Munich, Oxford, and Paris. 64 Some of these omew, eds., Aspects of the Notitia Dignitatum, British Archaeological
Reports, Supplementary Series 15 (Oxford: British Archaeological Re-
are known, and all are thought, to have been copied ports, 1976); Erich Polaschek, "Notitia Dignitatum," in Realencyclo-
from a codex Spirensis (that is, of Speyer cathedral), padie, 17.1 (1936): cols. 1077-116 (note 13).
which was written in the tenth century but disappeared 64. This is the manuscript called Londiniensis by I. G. Maier, "The
in the sixteenth century. They are all illustrated, and the Giessen, Parma and Piacenza Codices of the 'Notitia Dignitatum' with
Munich manuscript has two sets of illustrations. 65 Some Related Texts," Latomus 27 (1968): 96-141; it was presented
to the Fitzwilliam Museum by Professor F. Wormald. Important man-
The main divisions in the work are between the eastern uscripts are: Oxford, Bodleian Library, MS. Canon. Misc. 378, A.D.
and the western empires and between civil and military 1436; Paris, Bibliotheque Nationale, MS. Lat. 9661, fifteenth century;
officials. The official list was kept by the head of the Munich, Bayerische Staatsbibliothek, Clm. 10291, 1542-50; Cam-
civil service in the West, though it is disputed whether bridge, Fitzwilliam Museum, A.D. 1427; and Frankfort, Stadts- und
Universitatsbibliothek, Lat. quo 76, fifteenth century. There are many
the extant work is governmental or an amateur's copy.
other manuscripts, but without maps, the two labeled Tridentinus-
Its date is between 395 and 413, and it may have been Vindobonensis being not at Vienna but at Trento.
revised even later. 65. In 1548 Count Ottheinrich requested the loan of the alt Ex-
Illustrations consist of insignia of officials, personifi- emplar of the codex, but this was denied. Ottheinrich was clearly
cations of provinces, picture maps, and miscellaneous interested in seeing the original illustrations that accompanied the text
items. Many maps are such as only a bureaucrat unfa- and would not accept an untraced copy of them. In 1550 he received
permission for his artist to make tracings-now lost-of the original
miliar with the areas could have produced; and it is illustrations on geoldrenckt pappeir (oiled paper). See I. G. Maier,
considered likely that there has been much change or "The Barberinus and Munich Codices of the Notitia Dignitatum Om-
66
addition. Thus the map of Britain under the heading nium," Latomus 28 (1969): 960-1035, esp. 995-99 and 1024-30.
"vicarius Britanniarum" has five provinces arranged 1, 66. Polaschek, "Notitia Dignitatum" (note 63).
67. Seeck, Notitia Dignitatum, 171 (note 63).
2, 2, from north (top) to south, with "Britannia Prima"
68. Seeck, Notitia Dignitatum, 113 (note 63).
cluded twice, because each has two entries in the list of

military units.
Lower Egypt, sphere of the comes limitis Aegypti, has
the Nile as a single water channel on its map (fig. 14.5).
This is correct for the area around Memphis, but not
for Pelusium; and Thamudeni should be east of the Red
Sea. Upper Egypt, area of the Dux Thebaidos, is partic-
ularly muddled, with Syene (Aswan) shown incorrectly
on a tributary and not as near the frontier as it should
be. Coptos is shown in two different places, because
both a legion and a cavalry unit are listed there under
separate headings. The map follows neither topograph-
ical nor bureaucratic order and even, unlike the text,
misspells and misplaces the famous temple of Philae.
These shortcomings do not mean that the Notitia Dig-
nitatum was a useless document, but in fact the useful
information would have had to be gathered from the
text rather than from the maps.69
MAPS AS DECORATIVE AND SYMBOLIC

IMAGES
From the Roman period, as from the classical world in

general, the cartographic record extends to maps found
as images on diverse artifacts and in other than textual
sources. These maps have often gone unnoticed in the
literature, but they add a further dimension to an un-
derstanding of cartography in the classical period. Ex-
FIG. 14.5. THE NOTITIA DIGNITATUM: LOWER EGYPT.
amples include the well-known map on the Dura Eu- The area controlled by the comes limitis Aegypti is schemat-
ropos shield, the incised and inscribed stone named as ically represented, complete with its major features such as the
the Pesaro wind rose map, and the map images found Nile, Babylon, Memphis, and the Pyramids.
on coins, in frescoes, as part of the design of mosaics, Size of the original: 31 x 24 em. By permission of the Bay-
and even on Roman lamps. While such representations erische Staatsbibliothek, Munich (elm. 10291, foJ. 113r).
are perhaps peripheral to a reconstruction of the history
of scientific mapmaking, they help in assessing the dis- some in ground plan. The Claudian harbor appears in
semination (and extent of understanding) of maps in the plan on a coin of Nero, about A.D. 64 (fig. 14.6).71 Its
Roman period. Like the maps described in poems, they two jetties are shown as arcs of a circle, part of the right
suggest one way the idea of the map (if not the formal jetty perhaps being shown on arches. The harbor is on
knowledge underpinning the construction of maps) was an island, and at its mouth is a lighthouse surmounted
kept alive in a period that lacked scientific innovation by the statue of an emperor; ships are shown entering,
in the accepted sense. Their context, however, is usually leaving, or standing in the harbor. Trajan's harbor, in
an archaeological one, and the chance circumstances of a hexagonal basin with sides of 358 meters, appears in
their survival and discovery, often unrelated to the doc- plan on a coin of Trajan, about A.D. 113 or after. 72 On
umentary record, have led us to organize the material the right are shown warehouses, on the left what is
typologically in terms of the artifacts carrying the map
69. For example, for Britain see Rivet and Smith, Place-Names,
images.
216-25 (note 9). We can plot the forts in northern England under the
Dux Britanniarum, with the title of the army unit stationed at each;
COINS WITH PLANS OF ROME'S HARBOR
we can reconstruct with reasonable probability the names of all the
As earlier in the Greek period, there are Roman examples Saxon Shore forts; and we can read of a gynaeceum (imperial textile
where maps were used emblematically to face coins. 7o factory) at Vema, but the question is whether it was at Vema Icenorum
(Caistor Saim Edmund) or at Venta Belgarum (Winchester).
About three kilometers north of ancient Ostia at the
70. See pp. 158 and 164 above.
mouth of the Tiber, Claudius and Trajan constructed 71. Russell Meiggs, Roman Ostia (Oxford: Clarendon Press, 1960),
new harbors. These were commemorated on coins and pI. XVIIIa.
medallions, some representations being in pictorial form, 72. Meiggs, Roman Ostia, pI. XVIIIb (note 71).
FIG. 14.6. SESTERCE OF NERO SHOWING OSTIA HAR- Diameter of the original: 3.5 cm. By permission of the Trustees
BOR. Issued ca. A.D. 64, this coin shows the Claudian harbor of the British Museum, London (BMC Emp.I, Nero 132).
of Ostia. Two jetties are represented by arcs of circles, together
with a lighthouse surmounted by a statue, Roman ships, and
Neptune.
thought to be a series of buildings (excavated in the of the delta than is to be seen in the Madaba mosaic.
nineteenth century) including a temple, a small theater, Since the pontoon bridge is above the lowest point on
and an atrium; but the detail is far less precise than on the undivided river, one may conjecture that it was be-
the Neronian coin. tween Memphis, the dynastic capital, which was still of
some importance under the Roman Empire, and Babylon
MAPS IN MOSAICS
(Old Cairo). The military trophies would be appropriate
to Babylon, which was fortified as a legionary camp
Although the mosaic map of Byzantine age at Madaba under Augustus, while Memphis was the center for the
has appeared in many popular works on the history of export of wild animals from Egypt to Rome.?5
cartography, it is by no means the earliest map to occur
in this medium. An example dating to the early empire
is found in Ostia, associated with the many mosaic floors
73. Giovanni Becatti, ed., Mosaici e pavimenti marmorei; 2 pts.
that have been preserved there in situ.?3 They are in the (1961); both are vol. 4 of Scavi di Ostia (Rome: Istituto Poligrafico
"Forum of the Corporations," next to the theater, which dello Stato, 1953-).
had a temple of Ceres in the middle and a series of offices 74. Foro delle Corporazioni, Statio 27: Becatti, Mosaici, 74, no.
belonging to trading corporations around the sides. 108 and pI. CLXXXIV (note 73).
From the names of foreign ports within the mosaics it 75. The mosaic at the temple of Fortuna Primigenia at Praeneste
(Palestrina), central Italy, mentioned in chapter 7, note 4 above, also
is possible to reconstruct many of Rome's trade routes. gives a pictorial representation of scenes around the Nile. It has been
Only one of these mosaics is in map form, but unfor- given various dates but is perhaps of the second century A.D. The
tunately it has no inscription (fig. 14.7).74 It shows a inscriptions, in Greek capitals, specify typical Egyptian animals. The
river spanned by a pontoon bridge, with three branches attempt to show that it is a map in oblique perspective does not appeal
that could be either tributaries or distributaries. The to most art historians, and it must remain in the sometimes debatable
borderland between the completely pictorial and the partly carto-
bridge is supported by three vessels, and on each side of graphic. See references in Wilhelm Kubitschek, "Karten," in Realen-
it is a gateway surmounted by military trophies. The cyclopiidie, 10 (1919): cols. 2022-2149, esp. 2023 (note 13); Levi
Nile delta seems the most probable location. Another and Levi, Itineraria picta, 44 n. 65 (note 18). A black-and-white re-
office in the same forum was occupied by an Alexandrian production is found in Moses Hadas, Imperial Rome (Alexandria,
corporation, and corn imports from Egypt were consid- Va.: Time-Life Books, 1979), 70-71. For the Alexandrian tradition
of landscape painting, which may have influenced the Palestrina mosaic
erable. If these two offices are linked, the likely inter- and others, see Roger Ling, "Studius and the Beginnings of Roman
pretation is that the three principal ancient distributaries Landscape Painting," Journal of Roman Studies 67 (1977): 1-16, with
of the Nile are depicted, giving a more simplified version bibliography in note on p. 1, and p. 14 n. 53.
FIG. 14.7. THE OSTIA RIVER MOSAIC. Preserved in the Size of the original: 7 x 3.5 m. From Giovanni Becatti, ed.,
Forum of the Corporations in Ostia, this mosaic without an Mosaici e pavimenti marmorei, 2 pts. (1961); both are vol. 4
inscription depicts a river, with three tributaries or distribu- of Scavi di Ostia (Rome: Istituto Poligrafico dello Stato,
taries, and a pontoon bridge with a military gateway at each 1953-), pt. 2, pI. CLXXXIV (no. 108).
end. Of three possible locations-the Nile delta, the lower
Tiber, and the Rhone delta-the first seems most likely. Museo
Ostiense, near Rome.
What have been recognized as maplike designs, in lines are drawn to the center, where a large depression
addition to those of mazes described below, also occur presumably served to insert a metal base for the flag of
occasionally in mosaics dated to the late empire. Mosaics the wind rose. On the rim, opposite the holes, are the
with views of estates, one category of such images, can names of the twelve winds.
hardly be regarded even as picture maps: the most con- The scheme probably arose from a reading of the
spicuous is the fourth-century mosaic from Carthage, passage from Aristotle mentioned above (pp. 145-46)
now in the Bardo Museum, Tunis, showing the estate without the benefit of a diagram. 82 Aparcias (Aparctias,
of Julius. 76 But one temple scene, also from Carthage, Septentrio) occupies a far larger sector than neighboring
is represented partly in plan. This is a late fourth- or Boreas; the two are identical in Aristotle. This is not
fifth-century mosaic known as the Offering of the important, since in antiquity the number, names, and
Crane. 77 In the center is a shrine facade with Apollo and directions of winds on a wind rose varied. 83 Neverthe-
Diana between two columns; at their feet is the sacrificial 76. Very frequently reproduced: references in Levi and Levi, Itin-
crane, and below it a series of almost concentric squares eraria picta, 68 n. 3 (note 18); Katherine M. D. Dunbabin, The Mosaics
clearly representing the ground plan of this same shrine. of Roman North Africa: Studies in Iconography and Patronage (Ox-
Some mosaics also contain celestial maps in the form ford: Clarendon Press, 1978), 119-21, pI. 109.
of zodiac diagrams. The earliest Palestine zodiac mosaic, 77. Dunbabin, Mosaics, 57-58, pIs. 35-37 (note 76).
78. Michael Avi-Yonah, Ancient Mosaics (London: Cassell, 1975),
for example, thought to be of the third century A.D., is 51-53. For the difference in approach between Christian and Jewish
in the synagogue at Hammath, south of Tiberias. 78 It mosaicists see Dunbabin, Mosaics, 232-33 and n. 174 (note 76), and
represents the twelve signs of the zodiac surrounding the for the zodiac in North African mosaics see her Index IV, S.v. "Zo-
sun, with personifications of the four seasons occupying diac."
the corners. The inspiration of this Hebrew mosaic is 79. Italo Zicari, "L'anemoscopio Boscovich del Museo Oliveriano
di Pesaro," Studia Oliveriana 2 (1954): 69-75; Robert Boker, "Wind-
Greek, with the sun as Phoebus Apollo driving a four- rosen," in Realencyclopadie, 2d ser., 8.2 (1958): cols. 2325-81, esp.
horse chariot, Virgo as a veiled Persephone carrying a 2358-60 (note 13); Antonio Brancati, La biblioteca e i musei Oli-
torch, and Libra as a king, clearly Minos or Rhada- veriani di Pesaro (Pesaro: Banca Popolare Pesarese, 1976), 210-11;
manthys, holding a scepter and balance. Other similar Dilke, Greek and Roman Maps, 110-11 (note 4).
mosaics are of very much later date, when Palestine was 80. Wind rose maps are also found in textual sources. A wind rose,
for example, is preserved in the oldest manuscript of Vitruvius De
part of the Byzantine empire. architectura (note 33), in which it is the only illustration (1.6.4 ff.).
In the text he says it is "so delineated that it is clear where winds
THE PESARO WIND ROSE MAP come from": London, British Library, Harl. MS. 2767; see O. A. W.
Dilke, Roman Books and Their Impact (Leeds: Elmete Press, 1977),
In some cases an unusual physical object carries a map, 26-27, quotation on 26. A possible parallel to the Boscovich ane-
moscope is a small Greek world map to be found in manuscripts of
so that there is a temptation to describe it as "unique" the scholia to Ptolemy's Handy Tables; see Otto Neugebauer, "A
in artifactual terms although its image may also be fa- Greek World Map," in Le monde grec: Hommages a Claire Preaux,
miliar from other sources. Such an object is the "Bos- ed. Jean Bingen, Guy Cambier, and Georges Nachtergael (Brussels:
covich" anemoscope,79 a cylindrical block of Luna mar- Universite de Bruxelles, 1975), 312-17 and pI. III.2. The manuscripts
ble 55.3 centimeters in diameter and 6.8 centimeters in date from the thirteenth century onward, but the map clearly origi-
nated under the Roman Empire, probably in Egypt, since it names
width, on which is engraved a wind rose map (fig. certain southerly places such as Syene (Aswan) and Hiera Sykaminos
14.8).80 It was found near the Via Appia at Rome, out- (above Syene on the Nile). These are places from which astronomical
side the Porta Capena, in 1759. The name is inexact, observations could be made. Ptolemy's instructions for making tables
since the astronomer R. G. Boscovich only helped the were later worked up by Theon of Alexandria (fl. A.D. 364). Unlike
owner, P. M. Paciaudi, in his researches. Now in the the anemoscope, this world map, which is in the form of a rough
sketch, does not have lines linking the wind points to the center. It
Oliveriano Museum at Pesaro, it is thought to date from may go back to an archetype earlier than Theon.
about A.D. 200. It is inscribed Eutropius feci (I, Eutro- 81. But a miniature in a manuscript of Aristotle's Meteorologica at
pius [a Greek name], made it). the Royal Library in Madrid makes the spacing of the zones rather
To construct his wind rose map, Eutropius engraved even by attaching lines to small semicircles inscribed within the
a meridian, divided this equally into six, contrary to perimeter: Charles Graux and Albert Martin, "Figures tirees d'un
manuscrit des Meteorologiques d'Aristote," Revue de Philologie, de
Aristotle,81 and from the dividing points drew five lines
Litterature et d'Histoire Anciennes, n.s., 24 (1900): 5-18.
at right angles to the meridian. These are labeled, in 82. In modern reconstructions of Aristotle's diagram, four out of
descending order: TOTVS INFRA TERRA(M) (the Antarctic five of Eutropius's horizontal lines are visible, though not equally
Circle); BRVMALIS (the Tropic of Capricorn); AEQVI- spaced. Aristotle himself may have inserted only the diameters. Ar-
NOCTIALIS (the equator); SOLI(S)TITIALIS (the Tropic of istotle Meteorologica 2.6; see Meteorologica, trans. H. D. P. Lee, Loeb
Classical Library (Cambridge: Harvard University Press; London: Wil-
Cancer); and TOTVS SVPRA TERRA(M) (the Arctic Circle). liam Heinemann, 1952).
At each end of the meridian and of each line are small 83. Vitruvius De architectura 1.6.4 ff. (note 33), discussed four,
holes intended for bronze pegs. From these twelve holes, eight, and sixteen winds, the last corresponding to Etruscan usage. In
less, the anemoscope must have been intended as a me- are shown staging points, each having as a cartographic
teorological device, partly to help the traveler who, as sign a building with courses of pale green stonework.
he set out from Rome on the Via Appia, would be facing After each place-name was added the number of Roman
south as the map does. The flag would show the name, miles from the previous staging point for one traveling
origin, and direction of the wind. northward and then eastward. That coastal route ran
from Byzantium via Tomis to the mouth of the Danube
and beyond. On two distributaries of the Danube are
written the names Istros and Danubis, whereas properly
speaking Istros was the name given by the Greeks to the
whole of the lower Danube.
After an illegible entry the stages shown, with mileages
where preserved, are: Odessos, Bybona (Byzone), Kal-
latis, Tomea (Tomis) 33, river Histros 40, river Danu-
bios, Tyra 84, Borysthenes (albia is meant), Cherso-
nesos, Trap ... , and Arta. Chersonesos is the Tauric
Chersonese, the Crimea (fig. 14.10). But the following
two have been misunderstood. Trap does indeed stand
for Trapezus, but this is not Trebizond, which is on the
southern shore of the Black Sea, but the "Table Moun-
tain" of antiquity, the Krymskie Gory. Finally, Arta is
not Artaxata, capital of Armenia, which is nowhere near
the Black Sea, but the Latin word arta (narrows) tran-
sliterated into Greek. 85 This must obviously refer to the
Straits of Kerch, where the chief ancient settlement was
Panticapaeum (Kerch). We may therefore consider this
a "painted itinerary" (if that and not a picture is what
Vegetius had in mind by his phrase itineraria pieta) tak-
ing the soldier to Panticapaeum, ruled by a native prince
FIG. 14.8. THE PESARO WIND-ROSE MAP. A diagram of
winds engraved on a cylindrical marble block, probably dating but with a Roman garrison.
from about A.D. 200. There is a central hole for a pole sup- The shield may be regarded then, like the Peutinger
porting a pennant and small holes near the rim for wooden map, as an itinerarium pietum. Unlike the latter, it is
pegs indicating the winds. reasonably orthomorphic, but somewhat oversimplified.
Diameter of the original: 55 em. By permission of the Museo It has some semblance of reliability on the northwest
Archeologico Oliveriano, Pesaro (inv. 3.302).
coast of the Black Sea, and even eastward of that it is
THE DURA EUROPOS SHIELD
more reliable than has been thought. From its orna-
mental character and the use of Greek rather than Latin,
At Dura Europos on the Euphrates a parchment frag- it is likely to have been an unofficial composition.
ment, now housed in the Bibliotheque Nationale, Paris,
was excavated in 1923. 84 On the sheet of parchment, other cases, in addition to the cardinal points, summer and winter
which had covered a soldier's shield, had been painted sunrises and sunsets often appear as directions, though these, as Strabo
observes, vary according to latitude: Strabo Geography 1.2.20-21;
a rough map of the Black Sea and surrounding areas, see The Geography of Strabo, 8 vols., ed. and trans. Horace Leonard
the extant part showing the west and north coasts (fig. Jones, Loeb Classical Library (Cambridge: Harvard University Press;
14.9). Although the wording is in Greek, the shield must London: William Heinemann, 1917-32). See also Strabo, Geographie,
have belonged to a soldier in the Roman army. It is ed. Fran<;ois Lasserre, Germaine Aujac, et al. (Paris: Belles Lettres,
thought to date to shortly before A.D. 260, when the 1966-). But for mainland Greece the summer and winter sunrises,
together with the summer and winter sunsets, make an angle of about
Romans withdrew from Dura Europos. The measure- 30° with the east-west axis; and it is likely that Aristotle situated the
ments of the portion discovered are forty-five (originally winds at 30° from the north-south axis: Aristotle Meteorologica 2.6
about sixty-five) by eighteen centimeters, and it has (note 82).
roughly west-southwest at the top. 84. Bibliotheque Nationale, Gr. Supp. 1354, no. 5. Franz Cumont,
On the left, in blue, is the Euxine (Black) Sea, which "Fragment de bouclier portant une liste d'etapes," Syria 6 (1925): 1-
15 and pI. I; idem, Fouilles de Doura Europos (1922-1923), text and
has two large ships and four heads appearing out of the
atlas (Paris: P. Geuthner, 1926), 323-37, pis. CIX, CX.
water that may be those of sailors from other vessels. 85. Richard Uhden, "Bemerkungen zu dem romischen Kartenfrag-
The shore is indicated by a pale curved line, with no ment von Dura Europos," Hermes 67 (1932): 117-25. The suggested
promontories or indentations. Along the coastal route interpretation of arta given here is the author's.
FIG. 14.9. THE MAP ON THE DURA EUROPOS SHIELD. Size of the original: 18 x ca. 65 em. Photograph from the
Dating to shortly before the Roman withdrawal from Dura Bibliotheque Nationale, Paris (Gr. Suppl. 1354, no. 5).
Europos in A.D. 260, this design was drawn upon parchment
found covering a Roman soldier's shield. It shows the coastal
route from Byzantium to the mouth of the Danube and beyond,
complete with the mileages between staging posts.
o 100 200 miles

PLANS ON LAMPS I----+--r'
a 150 300 km
An unusual series of designs on Roman lamps, incor-

porating rudimentary plans, has been discovered in Pal-
estine, probably datable to the early fourth century
86
A.D. One from Samaria (Sebastye) shows what appears
to be a Roman fort with rooms around the four sides
and with intersecting central roads (fig. 14.11). The cor-
rectness of the identification as a military diagram is
shown by the emblem of two crossed swords on similar
lamps from the same area. A fragmentary lamp from the
Ophel (near Jerusalem's Old City) appears to depict two
L-shaped Roman road stations, and other less certainly
identifiable plans exist. 87
Dura Europos
CLASSICAL PLANS OF MAZES
The maps associated with the artifacts described above

show that they were adapted as motifs in a variety of FIG. 14.10. PRINCIPAL PLACE-NAMES FROM THE DURA
historical contexts. Such maps should be thought of as EUROPOS SHIELD.

different from, rather than historically less interesting
than, the formal maps of the textual sources. They reflect 86. Mordechai Gichon, "The Plan of a Roman Camp Depicted upon
alternative ways of thinking about maps and confirm a Lamp from Samaria," Palestine Exploration Quarterly 104 (1972):
their more general acceptance within Roman society. At 38-58, who also illustrates the lamp from the Ophel.
87. Robert Alexander Stewart Macalister, The Excavation of Gezer,
the same time, this evidence is important in showing
1902-1905 and 1907-1909, Palestine Exploration Fund, 3 vols. (Lon-
how some maps were recurrent motifs in the classical don: John Murray, 1911-12), 3:pl. LXXII, 18.9: perhaps this depicts
world. This is illustrated by a common category of clas- a barracks block.
S em.
I
FIG. 14.11. PLAN ON A ROMAN LAMP. Found at Samaria FIG. 14.12. LABYRINTH DESIGNS ON CRETAN COINS.
(Sebastye), in Palestine, and dating from the early fourth cen- These silver coins show both a square maze (ca. 80 B.C.) and
tury A.D .. The illustration is a view from above. a rarer circular one (300-280 B.C.).
Size of the original: 6.5 X 8.9 x 3.2 cm. After Mordechai Diameters of the originals: 2.3 cm and 3 cm respectively. By
Cichon, "The Plan of a Roman Camp Depicted upon a Lamp permission of the Trustees of the British Museum, London
from Samaria," Palestine Exploration Quarterly 104 (1972): (BMC Cnossus 24 and 41).
38-58, esp. 39 (top).
sical map-the maze-that was associated with several

forms of emblematic representation.
The maze, or labyrinth, a feature well known from 88. "Labirinto," in Enciclopedia dell'arte antica, classica e orientale,
7 vols. (Rome: Istituto della Encidopedia Italiana, 1958-66), W. H.
dynastic Egypt and elsewhere, was particularly associ-
Matthews, Mazes and Labyrinths: A General Account ofTheir History
ated in classical antiquity with the palace of Minos at and Developments (London: Longman, 1922); Janet Bord, Mazes and
Cnossos. 88 Sir Arthur Evans thought it was connected Labyrinths of the World (London: Latimer New Dimensions, 1976);
there with the double ax (labrys) , and a maze pattern Karl Kerenyi, Labyrinth-Studien: Labryrinthos als Linienreflex einer
appears on one of the wall frescoes of the palace. 89 The mythologischen Idee (Zurich: Rhein-Verlag, 1950).
89. Arthur Evans, The Palace ofMinos, 4 vols. (London: Macmillan,
city of Cnossos, proud of its Minoan heritage, minted
1921-35); Matthews, Mazes, 32, fig. 8 (note 88).
coins in the classical period with rough plans of mazes, 90. Jean N. Svoronos (Ioannes N. Sboronos), Numismatique de la
mostly having rectilinear paths, though the circular type Crete ancienne accompagnee de l'histoire, la geographie et la myth-
also exists (fig. 14.12).90 An early artifact showing one ologie de l'Aile, text and plates published separately (Macon: Impri-
is a clay tablet from the palace of Nestor at Pylos in the merie Protat Freres, 1890; reprinted Bonn: R. Habelt, 1972), 65 H.
Peloponnese. 91 On the reverse is a maze, thought to have and pIs. IV-VIII; Warwick Wroth, A Catalogue of the Greek Coins
of Crete and the Aegean Islands, ed. Reginald Stuart Poole (Bologna:
been incised earlier than the obverse, that contains a list A. Forni, 1963), 18-19, pis. 5 and 6; Matthews, Mazes, figs. 20-31
of ten men in the same Linear B script as occurs at (note 88).
Cnossos. Another example from Greece is a tile of the 91. Mabel Lang, "The Palace of Nestor Excavations of 1957: Part
classical period from the acropolis at Athens, where II," American Journal of Archaeology, 2d ser., 62 (1958): 181-91.
The day tablet is Inventory Cn. 1287, discussed on p. 190 and pictured
there had much earlier been a Mycenaean stronghold. 92
in pI. 46.
An Etruscan vase found at Tragliatella, near Rome,93 92. Ernst Buschor, Die Tondiicher der Akropolis (Berlin and Leipzig:
shows among other features infantry performing a dance Walter de Gruyter, 1929-33), 1.45 H., K 108; Hermann Kern, La-
and cavalry next to a circular maze labeled TPVIA (Etrus- birinti: Forme e interpretazione, 5000 anni di presenza di un archetipo
can Truia = Troy). Here we have an allusion not merely manuale e file conduttore (Milan: Feltrinelli, 1981), German ed., Laby-
rinthe: Erscheinungsformen und Deutungen, 5000 Jahre Gegenwart
to an early palace but obviously to the "Game of Troy" eines Urbilds (Munich: Prestel-Verlag, 1982).
that, whether rightly or wrongly attributed to that city, 93. Matthews, Mazes, figs. 133-35 and refs. (note 88).
is described by Virgil as a Roman game having labyrin- 94. Virgil Aeneid 5.580 H.; see The Aeneid of Virgil, trans. Cecil
thine movements. 94 Day-Lewis (London: Hogarth Press, 1952).
One of the best-known Roman specimens is a graffito office, was to save users unnecessary travel. The map
at Pompeii showing a square mosaic with the wording came to be recognized as a legal document, both in land
LABYRINTHUS: HIC HABITAT MINOTAURUS (Labyrinth: survey and for determining the use of aqueducts by prop-
the Minotaur lives here), referring to the Cnossos pal- erty owners. Both writers on water supply, Vitruvius
ace. 95 In fact the most common use of the motif in Ro- and Frontinus, show themselves familiar with maps: Vi-
man times is in mosaics. 96 One such is also at Pompeii, truvius mentions river sources as painted and written
in the Villa di Diomede. Another particularly well pre- about in world cosmographies;101 for the use of aque-
served mosaic at Salzburg shows, in the center of a duct maps as mentioned by Frontinus, see page 232.
square maze, Theseus about to kill the Minotaur. This Another function of mapping is shared by religious
same theme in the center of a circular maze, with re- and legal or surveying aspects: often a tomb and sur-
cumbent towers at the corners, is found on a mosaic at rounding plot were given their measurements on an in-
Cormerod, Switzerland. 97 A partly preserved square mo- scription. An actual plan of these was only very rarely
saic, with a vase and scrolls surrounding the maze, was attached. Such a plan could have helped lawyers in any
found in the churchyard of Caerleon (Isca).98 A lost dispute, such as about whether the land around a tomb
specimen from Sousse (Hadrumetum), Tunisia, included belonged to the heirs. The users of the Forma Urbis
Theseus's ship and the words HIC INCLUSUS VITAM PER- Romae were on the one hand public organizations, on
DIT (one enclosed here loses his life).99 the other hand private individuals living in Rome. It
The early Christian use of maze mosaics can be ex- seems unlikely that builders of new roads normally con-
emplified from the fourth-century basilica of Saint Re- sulted maps, though there was a civil service maps de-
paratus at al-A~nam, Algeria, where the center of the partment under the late empire. 102
square maze has a play on words on SANCTA ECCLE-
SIA.
100 The classical tradition of maze plans continued
MAPS AS STRATEGIC DOCUMENTS
into the Middle Ages; the connection of the labyrinth
with Theseus and the Minotaur, sometimes then mis- Contrary to Lloyd Brown's view, the story that a Phoe-
interpreted as a centaur, was not forgotten in medieval nician captain was publicly rewarded for running his
church ornamentation. ship aground so as not to reveal to a following Roman
ship his route from Gades (Cadiz) to the Cassiterides
THE USE OF MAPS IN THE ROMAN PERIOD gives no indication of the use of maps.l03 It is thought
Much has already been either said or implied about the

widespread use of maps in Roman society. There is more
95. E. Pottier, "Labyrinthus," in Dictionnaire des antiquites grec-
evidence for the use of maps in the Roman period than
ques et romaines, 5 vols., ed. Charles Daremberg and Edmond Saglio
in other periods of antiquity, but the evidence is still (Paris: Hachette, 1877-1919), 3.2:882-83.
largely fragmentary. Such evidence is partly from literary 96. Pliny Natural History 36.19.85 (note 41) says one should not
or technical works and partly from inscriptions. It sug- compare the Cretan labyrinth to "our mosaic pavements or to the
gests that there was an expansion in the use of maps, mazes made in fields to amuse children"; author's translation.
97. C. Bursian, "Aventicum Helvetiorum," Mittheilungen der Anti-
even in periods that lacked marked technical or scientific
quarischen Gesellschaft zu Zurich 16, no. 1 (1867-70), esp. pi. 29;
advances in mapmaking. Thus, as a conclusion to the Matthews, Mazes, fig. 36 (note 88).
period as a whole, an attempt is made here to classify 98. George C. Boon, Isca: The Roman Legionary Fortress at Caer-
the uses of maps in Roman society, while accepting that leon, Mon., 3d ed. (Cardiff: National Museum of Wales, 1972); Mat-
there is clearly much overlap. thews, Mazes, 42 (note 88).
99. Salomon Reinach, Repertoire de peintures grecques et romaines
(Paris: E. Leroux, 1922), 214.1, fig. 1.
MAPS AS CADASTRAL AND LEGAL RECORDS 100. Henri Leclercq, "Labyrinthe," in Dictionnaire d~archeologie
chretienne et de liturgie, 8.1 (1928): cols. 973-82, esp. 974-75 and
It was probably for land survey that Rome produced its fig. 6549 (note 17).
earliest working maps, one lost specimen (pp. 209-10) 101. Vitruvius De architectura 8.2.6 (note 33).
102. See p. 244 above and n. 68 of this chapter.
dating from the second century B.C. The Roman land
103. The story is in Strabo Geography 3.5.11 (note 83). The Cas-
survey treatises mention maps available both in Rome siterides are called a group of islands, and their name is derived from
and locally; these were used among other purposes by the Greek word for tin; they are usually thought to have been the
lawyers (or surveyors experienced in land law) to contest Scilly Islands, but Rivet and Smith, Place-Names, 42-43 (note 9),
property disputes, by emperors to decide boundary dis- prefer to place them outside the British Isles. Lloyd A. Brown, The
Story of Maps (Boston: Little, Brown, 1949; reprinted New York:
putes between local authorities, and for levying central
Dover, 1979), 9, has a misleading account of this episode: in Strabo
or regional taxes. One object of depositing two copies the sea captain is called Phoenician, not Carthaginian, and Strabo does
of survey maps, one in Rome and the other in a local not mention any ship's log or charts.
to refer to the period between the First and Second Punic "shape" or "map." If it is taken to be the latter, the
Wars, 241-218 B.C. If Roman ships' captains wanted sentence may be rendered: "The map of Ethiopia became
to consult a work that would help them navigate outside known, as mentioned, and when recently brought to the
the Straits of Gibraltar, they could perhaps have turned emperor Nero it showed that for 996 miles from Syene,
to the periplus of pseudo-Scylax (p. 383), though in the the boundary of the empire, to Meroe there were few
form in which it has come down to us it contains nothing trees and that all of these were species of palm."109 Since
on sea areas north of Cadiz. trees were often drawn on maps, though sometimes in-
The only extant admonition to the soldier to use maps correct species, palms may have been drawn in as far as
is in the late military writer Vegetius. Since, however, Syene and their dearth deduced from the appearance of
both Julius Caesar and Agrippa, the one a general, the very few upstream from there.
other an admiral, were promoters of maps, there is good Such maps are likely to have been drawn by military
reason to think that the use of these, as well as of itin- surveyors but used with caution by strategists back in
eraries and periploi (see "Maps for Traveling," below) Rome. Certainly, by the second century A.D., educated
was established relatively early. Romans were becoming aware of the limitations of maps
The Augustan period was one in which, perhaps for as a basis for action on the fringes of unexplored ter-
the first time in the Roman world, the use of maps by ritory. Thus Plutarch (ca. A.D. 46 to after 120) dedicates
the man in the street was taken for granted. Rough ones his Parallel Lives to Sosius Senecio, consul four times
could be traced on the spot. The first poem of Ovid's between A.D. 98 and 107, with this simile: "Just as his-
Heroides is supposedly a letter from Penelope to Ulysses, torians, Sosius Senecio, in their geographies squeeze onto
but typically the poet makes his Trojan heroes behave the edges of their maps [pinakes] parts of the earth that
like contemporaries. One such, who has reached home escape their knowledge, with notes explaining 'Every-
after the sack of Troy, describes the course of events as thing beyond is sandy desert with no water or full of
he sketches the Troy area in wine on the table: "This is wild animals' or 'unexplored marsh' or 'Scythian frost'
where the river Simois flowed; this is the land of Sigeum; or 'frozen sea,' so in writing my Lives ... I might as
here stood the high palace of old Priam; that is where well say of prehistory: 'Everything beyond is full of won-
Achilles encamped, that is where Ulysses encamped; this ders.' ,,110
is the point where the mangling of Hector terrified the Such references suggest that the value of maps to the
galloping horses." 104 Propertius (ca. 50 B.C. to ca. 16 state and its generals was widely accepted. Together with
B.C.) also uses the literary format of the letter, but from coins and such monuments as Trajan's column, with its
a Roman lady to her lover far away in the army. His graphic representation of Rome's campaigns on the Dan-
Arethusa spends the winter nights studying maps: "I ube, they could have had great propaganda value. But
learn in what area the river Araxes, which is to be con- there were possible dangers also to an autocracy from
quered, flows, how many miles the Parthian horse runs knowledge that could be extracted from maps by move-
without water. I compel myself to learn painted worlds ments hostile to the imperial system, so that in the wrong
from the map [tabula] ... ; which land freezes up, which hands they could become a threat to security. Domitian,
crumbles from heat, which wind gives good sailing to emperor from A.D. 81 to 96, was by nature autocratic,
Italy."105 Editors mostly treat tabula here as map rather and knowing of previous conspiracies he was quick to
than picture, though the expression may include both a
map and its commentary.
The extent to which Roman expeditions carried maps
104. Ovid Heroides 1.33 ff., author's translation; see Heroides [and]
is open to dispute. If the word forma means "map" in
Amores, trans. Grant Showerman, Loeb Classical Library (Cambridge:
several expedition accounts, both compilation and use Harvard University Press; London: William Heinemann, 1958); cf.
are attested; but in any case there is at least one reference, Tibullus Elegies 1.10.29-32; see Elegies, trans. Guy Lee, Liverpool
and probably two, during the early empire to expedition Latin Texts 3, 2d ed. (Liverpool: Francis Cairns, 1982).
maps sent back to Rome by commanders in the field. 106 105. Propertius Elegies 4.3.33-40, author's translation; see Pro-
pertius, trans. H. E. Butler, Loeb Classical Library (Cambridge: Har-
The elder Pliny insists that the Caucasian Gates pass
vard University Press; London: William Heinemann, 1912).
should be so called, not Caspian Gates. The mistake, he 106. Robert K. Sherk, "Roman Geographical Exploration and Mil-
says, arose in expeditions to the East by Domitius Cor- itary Maps," in Aufstieg und Niedergang dey romischen Welt, ed.
bulo between A.D. 58 and 63; and "maps of the area Hildegard Temporini (Berlin: Walter de Gruyter, 1972-),2.1 (1974):
[situs] painted and sent home have this name drawn on 534-62, esp. 537-43.
107. Pliny Natural History 6.15.40 (note 41), author's translation.
them.,,107 The other reference is also in Pliny and con-
108. Pliny Natural History 6.35.181 (note 41).
cerns a party of Praetorian guards who explored the 109. Pliny Natural History 12.8.19 (note 41), author's translation.
upper Nile south of Khartoum. l08 Unfortunately the 110. Plutarch Theseus 1.1; cf. Christian Jacob, "Lectures antiques
word forma is ambiguous, since it can mean either de la carte," Etudes Fran~aises 21.2 (1985): 21-46, esp. 44-45.
suppress any hint of one against himself. Mettius Pom- drawings [diagrammata] than previous ones, but you
pusianus was put to death by Domitian because, in ad- have made it more attractive by adding iambic
dition to being born under an imperial constellation and verses.,,114 This may have been a map of Britain, since
giving Carthaginian generals' names to his slaves, he Alypius was or had been viearius of the British provinces.
carried around a parchment map of the world, together The Greek verses remind us of the Latin ones attached
with speeches of kings and generals from Livy.lll This to the Theodosian map (see below, pp. 258-59).
was a period when parchment was increasing in use as
against papyrus;112 it was certainly more transportable. RELIGIOUS AND PROPAGANDA FUNCTIONS FOR MAPS
Suspicious of spies, Domitian may have linked the map
with fears of an uprising in North Africa; in an expanded One of the first Roman maps we hear of, the 174 B.C.
empire there was plenty of scope for rebellion. map of Sardinia (p. 205), served both as a form of
thanksgiving to the gods for victory and as a useful piece
MAPS FOR TRAVELING
of propaganda showing Rome's expansion, just as Mus-
solini displayed maps of the Roman empire. Mosaics at
Ptolemy's Geography, with its tables of coordinates, was Ostia (pp. 246-47) advertised a shipping corporation or
never intended for the traveler; but there is reason to the transport guild. A somewhat different element of
think that the latitudes of Roman provinces on the republic relations is visible in the chief Christian map of
verse of portable sundials were indebted to Ptolemy, and the early Byzantine period, the Madaba map (pp. 263-
these sundials were used by travelers and surveyors. 65), where Jerusalem is given specially enlarged and de-
Itineraries for land journeys and simple periploi for tailed treatment; the map was oriented so as to be easily
sea journeys were also commonly used. A governor such seen by the congregation.
as Arrian, second century A.D. (who was also a man of
letters), chose to compile an expanded periplus for his DIDACTIC AND SCHOLARLY USES FOR MAPS
sail around the Black Sea. This may have appeared in
Latin for official purposes as well as in Greek for his It is clear from Varro that at least the more progressive
reading public. 113 Whether it was based on a map we landowners in Italy were regarded as being familiar with
cannot be sure, but the Dura Europos shield proves that the use of maps. Both he and the elder Pliny were en-
at least road maps of that area existed. It is also difficult cyclopedic writers who absorbed vast quantities of
to establish how far land itineraries were derived from Greek scholarly exposition. Varro mayor may not have
maps; since road maps are called by Vegetius itineraria cited sources (his encyclopedic writing exists only in
pieta, painted itineraries, the priority may be the other fragments). Pliny, in book 1, lists all his sources and
way around. The most famous are the Antonine land thereafter frequently refers to them, but often in such a
and sea itineraries (pp. 235-36). These are not repre- way that he does not tell us whether he is referring to
sentative of the type carried around by the traveling a text or a map. Thus for the circumference of the Black
public. They obviously detailed particular journeys made Sea, apart from Agrippa's map and commentary (which
by people like the emperor Caracalla, whereas other he calls simply Agrippa), he cites four estimates, dating
travelers might want only a section of these. Among from about 50 B.C. to about A.D. 70, all of which may
Christian itineraries, that of the journey from Bordeaux come from texts rather than maps. In the case of lengths
to Jerusalem (p. 237), by expanding on the basic list of the shores of the Red Sea, his object seems to have
form, was designed to help the pilgrim traveling to the
Holy Land. 111. Suetonius Domitian 10.3, book 8 of De vita Caesarum (The
lives of the Caesars), in Suetonius, 2 vols., trans. J. C. Rolfe, Loeb
Clearly anyone making a complicated road journey Classical Library (Cambridge: Harvard University Press; London: Wil-
would have benefited from having an itinerarium pictum liam Heinemann, 1913-14). Pascal Arnaud, "L'affaire Mettius Pom-
like the archetype of the Peutinger map. But ancient pusianus, ou Le crime de cartographie," Melanges de tEcole Fran{aise
maps may not have stood up very well to travel, since de Rome: Antiquite 95 (1983): 677-99, prefers to follow the account
they would normally be carried around loose, not in a of Dio Cassius and Zonaras. According to this, Pompusianus had a
world map painted on the walls of his bedroom. See Dio Cassius
eapsa (cylindrical box), used for storing rolls of papyrus Roman History 67.12.4 (note 7).
in the owner's house. Perhaps, however, the increasing 112. C. H. Roberts, "The Codex," Proceedings of the British Acad-
use of parchment started to promote a greater mobility emy 40 (1954): 169-204.
of maps in the late empire. Julian the Apostate, emperor 113. Arrian Periplus Ponti Euxini, in Miiller, Geographi Graeci
360-63, thanks his friend Alypius of Antioch thus: "I minores, 1:370-423 (note 20).
114. Julian Epistles 7, author's translation; see vol. 3 of The Works
happened already to have recovered from my illness of the Emperor Julian, 3 vols., trans. Wilmer Cave Wright, Loeb
when you sent the geography; all the same I was glad Classical Library (Cambridge: Harvard University Press; London: Wil-
to receive the map you sent. Not only does it have better liam Heinemann, 1913-23).
been to correct the mapping by Eratosthenes with ref- republic but in the Greek geography of Strabo. Astro-
erences to the much higher figures given by Artemidorus nomical and mathematical instruments were intro-
of Ephesus (fl. 104-101 B.C.) and by Agrippa. Varro duced into Rome as contacts with the Greek world in-
had illustrations in some of his works, but none has creased. But it often took a long time for the proper use
survived; Pliny is not known to have had any. After Pliny of an instrument to be appreciated, as in the case of the
we have no extant encyclopedias until the fifth century sundial. In 263 B.C. one such was brought from Catana
A.D. and later (Martianus Capella, Boethius, and Isi- to Rome, where it was set up near the speakers' platform
dore); of these only Isidore, bishop of Seville A.D. 602- in the forum. Nevertheless, for ninety-nine years it dis-
36, dealt specifically with geography.115 played the wrong time, since the necessary adjustment
Of purely geographical Latin writings, the lost com- for a different latitude had not been made: only in 164
mentary of Agrippa was intended (as the many frag- B.C. did Q. Marcius Philippus put up by its side a sundial
ments show) to explain his map; it would appeal mainly correct for the latitude of Rome. 119 Lucretius was ex-
to those, including Strabo and Pliny, who visited his ceptional in expounding Greek physical theory in Latin
colonnade. Pomponius Mela's De chorographia was de- verse. Cicero's contribution of this type consisted only
signed for the ordinary reading public and had no maps. of a translation of the Phaenomena of Aratus of Soli (b.
But the writer probably had one before his eyes when, ca. 315 B.C.), a meteorological work. He did promise
for example, he writes of the Baltic: "The Codanus Gulf his friend Atticus a work on geography, but though
... is dotted with large and small islands.,,116 Atticus sent him, evidently for this purpose, a work of
An interesting insight into the teaching of geography mathematical geography by Serapion, it never materi-
with the use of a terrestrial globe is found in a pupil's alized;120 Cicero, in thanking him, commented, "Be-
addendum to the Cosmographia of Julius Honorius: "So tween ourselves, I hardly understand one line in a thou-
as to avoid errors, as the teacher has said, this book of sand." 121 This is not, however, to say that Cicero never
extracts should not be separated from the globe."11? The consulted maps; he probably did in at least one case.
use of the map in schools continued after the fall of When Atticus criticizes him for having written that al-
Rome: Cassiodorus (ca. A.D. 490 to ca. 583), who after most all the city-states of the Peloponnese were by the
many years as an important administrator under kings sea,122 he replies that he got this from the tabulae of
of the Goths in Italy devoted himself to Christianity and Dicaearchus. 123 Since he goes on to refer to Dicaear-
education, recommended for teaching purposes the chus's work on the underground oracle of Trophonius,
pinax (map) accompanying the geographical poem of
Dionysius Periegetes. 118 In addition to maps in schools,
the words quoted from the rhetorician Eumenius show
115. Isidore of Seville Etymologies; see Etymologiarum sive ori-
that, in Gaul of A.D. 298 at least, an idealist wanting to ginum libri XX, 2 vols., ed. W. M. Lindsay (Oxford: Clarendon Press,
foster culture among the young after the ravages of war 1911); idem, De natura rerum, in Traite de la nature, ed. Jacques
included a large map on a balcony wall as an important Fontaine, Bibliotheque de l'Ecole des Hautes Etudes Hispaniques, fasc.
teaching aid. 28 (Bordeaux: Feret, 1960), 164-327.
116. Pomponius Mela De chorographia 3.31 (note 40), author's
Mapmaking clearly constituted an essential part of
translation.
the training of land surveyors under the late Roman 117. Julius Honorius Excerpta 50 (note 60), author's translation.
Empire. First and foremost was drawing up centuriation 118. Cassiodorus lnstitutiones divinarum et saecularium litterarum
maps, which, although to some extent diagrammatic, 1.25.2; see Institutiones, ed. R. A. B. Mynors (Oxford: Clarendon
had to be very accurate from the legal and administrative Press, 1937), or, for an English translation, An Introduction to Divine
point of view. For this purpose apprentice surveyors and Human Readings, ed. and trans. Leslie Webber Jones (New York:
Columbia University Press, 1946). For Dionysius Periegetes, see pp.
would consult the official copies; yet in the Corpus Agri- 171-73.
mensorum we find some maps that are unrealistic in 119. Pliny Natural History 7.60.214; see corresponding notes in
topographical terms. This must represent a tradition of Schilling, Histoire naturelle, 261-62, esp. n. 214.3 (note 41). There
decades of specimen teaching maps, in which topograph- was a dispute even in antiquity whether this was the earliest sundial
erected in a public place in Rome: one of 293 B.C. is also mentioned
ical accuracy and the representation of real landscapes
by Pliny (Natural History 7.60.213); again, see Schilling, 261, n.
were of secondary importance. 213.1. Cf. Censorinus De die natali 23.6-7 (cited by Schilling) in De
It must, however, be admitted that educated Romans, die natali liber, ed. F. Hultsch (Leipzig: Teubner, 1867).
although they respected Greek scientific research, did 120. Cicero Letters to Atticus 2.4.3; see Cicero's Letters to Atticus,
not always understand the mathematical principles be- 7 vols., ed. and trans. D. R. Shackleton Bailey (Cambridge: Cambridge
hind it. We do know that about 168 B.C. Crates of University Press, 1965-70).
121. Cicero Letters to Atticus 2.4.1 (note 120).
Mallos gave many lectures in Rome, illustrated by a 122. Cicero The Republic 2.4.8; see La republique, 2 vols., trans.
globe (see description above, pp. 162-64). Even so, his Esther Breguet (Paris: Belles Lettres, 1980).
globe is reported not in extant Latin writings of the 123. Cicero Letters to Atticus 6.2 (note 120).
the latest editor, Shackleton Bailey, doubts whether these or Augustus they were used by a wide range of people,
tabulae were maps124; but Cicero is probably referring had become indispensable for the surveying of land, pub-
to Dicaearchus's lost Ges periodos, a "geographical lic works, and other engineering projects, and were im-
tour" of the world. portant for legal, strategic, traveling, scholarly, and di-
In general it may be concluded that, as far as we know dactic purposes.
at present, up to about 170 B.C. maps were relatively
unfamiliar to most Romans. From that time onward
their use steadily increased. By the time of Julius Caesar 124. Shackleton Bailey, Cicero's Letters, 3:257 (note 120).
BIBLIOGRAPHY
CHAPTER 14 ITINERARIES AND GEOGRAPHICAL MAPS
IN THE EARLY AND LATE ROMAN EMPIRE Cumont, Franz. "Fragment de bouclier portant une liste
d'etapes." Syria 6 (1925): 1-15.
Avienius. Ora maritima. Edited by Andre Berthelot. Paris: H. Cuntz, Otto, ed. Itineraria Romana. Vol. 1, Itineraria Antonini
Champion, 1934. Augusti et Burdigalense. Leipzig: Teubner, 1929.
Boker, Robert. "Windrosen." In Paulys Realencyclopadie der Dilke, O. A. W. Greek and Roman Maps. London: Thames
classischen Altertumswissenschaft, ed. August Pauly, Georg and Hudson, 1985.
Wissowa, et aI., 2d ser., 8.2 (1958): cols. 2325-81. Stuttgart: Elter, A. Itinerarstudien. Bonn: C. Georgi, 1908.
J. B. Metzler, 1894-. Finkelstein, 1. "The Holy Land in the Tabula Peutingeriana."
Bosio, Luciano. La Tabula Peutingeriana: Una descrizione pit- Palestine Exploration Quarterly 111 (1979): 27-34.
torica del mondo antico. I Monumenti dell'Arte Classica, Frank, Johannes. "Beitrage zur geographischen ErkHirung der
vol. 2. Rimini: Maggioli, 1983. Ora maritima Aviens." Dissertation, Wiirzburg. Sangerhau-
Carmody, F. J. La Gaule des itineraires romains. Berkeley: sen: Schneider, 1913.
Carmody, 1977. Gichon, Mordechai. "The Plan of a Roman Camp Depicted
Castagnoli, Ferdinando. "L'orientamento nella cartografia upon a Lamp from Samaria." Palestine Exploration Quar-
greca e romana." Rendiconti della Ponti{icia Accademia Ro- terly 104 (1972): 38-58.
mana di Archeologia 48 (1975-76): 59-69. Goodburn, R., and P. Bartholomew, eds. Aspects ofthe Notitia
Clemente, G. La Notitia Dignitatum. Saggi di Storia e Letter- Dignitatum. British Archaeological Reports, Supplementary
atura 4. Cagliari: Fossataro, 1968. Series 15. Oxford: British Archaeological Reports, 1976.
Gross, H. Zur Entstehungs-geschichte der Tabula Peutinger- Polaschek, Erich. "Notitia Dignitatum." In Paulys Realency-
iana. Bonn: H. Ludwig, 1913. clopadie der classischen Altertumswissenschaft, ed. August
Itineraria et alia geographica. In Corpus Christianorum, Series Pauly, Georg Wissowa, et aI., 17.1 (1936): cols. 1077-116.
Latina, vols. 175 and 176 (1965). Stuttgart: J. B. Metzler, 1894-.
Janni, Pietro. La mappa e il periplo: Cartografia antica e spazio Reed, N. "Pattern and Purpose in the Antonine Itinerary."
odologico. Universita di Macerata, Pubblicazioni della Fa- American Journal of Philology 99 (1978): 228-54.
colta di Lettere e Filosofia 19. Rome: Bretschneider, 1984. Riese, Alexander, ed. Geographi Latini minores. Heilbronn,
Kubitschek, Wilhelm. "Itinerarien." In Paulys Realencyclo- 1878; reprinted Hildesheim: Georg alms, 1964.
padie der classischen Altertumswissenschaft, ed. August Rodwell, Warwick. "Milestones, Civic Territories and the An-
Pauly, Georg Wissowa, et aI., vol. 9 (1916): cols. 2308-63. tonine Itinerary." Britannia 6 (1975): 76-101.
Stuttgart: J. B. Metzler, 1894-. Seeck, Otto, ed. Notitia Dignitatum. Berlin: Weidmann, 1876;
- - - . "Karten." In Paulys Realencyclopadie der classischen reprinted Frankfort: Minerva, 1962.
Altertumswissenschaft, ed. August Pauly, Georg Wissowa, Stewart, Aubrey, trans. Itinerary from Bordeaux to Jerusalem:
et aI., vol. 10 (1919): cols. 2022-2149. Stuttgart: J. B. Met- HThe Bordeaux Pilgrim." Palestine Pilgrims Text Society,
zler, 1894-. vol. 1, no. 2. London: Palestine Exploration Fund, 1896.
Levi, Annalina, and Mario Levi. Itineraria picta: Contributo Temporini, Hildegard, and Wolfgang Haase, eds. Aufstieg und
aUo studio della Tabula Peutingeriana. Rome: Erma di Bret- Niedergang der romischen Welt. Berlin: Walter de Gruyter,
schneider, 1967. 1972-.
Maier, I. G. "The Giessen, Parma and Piacenza Codices of the Uhden, Richard. "Bemerkungen zu dem romischen Karten-
'Notitia Dignitatum' with Some Related Texts." Latomus fragment von Dura Europos." Hermes 67 (1932): 117-25.
27 (1968): 96-141. Van der Poel, Halsted B., et aI., eds. Corpus topographicum
- - - . "The Barberinus and Munich Codices of the Notitia pompeianum. Part 5, Cartography. Rome: Edizione
Dignitatum Omnium." Latomus 28 (1969): 960-1035. dell'Elefante, 1981.
Miller, Konrad. Itineraria Romana. Stuttgart: Strecker und Ward, J. H. "The British Sections of the Notitia Dignitatum:
Schroder, 1916. An Alternative Interpretation." Britannia 4 (1973): 253-63.
- - - . Die Peutingersche Tafel. Stuttgart: F. A. Brockhaus, Weber, Ekkehard, ed. Tabula Peutingeriana: Codex Vin-
1962. dobonensis 324. Graz: Akademische Druck- und Verlag-
Pliny. Natural History. 10 vols. Translated by H. Rackham et sanstalt, 1976.
al. Loeb Classical Library. Cambridge: Harvard University Zicari, Italo. "L'anemoscopio Boscovich del Museo Oliveriano
Press; London: William Heinemann, 1940-63. di Pesaro." Studia Oliveriana 2 (1954): 69-75.

HOC VOLUME1 Chapter14

Uploaded by

Copyright:

Available Formats

HOC VOLUME1 Chapter14

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

HOC VOLUME1 Chapter14

Uploaded by

Copyright:

Available Formats

14 · Itineraries and Geographical Maps in the Early and

Late Roman Empires

o 100 200 300 400 miles

FIG. 14.1. PRINCIPAL PLACES ASSOCIATED WITH ITIN-

cluded twice, because each has two entries in the list of

MAPS AS DECORATIVE AND SYMBOLIC

From the Roman period, as from the classical world in

o 100 200 miles

An unusual series of designs on Roman lamps, incor-

The maps associated with the artifacts described above

historical contexts. Such maps should be thought of as EUROPOS SHIELD.

sical map-the maze-that was associated with several

Much has already been either said or implied about the

You might also like