Roman Roads [Reprint 2020 ed.] 9780520348882

Citation preview

Roman Roads

Frontispiece

Aerial v i e w o f W a t l i n g S t r e e t west o f P e n n o c r u c r u m , t a k e n b y A r n o l d B a k e r

Roman Roads R A Y M O N D CHEVALLIER Director of School of Latin Studies, University of Tours Translated by N. H. Field

University of California Press Berkeley and Los Angeles 1976

First published in this edition 1976 Raymond Chevallier Translation © B . T. Batsford Ltd 1976 University of California Press Berkeley and Los Angeles, California I S B N 0-520-02834-1 Library of Congress Catalog Card Number 74-82845 Printed in Great Britain

Foreword by Professor A. L. F. Rivet

It is a great pleasure to be invited to introduce a b o o k b y m y friend R a y m o n d Chevallier to the English reader. M . Chevallier is already k n o w n to a widening circle of British archaeologists as the enterprising editor of the periodical Caesarodunum, as the indefatigable organiser of international colloquia on all aspects of the R o m a n o - C e l t i c w o r l d , and, perhaps above all, as the compiler of invaluable bibliographies of R o m a n Gaul. As the reader will quickly discover, the extensive bibliography is n o t the least useful part of the present w o r k . Julius Caesar remarked that in his time the manners and customs of the inhabitants of Britain (at least, of the south-eastern part of it) were in general similar to those of the Gauls. This could hardly be said to-day of British and French archaeologists, for the study and practice of archaeology have developed along rather different lines in the t w o countries, and especially in respect of things R o m a n . S o m e of the reasons for this m a y be f o u n d in our different cultural traditions and specifically in different degrees of continuity w i t h the R o m a n period. French is a R o m a n c e language, English is not. France has a m u c h greater wealth of upstanding R o m a n m o n u m e n t s than Britain and, unlike Hadrian's Wall, they are almost all of civil, n o t military, origin and so less likely to be regarded b y the ordinary Frenchman as the curious symbols of a r e m o t e and alien domination. A h u n d r e d French cities preserve in their very names the m e m o r y of the tribal states w h i c h m a d e up R o m a n Gaul. Besides this, there is a m u c h greater b o d y of c o n t e m p o r a r y written material, b o t h literary and epigraphic, relating to Gaul than there is relating to Britain. As a result, it can almost be said that while the British archaeologist digs to create his R o m a n history, the French archaeologist digs to illustrate it. A n d herein lies the particular value of this w o r k to the British reader. It is essentially a French b o o k , written f r o m the French point of view, and the publishers have wisely refrained f r o m trying to anglicise it in any significant w a y except in the language. It does n o t seek to instruct the British archaeologist in the identification of R o m a n roads in Britain - t h o u g h even here he m a y find f o o d for t h o u g h t in some of the methods w h i c h M . Chevallier advocates for France. Rather it is concerned to discuss R o m a n roads in their wider imperial setting and in all their aspects - military,

Foreword by A. L. F.

VI

Rivet

commercial and administrative - and in so doing brings to notice a vast wealth of comparative material not only from Gaul but f r o m the whole of the Roman Empire. Accordingly, while there is much here that is familiar to the British archaeologist the use of maps and place-names, for example, and parish or commune boundaries, of air photographs and of the spade - the product of these techniques is all related to its proper historical background and a new picture emerges not only of the roads of Gaul but of those of Britain too. In both cases they are presented not as the subject of a parochial exercise in archaeological expertise but as parts of the vital framework of the great Empire of which both countries once formed part and the book is thus one to be read not merely by archaeologists but by all w h o take an interest in Roman imperial history. University

of Keele

A . L. F. R I V E T

Contents

Foreword by Professor A. L. F. Rivet

v viii

The Illustrations Preface to the English Edition

ix

Introduction

xi

I

Evidence from literature and inscriptions

16

2

The archaeology of Roman roads

65

3

Roads in the Roman Empire

131

4

Life on the road

178

5

The function of Roman roads

202

References

210

Bibliography

227

Index

258

vii

The Illustrations

Aerial view of Watling Street west of Pennocrucrum frontispiece 1 The Peutinger Table 2 Illustrations from the Peutinger Table 3 Lists of stations from the Antonine Itinerary 4 Main roads in Gaul 5 A milestone from Northern Italy 6 Inscription from a Vicarello goblet 7 Inscription and motif from the Amiens patera 8 Extract from La Guide des Chemins de France by Charles Estienne 9 Plan of R o m e 10 Paved street in Pompeii 1 1 Map of Autun, ancient and modern 1 2 Part of the cadastral inscription from Orange 13 Roman road from Rheims to Trier 14 Section of the via Mansuerisca 15 A mosaic from Ostia: the Aries bridge 16 The Fabricius bridge in Rome 17 The Trier bridge 18 The 'route des Gaules' in the Aosta valley 19 Geophysical traverses across a Roman road at Izernore 20 Use of a large-scale map: St Just-en-Chausee 21 Vertical air photograph of a Roman road 22 Oblique air photograph of a Roman road 23 Use of the 1/20,000 large-scale map: abandonment of a Roman road near a settlement 24 Relationship between Roman road, settlements and land-holdings at Voves 25 Plan of the road-system at Bologna 26 Field-system and the Roman road from Saintes to Poitiers 27 The main Roman roads in Italy 28 The Roman roads of northern Italy

viii

The Illustrations 29 30 31 32 33 34 35 36 37 38 39 40

IX

The main Roman roads in the Middle East The Roman road system in northern Syria Roman roads in the province of Africa Roman roads in Spain Roman roads in southern Britain The main Roman roads in Gaul Early road system around Chartres Roman roads in Belgium Roman roads between the Rhine and Danube Trade routes into Free Germany Finds of Roman coinage in Free Germany A horse-drawn vehicle

Preface to the English edition

A s I finish reading the English version o f m y b o o k on R o m a n roads I am most anxious to express m y gratitude to the author o f the f o r e w o r d , the translator and the publishers. I am happy to say that the w o r k has had a v e r y favourable reception 1 , apart f r o m a couple of critics. O n e is y o u n g , as yet without an international reputation, and appears to be looking for an argument with me : as far as I am concerned, a manual (Handbuch) is not a Forschungsbericht, a f o r m o f publication a n y w a y w h i c h I have contributed to the Mélanges V o g t . In any case it is nonsensical to suggest that the present manual (which has never claimed, be it noted, to be competing with the irreplaceable - and unobtainable - v o l u m e b y Grenier) is a w o r k with a message. W e do not need to think deeply about history to realise that all roads, R o m a n or not, have fulfilled, either successively or simultaneously, strategic or economic roles: while the via Aemilia probably answered 'socio-economic' problems, it also served as a strategic road against the Ligures o f the Apennines. Furthermore, in m y v i e w it is valid to compare streets in towns with roads in the countryside, if only because neglecting the continuity o f usage linking both groups is the best w a y o f not understanding the influence o f topography on settlement 1

The book has been awarded the Prix Broquette-Gonin by the French Academy.

The Illustrations 29 30 31 32 33 34 35 36 37 38 39 40

IX

The main Roman roads in the Middle East The Roman road system in northern Syria Roman roads in the province of Africa Roman roads in Spain Roman roads in southern Britain The main Roman roads in Gaul Early road system around Chartres Roman roads in Belgium Roman roads between the Rhine and Danube Trade routes into Free Germany Finds of Roman coinage in Free Germany A horse-drawn vehicle

Preface to the English edition

A s I finish reading the English version o f m y b o o k on R o m a n roads I am most anxious to express m y gratitude to the author o f the f o r e w o r d , the translator and the publishers. I am happy to say that the w o r k has had a v e r y favourable reception 1 , apart f r o m a couple of critics. O n e is y o u n g , as yet without an international reputation, and appears to be looking for an argument with me : as far as I am concerned, a manual (Handbuch) is not a Forschungsbericht, a f o r m o f publication a n y w a y w h i c h I have contributed to the Mélanges V o g t . In any case it is nonsensical to suggest that the present manual (which has never claimed, be it noted, to be competing with the irreplaceable - and unobtainable - v o l u m e b y Grenier) is a w o r k with a message. W e do not need to think deeply about history to realise that all roads, R o m a n or not, have fulfilled, either successively or simultaneously, strategic or economic roles: while the via Aemilia probably answered 'socio-economic' problems, it also served as a strategic road against the Ligures o f the Apennines. Furthermore, in m y v i e w it is valid to compare streets in towns with roads in the countryside, if only because neglecting the continuity o f usage linking both groups is the best w a y o f not understanding the influence o f topography on settlement 1

The book has been awarded the Prix Broquette-Gonin by the French Academy.

X

and the salient features o f the landscape, too often unrecognised until the very recent problems spotlighted by the Congress on Greater Greece. In general, m y critics have appreciated the range o f information; the use o f illustrative material that has been taken not just from archaeological works, but also from literary sources, including some technical details (though the latter admittedly are not numerous enough, even allowing for the need to select); the importance o f the bibliography (although one critic considers it too long) and its classification, always allowing that some titles might figure under different headings. But it is no good blaming me for not listing all the entries in the Real-Encyclopadie that have any connection with the subject. Is this the place to justify shortcomings? Given the fact that in describing the road network o f the Empire the accent w o u l d be on Gaul and Italy, I was well aware that I w o u l d displease the specialist interested in Andorra or Lichtenstein. B u t was the purpose o f filling pages and pages with lists o f stations to be found in the various editions o f the itineraries? In all event I thank m y teachers, friends and colleagues, particularly Monsieur Hano, w h o have been kind enough to point out a number o f errors. These have been corrected in the present edition, which also includes additions to the index. R A Y M O N D CHEVALLIER

Introduction

H O W ROADS BEGAN The earliest roads were the work of the elements and wild beasts, mere huntsmen's trails, as Fustier has recalled in a recent book. The hunter is well aware o f the fairly regular movements o f game animals, instinctively drawn towards country where they may find their prey or where there is pasture or water. In uplands such tracks will fit in as a normal course with the rise and fall o f the ground. It is an interesting experiment to study how footpaths, with their smooth curves, always aim at avoiding natural obstacles. W e should take a closer look at ancient ways along which transhumance occurred. It is in this light that Dion has given a masterly introduction to the study o f the geography o f traffic in France 1 . If the use o f the horse and the wheeled vehicle seems right enough on the steppes, such was not the case in Gaul until the forest cover had been penetrated. In the forest one has to walk in single file. In this way arose the ridge-way, little affected by erosion, and the 'drailles' o f the South o f France may have been brought about by animal migration. Changes in climatic conditions play an important part here: we know that the horse was first driven towards the grasslands o f Central Europe by the extension o f forests at the end o f the Palaeolithic. The horse was seen again in the West as a domesticated animal after the arrival o f the ox at a time when forest clearance was taking place at the hands o f the earliest farmers. 'The history o f human movement is closely connected with the development o f agriculture, and we may usefully note that there has been a contrast in France since the Bronze Age between those regions already converted into open country and provided with a stock o f sheep and horses and with an array o f wheeled vehicles and other regions which by contrast were strongholds o f the forest belt.' In the first group Dion gives as an example Champagne, where La Tene burials show that swift, light war chariots with two wheels were in use. The broad plains o f North-Western Europe, together with gaps and valleys, have usually encouraged human movement. On the other hand, claylands have been a handicap. What archaeological evidence is at our disposal? The wheel seems to have been XI

12

Introduction

known in Mesopotamia as early as 3000 BC. There are clay models of four-wheeled waggons in the Danube valley about 2000 BC and wheels of solid wood, belonging to the same period, have been found in the L o w Countries. This progress westwards of the wheel is certainly connected with the movement of prehistoric peoples. In this way the two-wheeled war chariot was introduced into Western Europe (it is mentioned in Caesar's account of the Gallic W a r s 2 and was still in service in Britain during the first century AD, as Tacitus makes clear in his Agricole). From the end of the Bronze Age, there were wheels with felloe and spokes made of metal. The technique was still being perfected in the Early Iron Age. It has been possible for J o f f r o y to put together the parade chariot of the princess of V i x thanks to the careful recording of the position of all the metallic parts, followed by comparison with similar finds elsewhere (the chariots from Ohnenheim, Dejbjerg and Bell im Hunsrlick), and the support of the Nancy laboratory (A. France-Lanord). W e may refer to the long description given by the author of The Treasure of Vix4: wheels with iron tyres and spokes jointed into the felloe and bosses covered with overlapping bronze plates, together with a steerable limber and a chariot frame that could be taken to pieces. Other evidence could also be quoted: the early chariot from La Côte-Saint-André 5 , the Gallo-Roman chariot (second or third century AD) from Fa (Aude), the bronze-cast wheels of which are on show in the Musée Saint-Raymond at Toulouse, as well as a piece of decorative gear from the same source. What kinds of road may be associated with these vehicles? It is difficult to say, but there were expedients certain to have been used: on swampy ground, corduroy roads and, in mountainous country, rutted tracks. R O A D S IN E A R L Y H I S T O R Y A N D

LEGEND

Happenings in myth and legend may well recall actual events of long ago. Just as the Odyssey has given epic form to early Mycenean voyages, so the travels of the Argonauts and the myth of the Hyperboreans merely seem to mask the sea-routes along which Mediterranean bronze was exchanged for Baltic amber. In the same way, the myths that have as their theme the journeys made by Dionysos allow us to follow the movement of Greek wine on its w a y to the Barbarians, for one of the aims of colonisation was always to produce wine in the immediate neighbourhood of those peoples who had been buying it at a high price. The Myceneans had already fallen under the spell of the Mediterranean Far West, as represented by Tartessos, if w e are to believe Ulysses' visit to Calypso. These contacts between both ends of the Mediterranean were even more ancient than this, judging by the legend of Heracles going to seize Geryon's cattle6. This story helped Greek merchants to stand up to Phoenician claims of complete control over the route to the Spanish orefields and the sea-ways to amber and gold. Pytheas was to undertake his exploration in the wake of the Hero. In his Promenades du pays de la Dame de Vix, Carcopino has suggested how the spread of the same folk tale brought the Hero as far as Alesia.

Introduction

13

Likewise, the return of the Argonauts, best known through the Argonautica of Apollonius Rhodius 7 , as they passed from the Danube across to the Adriatic, then up the Po and the Ticino and over the Alps, to come down the Reuss Valley, across Lake Lucerne, then along the Aar and the Rhine, from which point the goddess Hera forced them to return towards the Aar, Lake Neuchatel, Lake Geneva and the Rhone, provides an example in legend of the use of the trunk route along the Danube and the Sava, linking the Black Sea and the Adriatic by easy overland portage. Dion has emphasised a statement by Strabo giving credence to this journey: 'Everyone agrees that the expedition of the Argonauts and their halts on particular islands are facts of which the authenticity cannot be denied. Until the Roman period, the river Ister or Danube was very little known and there were no clear ideas about its upper course. The river and its tributaries were confused with the trade-routes opened up along their valleys.' This explains w h y w e find in Herodotus the notion of a Danube splitting Europe in two, along a line extended by the Po, the Durance and the route skirting the foothills of the Pyrenees. Dion has shown 8 that the story of Heracles illustrated this second part of the great highway across Europe. Jason, w e may remember, did not manage to go beyond the Swiss plateau, whereas he had been thinking of sailing round Western Europe by emerging from the mouth of the Rhine and then passing through the Strait of Cadiz. T o explain this anomaly, Jullian, in an inspired moment, has suggested: 'Let us imagine that La Tene was a great Celtic market-centre and further that the Greeks had either the right or the opportunity of going through it, but were forbidden to go beyond, then the fact that the Argonauts had to halt finds a satisfactory explanation and all that Juno did was to remind them of the regulations imposed by the local inhabitants.' Thus it was that along these routes, and especially where fairs were held 9 , the crafty Mediterranean traders made sure they were preceded by Greek Heroes, w h o were shown as being, on the one hand, tough towards the wicked - whether these were brigands or simply hostile to buying and selling 1 0 - and, on the other hand, well disposed towards those who welcomed them as guests. These Heroes, whose exploits illustrated the great trade routes, were to some extent guarantors under whose authority future dealings were to take place. It was thanks to their support that it was possible to establish openly vital claims to spheres of influence in the face of later competitors. A number of other legends appear to deserve an explanation of this kind, for example, the story of Diomedes in Italy that throws light on the roads crossing the Apennines. Some texts of classical date, which belong no longer to myth, but to history, and which hint at an ancient system of genuine trade routes, together with archaeological evidence, allow us to trace some of these highways, which may have specialised in the movement of one particular product, tin or amber. Thus Herodotus (v, 9 - 1 0 ) 1 1 draws attention to a strange folk, 'nomadic traders', w h o used to hawk their wares from waggons, along the Danube as far as the borders of the Veneti, and were known to the Ligures in the hinterland of Marseilles. This very route is mentioned in a text of Pseudo-Aristotle (De mirabilibus

H

Introduction

auscultis, 85): ' T h e y say that f r o m Italy as far as Céltica, in the territory o f the Celtoligures and the Iberians, there is a road bearing the name o f the Heraclean W a y . Should a Greek or a local man venture along it, the inhabitants make sure that no harm befalls him, as they w o u l d bear the penalty.' T h e profits drawn f r o m tolls and portage do explain w h y those w h o dwelt b y roads should have become expert, during the later prehistoric period, in maintaining and safeguarding some routes, particularly where these passed through ports, along rivers and over mountains, as, f o r example, the Celts in the case of the Alpine passes 1 2 . This role as guide, well attested in historical times 1 3 , has long been considered by mountain folk to be their prerogative 1 4 . This was pointed out in the sixteenth to eighteenth centuries b y travellers crossing the Alps or the Apennines. It is worth quoting the famous passage f r o m Diodorus, referring to the tin routes (v, 22, written between 60 and 30 BC, but using much earlier sources, such as Timeas, third century BC, and Posidonius, 100 BC): ' T h e traders w h o buy tin [in Britain] transport it over to Gaul, which they then cross on f o o t in 30 days, with their load on pack horses, to reach the mouth o f the Rhône.' Another passage confirms this statement : ' A large amount o f tin coming f r o m the British island finds its w a y over to the shore of Gaul opposite. T h r o u g h Céltica, the traders take it b y pack-horse to the people o f Marseilles and the t o w n called Narbonne.' D i o n 1 5 has shown that this route goes along the estuary o f the Gironde and the valley o f the Garonne b y w a y o f Bordeaux, Toulouse and Narbonne. B u t several highways were being used at the same time or one after another. As the numerous middlemen w e r e unaware o f the extremities o f the chain, discussion is by no means finished on the possible courses, particularly since the journeys w e r e often a combination of transport by land and b y water, going up the valleys o f the navigable rivers as far as the gaps, where portage became necessary : the Seine valley - V i x - the B u r g u n d y gap, the S a ô n e - R h ô n e or the Swiss plateau and the Alpine passes ; the Loire valley - portage possible f r o m several places (Roanne) - the R h ô n e valley ; the Garonne valley (Carcassonne or Naurouze gap). W h a t is certain is that the Mediterranean penetrated a long w a y , right into G e r m a n y w h e r e Tacitus 1 6 came upon the tracks o f Ulysses. So, f r o m that time on, there had been a spread o f cultural influence : the discovery o f the Hirschlanden warrior, the most northerly kuroi to be sculptured b y a native, and which has its counterpart in the Entremont statues, throws light, according to B e n o î t 1 7 , on the legend mentioned b y Tacitus that Northern Céltica had been hellenised. Ulysses had been attributed w i t h the foundation o f the naval station at Asciburgium, identified as Eschenburg, near Duisburg, on the Rhine, a depot on the amber and tin trade routes. A t the time o f the historian, there w o u l d still have been in existence monuments bearing Greek inscriptions at the cross-roads between G e r m a n y and Rhaetia. Since the d a w n o f history, the discovery o f relief sculpture has provided vital clues in tracing the great highways o f trade. In the same w a y , the find spots o f products f r o m Greece

Introduction

15

and Etruria and of coins and pottery from Marseilles pinpoint the roads along the valley of the Rhone and along the Pre-Alps (the upland route parallel with, and east of, the Rhone). The situation was the same throughout the Roman West on the eve of the Roman conquest. Everywhere, there were tracks in use by traders, who almost always preceded the soldiery. On the coinage of the Arverni, the amphora symbolises the economic stranglehold of Italic merchants on a Celtica that was still free; there are amphorae all along the road bringing wine from Marseilles. That such land tracks existed and were used by Hannibal 1 8 is supported by the migration made by the Helvetii in waggons and by the swiftness of Caesar's movements in Gaul. The Roman general also mentions engineering works, such as the bridges over the Loire 1 9 . It is clearly difficult to date these roads. Distribution maps can help in this direction by showing the archaeological material found in association with these routes, particularly at focal points such as fords and passes. It is helpful, too, to note links with the oppida 20, as for example in the case of the oppidum of the Caisses de Mouries (fourth to third centuries BC) , lying at the entrance of the valley of the Vallongue. This valley connected the Crau lowland with the fords over the Durance and was a prehistoric track used in the seasonal movement of herds and improved in the Roman period. Cassini's map, on which the extent of forest cover is clearly displayed, brings out a chain of clearings linked by age-old paths. If w e turn to the case of Vermand, the oppidum that was replaced by Saint-Quentin, w e find that each was the centre of a different road network, suggestive of development at successive times, probably within the Roman period. Agrippa may have been responsible during his residence in Gaul, first 39-37 then 19 BC. The time had come when a Roman town managed to attract to itself a pre-existing network of native roads, some of which were already in service close to the new centre.

I

Evidence from literature and inscriptions

i L A T I N W O R D S F O R R O A D S A N D STREETS A look at the ancient language at this point is extremely useful, since it throws light on the economy and in a more general way, on the civilisation of R o m e . Besides, this will enable us to understand some of the very revealing place names, whose importance will be emphasised later. André 1 has shown h o w the Indo-European name for road or track (its derivative in Latin, pons, developed a special sense) was replaced by groups of words closely connected in meaning with the ground surface and with everyday living, words to be expected f r o m a nation of peasant farmers. Thus w e find: via, a carriage road, allowing two vehicles to pass (from vehere) ; semita, a small path, one foot wide, dedicated to the goddesses, the Semitae, a way between t w o garden-beds; vicus, first a country township, then a city street; callis, a track used for seasonal transhumance (whence the Spanish 'calle') ; trames, a mountain path at right angles to the line of the crest. Some words have evolved f r o m the concrete to the abstract or vice versa : iter, a route, a connecting road, a right of way, a footpath (two feet wide, for travellers on foot, horsemen or litters); actus (from agere, to drive cattle) a right to drive animals, then a track for animals, later for vehicles, a one-lane road of four feet; limes, a road of bare soil (cf. limus, m u d ?), often acting as a boundary, a byroad, a subdivision for centuriation, then a road outside the ramparts, finally in the special military sense of a fortified frontier-work. T h e following words m a y also be mentioned: agger, a causeway f o r m i n g a road 2 ; compendiaria, a short cut; Jìexus, a fork or a turning. W i t h the development of t o w n life there appear words that are strictly urban : vicus takes the meaning of district, then street, while semita, originally footpath, in towns becomes pavement or sidewalk (margo and crepido are also found). In addition to via, for the street is simply the prolongation of the highway (for example, the city section of the Flaminia is called via Lata), w e find the following : ambulatio, a public walk open to the sky (whereas the porticus had a roof)

16

Evidence from literature and inscriptions

17

angiportus, a lane or alley between houses (according to ancient Roman law the owner had to be able to go right round his dwelling) area, a threshing-floor, then an open space clivus, an ascending or descending street forum, a market place, originally the enclosure within the Indo-European house fundula, a cul-de-sac pervium, a town thoroughfare platea, an approach road, then a square scalae, steps or stairs. Except for angiportus and fundula, these are all former rural words with changed meanings. The evolution of the words actus (cf. above) and limes (which has a variety of adjectives) reflect the way in which local services and dues were given legal form. Via, having entered general use, finds competition in the third century from strata (via is meant), indicating first of all an embanked road, so giving us 'strada', 'strasse', 'street'. In some districts, via is replaced by rupta (via), the beaten track, the root of the French 'route'. In Vulgar Latin, furca and quadrifurcus are found instead oibivium and quadrivium. Ruga ('rue') is used figuratively in early Romance in place ofvicus, whilst a Celtic word forms the root of the French 'chemin', the Italian 'cammino', and the Spanish 'camino'. It would be interesting to discover from official documents drawn up in Greek exactly how these words were translated from Latin. These equivalents would throw light on the way in which semantic changes occurred by means of analogy.

2 THE R O A D IN L A T I N L I T E R A T U R E One would expect to find, especially in the Histories, detailed information on the construction and use of roads, which were the basis of Roman power. But this is far from the case, since history was for the Romans above all the study of politics and of human behaviour. The following is the result of a close examination of what is left of Livy's works 3 . Only seven times is there mention of the building of roads by civil magistrates: ix, 29: the censor Appius Claudius, responsible for the Appian Way. ix, 43: the censors Bubulcus and M . Valerius Maximus, builders of local roads (viae per agros) at State expense. x , 23 : the aediles curules, with the money obtained from confiscating the wealth of usurers, had the road from the Capena Gate to the Temple of Mars resurfaced with paving-stones. x , 47: use by the aediles of fines levied on farmers grazing animals on public

i8

Evidence from literature and inscriptions pasture in order to pave the via Lavicana from the Temple of Mars as far as Bovillae. XXXVIII, 28: the repaving of a Roman street leading from the Capena Gate to the Gate of Mars undertaken by private enterprise. x x x i x , 44: the censor Flaccus had a highway built, in the interests of the people, leading to the waters of Neptune and had a road tunnelled through the mountain of Formies. XLI, 32: the censors were the first to put out for tender paving the city streets, gravelling and embanking the highways, and building bridges at a number of points. . .; they had the way up to the Capitol paved.

One passage only mentions the building of roads by the army: x x x i x , 2: After having pacified Liguria, Aemilius had his army build a road from Piacenza to Rimini to join the via Flaminia . . . After restoring peace to his province, the consul Flaminius would not allow his soldiers to remain inactive and had them build a road from Bologna to Arretium. Some references are incidental: 2nd, 3rd and 8th milestones of the via Gabina (11, 1 1 ; in, 6; v, 49). 3rd mile of the via Salaria close to the Arno bridge (vn, 9), 8 th milestone of the via Appia (VII, 39), statue of Mars on the same road (XXII, 1 , in connection with a portent), the via Nomentana formerly named Ficulensis (111, 52, at the time of the mutiny on the Sacred Mountain). What is shown in particular is the manner in which the army made use of the roads: IV, 4 1 : a consul returning from campaign along the via Lavicana, close to Fanum Quietis. Horses and waggons are sent there from the city to pick up the troops exhausted by battle and a night's marching. XXII, 1 1 : the dictator going off along the via Flaminia to meet the consul and the army. XXII, 55 : after Cannae, Q. Fabius Maximus has light cavalry dispatched along the via Appia and the via Latina to question the fugitives. x x v i , 8: a more enlightening text: the proconsul Q. Fulvius proceeded along the via Appia, sending messengers ahead to all the towns on that road, such as Setia, Cora and Lanuvium, so that there should be food ready in those towns, brought in from the surrounding countryside. x x x i x , 2: the pursuit of the Ligures in the Apennines over difficult tracks. Roman roads were sometimes used by the enemy: 11, 39: C . Marcius on the via Latina. ix, 43 : the Samnites cut the roads and occupy the passes. x , 36: the Samnites try to seize Interamna on the same road, x x v i , 8: Hannibal goes along it.

Evidence from literature and inscriptions

19

Except for the previous references, the streets o f R o m e are seldom mentioned: a statue of Clelia on horseback at the top of the Sacred W a y (11, 1 3 ) ; people killed by lightning in the via Fornicata ( x x n , 36); demolition b y order of the censors o f houses projecting onto the public h i g h w a y ( x x x i x , 44). T h e geographers (Strabo, Pliny the Elder, Pausanias) provide more information but a great deal again can be found in the vast number o f written sources. A s regards Syria, f o r example, the authors used b y Father Poidebard are Josephus, A m m i a n u s Marcellinus 4 , Plutarch, D i o Cassius and Procopius, without counting the R o a d Itineraries. " TRAVELS OF A RICH CITIZEN - PLINY THE YOUNGER

A s an example o f the movements of a wealthy R o m a n , w e m a y consider those o f Pliny the Y o u n g e r , whose Letters provide us with some details. His main reason f o r travelling was to visit his estates, f o r 'there is enchantment in a change o f air and landscape and even in merely j o u r n e y i n g around one's properties' (ill, 19, 4, cf. 1, 3, 2 ; i x , 1 5 , 3 : 'I mount m y horse and I play m y part as landlord b y riding around'; iv, 14, 8: 'I travel about m y holdings and listen to m a n y a countryman's tale o f woe'). M o r e often all he is seeking is the physical exercise and the uplift f r o m seeing n e w places (iv, 1 4 ; VII, 4 - 8 ; i x , 36: 'I get into m y carriage.' H e has the same interest w h e n passing along the garden walks in his villas as w h e n out riding, cf. v , 6, 1 7 . ' M y mind maintains its alertness, refreshed b y the v e r y change.'). In this respect he f o l l o w e d the example o f his uncle, Pliny the Elder, w h o used to w o r k with his secretary on his travels, and even in R o m e , borne along in his litter (111, 5). T h e reasons for m o v i n g around might be trivial - sight-seeing, a delight in the oddities o f nature, 'which set us going on our travels, d o w n the road and across the sea, yet leave us u n m o v e d if they lie within sight.' (vm, 8, 20). O r the reasons m a y be serious: 'I had been a w a y in Tuscany to preside over the first stages in the construction o f a public monument being built at m y expense, having obtained leave o f absence as Prefect at the Treasury', while it is his important mission to Bithynia, on which B o o k x o f the Letters gives us such useful details (journeys into the Province, official reception o f magistrates, embassies, travel permits, messengers). In a letter about the illness of the freedman Zosmius (v, 19), w h o m Pliny sent first to E g y p t , then to Frejus, because he was spitting blood, w e learn also that people used to travel for their health. It has been possible to find only three other quotations about travelling conditions : the employment o f a closed carriage one day w h e n Pliny was suffering f r o m eye-strain (vn, 2 1 ) the sandy build-up o f the track which, forking o f f f r o m the via Laurentina or the via Ostiensis, led to his villa of the Laurentes ( 1 1 , 1 7 : 'Horse-carriages m o v e along it slowly and with some difficulty, but it is a short road and a g o o d one f o r a horseman'). Finally a w o r r y i n g item o f news, which shows that brigandry had

20

Evidence from literature and inscriptions

remained endemic in Italy, n a m e l y the disappearance o f a centurion in the course o f a j o u r n e y (vi, 25).

R O M A N ROADS A N D THE EMPEROR, A C C O R D I N G T O SUETONIUS

T h e Lives of the Twelve Caesars b y Suetonius acquaints us w i t h the manner in w h i c h the Emperors travelled and their reasons for so d o i n g . W e pick o u t at r a n d o m : Caesar, x x x i v : f r o m C o r f i n i u m to Brindisi, passing d o w n the Adriatic coast, then o f f to Spain, LVIII: the crossing f r o m Brindisi to D y r r a c h i u m in winter. Augustus, w h o preferred to travel b y w a t e r , did n o t hesitate to face storms (LXXXII), but he travelled to E g y p t f o l l o w i n g the coast o f Syria and Asia (xvn). In his later years, he used a litter f o r travelling, almost always at night, at a s l o w speed and in short stages, taking t w o days to j o u r n e y to Praeneste or T i b u r . Tiberius, w h o , as a y o u n g man, travelled extensively (cf. x i : s e a - v o y a g e as far as the R h o n e , x i v : crossing o f M a c e d o n i a for an expedition into Syria), did n o t m o v e far in his old age (XXXVIII, XL, LXXII, returning f r o m C a p r i to the outskirts o f R o m e in a trireme and on another occasion approaching the city along the via Appia). H e w a s not, h o w e v e r , free f r o m sudden impulses, and in case he wished to m a k e j o u r n e y s into the provinces, m a d e sure that necessary supplies w e r e available in the municipia and the colonia. ( x x x v i i i ) 5 . Caligula, w h e n leaving on an expedition, hesitated b e t w e e n t w o extremes, either so swift a travel rate that the praetorian cohorts w e r e obliged, in order to keep up w i t h h i m , to put their standards on pack animals, or so slow a speed that he w a s able to be borne along in a litter b y eight bearers, w h i l e ensuring that ahead o f h i m the local plebs swept and w a t e r e d the road to keep d o w n the dust (XLIII) . H e t o o w e n t in f o r sight-seeing and visited the river C l i t u m n u s and its sacred g r o v e at M e v a n i a (ibid.). Claudius, on his w a y f r o m Ostia to Britain, w a s t w i c e almost s h i p w r e c k e d because o f storms. So he travelled o v e r l a n d f r o m Marseilles to B o u l o g n e (xvn). A keen gambler, he kept up his dicing in a carriage specially arranged f o r the purpose (XXXIII). Vitellius w o u l d display his g l u t t o n y in w a y s i d e taverns 'seizing almost f r o m the v e r y fire dishes that w e r e still steaming o r leftovers f r o m the day b e f o r e and halfeaten f o o d ' . T h e same Life (xiv) i n f o r m s us that the E m p e r o r w a s n o t e x e m p t f r o m road tolls. Vespasian (xxi) w e n t about in a litter. Titus (v) m a d e his w a y to Paphos to consult the oracle o f Venus. D o m i t i a n ( x i x ) during campaigns or w h e n travelling seldom rode a horse but n o r m a l l y used a litter. T h e s e snippets o f i n f o r m a t i o n (there are m a n y others in the Scriptores Historiae Augustae) lack some o f the details w e should like today. H o w e v e r , Suetonius gives us s o m e interesting information about imperial p o l i c y f o r roads, w h i c h w e r e m a i n l y

Evidence from literature and

inscriptions

21

constructed for the purpose of prestige. Such a policy had already been thought up by one C. Gracchus, who, according to Appian, 'had great roads built the length and breadth of Italy and set to work on his behalf a horde of workmen of every kind, ready to do all he wanted' 6 . W e may quote from Suetonius: Caesar's plans (XLIV) : the construction of a road f r o m the Adriatic Sea to the Tiber over the ridge of Apennines, the digging of a canal across the Isthmus of Corinth (an idea taken up by Nero, xix), the enthusiastic erection of a bridge over the Rhine (xxv). the more realistic and successful efforts of Augustus: x x x : repairs at his o w n expense to the via Flaminia as far as Arminum, the other roads being 'shared amongst the generals honoured by a triumph, w h o had to find the money for this out of their battle-spoils'. XXXII : the suppression of brigandry by setting up military posts along the roads, for 'in the country districts travellers were being kidnapped and were being held within the ergastula of the villa-owners themselves'. XXXVII : a department was created to look after public works, highways, watersupplies and the river Tiber. These measures were repeated by Tiberius: VIII :

Suetonius comes back to the subject of lawlessness on the roads: 'Workshop owners were suspected of holding on to travellers as slaves after having received them as guests'; x x x i : the Emperor made an unsuccessful request to the citizens of Treba that they should authorise for the upkeep of a road a sum of money bequeathed to them for the construction of a theatre; x x x v n : he increased the number of military posts throughout Italy, a move justified when the inhabitants of Pollentia halted a supply column in the charge of a senior centurion. Caligula (xix) 'caused to be built between Baiae and the Pozzuoii jetty a bridge that was made up of barges covered with soil so that the whole thing looked like the via Appia. According to most reports, Caligula's idea was that such a bridge would rank with the one built by Xerxes, but others held that he was trying to terrify the Germans and Britons, then making warlike threats, with the news of a vast engineering feat.' Claudius (xxv) issued a decree forbidding travellers to go through Italian towns other than on foot, in a sedan chair or in a litter. The adventus evolved considerably in the Early Empire. W h e n Caligula came back from Germany, the praetorian cohorts marched ahead of him and the people of R o m e ventured out as far as the twentieth milestone (cf. the triumphal entries made by emperors on horseback or in a chariot: Otho, x, and Vitellius, x). Likewise the Historia Augusta and inscriptions yield information on Hadrian's journeys.

22

Evidence from literature and inscriptions THE SUBJECT OF TRAVEL IN THE 'METAMORPHOSES' BY APULEIUS

T h i s w o r k suggests a v e r y m o b i l e society. W h o w a s t r a v e l l i n g and w h y ? sightseeing: 'I w a s still in the c h a r g e o f m y guardian, w h e n , h a v i n g c o m e f r o m M i l e t u s to attend the O l y m p i c G a m e s , I m a d e u p m y m i n d to l o o k r o u n d that part o f the illustrious p r o v i n c e in w h i c h w e l i v e d ' (n, 2 1 ) ; religious p u r p o s e s : L u c i u s w e n t t o R o m e f o r initiation ( x i ) ; the priests o f the S y r i a n g o d d e s s w e r e b e g g i n g their w a y f r o m t o w n to t o w n (VIII, 24); business reasons: 'I w a s p r o c e e d i n g t o Thessalia o n business' (1, 2); 'I trade in h o n e y , cheese and o t h e r f o o d s t u f f s o f that k i n d used b y i n n k e e p e r s and I g o all o v e r Thessalia' (1, 5). B e f o r e l e a v i n g o n a j o u r n e y , as p e o p l e w e r e superstitious - 'I h a d started o f f left f o o t first and so I w a s f o i l e d o f m y e x p e c t e d p r o f i t ' (1, 5) - the a d v i c e o f the g o d s was sought: ' A C h a l d e a n states w h i c h d a y is best c h o s e n f o r t r a v e l l i n g b y r o a d o r p r o p i t i o u s f o r p u t t i n g to sea' (11, 1 2 ) 1 ; ' A business m a n consults a f o r t u n e - t e l l e r to k n o w the best d a y f o r a j o u r n e y ' (n, 13); ' q u a c k priests m a d e u p a single a u g u r y w h i c h c o u l d be applied in a n u m b e r o f cases: " I f o x e n are y o k e d and s t r a i g h t a w a y p l o u g h their f u r r o w , that is a sign that their efforts w i l l y i e l d a rich h a r v e s t " . W a s there s o m e o n e w i t h a j o u r n e y in m i n d w h o w i s h e d to receive the o m e n s v o u c h s a f e d b y the g o d d e s s ? T h e n the m o s t docile animals in the w o r l d , already y o k e d , w e r e r e a d y to h a n d f o r h i m and w h a t the soil w o u l d y i e l d g u a r a n t e e d a successful o u t c o m e ' (ix, 8). H o w did p e o p l e travel? M o v i n g a r o u n d w a s e x p e n s i v e : w h e n L u c i u s w e n t to R o m e , his o u t l a y ' m e l t e d a w a y his h u m b l e inheritance' (xi, 28). P e o p l e travelled o n f o o t , o n h o r s e b a c k or o n a m u l e , in t w o - o r f o u r - w h e e l e d vehicles (carpentum, raeda), the rich h a v i n g a w h o l e escort: ' T h i a s u s s h o w e d disdain f o r his l u x u r y carriages and used to turn his nose u p at their rich furnishings. T h e useless vehicles w e r e pulled a l o n g in the rear o f his retinue, w i t h their curtains i n d i f f e r e n t l y u p o r d o w n , w h i l e he displayed the same little consideration f o r his Thessalian steeds and f o r his Gallic horseteams, creatures o f n o b l e b r e e d and w e l l p r i z e d ' (he p r e ferred his m u l e n a m e d Lucius) ( x , 18). O n the road, the g o i n g w a s r o u g h and there w e r e m a n y obstacles: ' D i d w e h a v e t o cross the r i v e r flowing beside the r o a d ? ' (vn, 18). T h e track is s o m e t i m e s ' h e a v i l y rutted, n o w a morass o f stagnant w a t e r , n o w c o v e r e d b y a l a y e r o f slippery m u d ' . T h i s m u s t o f t e n h a v e b e e n the case in w i n t e r 8 . E l s e w h e r e there w a s 'a lane that w a s a mass o f stones and o b s t r u c t e d b y tree-stumps o f e v e r y k i n d ' (ix, 9) 9 . A t the s t a g i n g - p o i n t , hospitality w a s e x c h a n g e d at the inn (1, 4), and s t o r y - t e l l i n g w a s c o m m o n (ix, 4), b u t c o m f o r t w a s l i m i t e d : ' M y c o u c h , besides b e i n g rather short, had a l e g missing and w a s w o r m - e a t e n ' (1, u ) 1 0 . Self-respecting travellers p u t t i n g u p in t o w n s w e n t t o houses f o r w h i c h t h e y h a d

Evidence from literature and inscriptions

23

letters of introduction (1, 21), for example the military courier who went to a decurion's house (x, 1). At the end of the day, the weary traveller made a few purchases at market and in particular went to recover at the baths (1, 5 ; 1, 24), before giving his hosts 'news of his native country, its leading men, its governor' (1, 26), which was then a common way of learning about current events. Road travelling was indeed arduous. It was little use whiling away the length of a journey by exchanging stories, whether they were true or imaginary: 'He held us so spellbound with an entertaining tale that I came to the end of that long and difficult climb without discomfort and without boredom' (1, 3, 20). Many dangers lay in wait for the traveller both on the road and at the posting-station. There was the misadventure of the trader w h o travelled through Macedonia on business. After nine months of hard work, he came back with plenty of cash, only to be attacked by brigands, then looked after by an old woman who was running an inn (1, 6f.). Then w e have the case of the gardener w h o was going to the neighbouring town and met a soldier (ix, 39) travelling to R o m e by order of his senior officer with a letter for the Emperor (x, 13). The legionary 'asked him in a haughty and arrogant manner where he was taking that unburdened ass', then managed to seize the animal (x, 1) and loaded it with his gear, helmet, shield and javelin. Other awkward encounters took place with dogs and wolves : 'When the farmers on a country estate near which we were passing 11 saw us in such numbers, they thought w e were brigands... They sent after us huge dogs' (vm, 17, to be compared with ix, 36, where w e learn o f ' f a r m dogs, enormous, savage creatures, accustomed to feeding on corpses left in the fields and trained in addition to bite without any distinction all passing wayfarers') ; 'packs of huge wolves had reached the point of besieging the roads and attacking travellers in the same w a y as b a n d i t s . . . Travellers must walk in groups and keep close together' (vm, 15). Sometimes there were meetings with police patrols: 'and behind us comes a squad of armed horsemen' (ix, 9). More often the meeting could be with robbers: 'Don't you realise that the roads are swarming with brigands, starting on your w a y at this time of night?' (1, 1 5 ) 1 2 . Long journeys included sea-crossings: ' M y brother, tell me how your journey has gone, by land and sea, since you left the isle of Euboea' (11, 13) and 'Travelling partly by land, partly by sea, we finally reached Corinth' (x, 19). Such crossings were hazardous and the writer pointed out 'the thousand and one difficulties of the trip and the terrors of the sea' (vu, 6) even when it was only a matter of coastal traffic: 'The passengers, seeing that the night was well advanced, had retired to bed to avoid the rough sea in a small inn near to the coast and the ship' (vu, 7). Regarding the cult of the mare clausum and the navigium Isidis w e may simply refer to Rougé's commentary : Recherches sur l'organisation du commerce maritime en Méditerranée sous l'Empire romain13.

24

Evidence from literature and inscriptions

In spite of all the handicaps so depicted, the Metamorphoses show that people enjoyed travelling, because mobility was a keynote o f society, and also because the author, inquisitive to see everything, liked n e w surroundings. W e may conclude the theme of travel as an aid in character-training, with this reference to the beginning of the Odyssey: 'Wishing to present to us a man of matchless wisdom, H o m e r wrote of him that he had acquired the highest virtue b y visiting many cities and making friends amongst many nations' (ix, 1 3 ) 1 4 . THE SYMBOLISM OF T R A V E L

It is impossible to study the subject o f travel in the literature of Antiquity without recalling the symbolic meaning attached to it, since at least as far back as Parmenides' poem, which was wholly based on the figurative image o f the R o a d and the Gate. T h e R o m a n liked travelling on business or f o r amusement. O n e of Seneca's letters (111, 28) rests on this assumption and might be given the title: 'Journey into the Land o f the S o u l ' : ' Y o u are surprised that y o u have made such a long j o u r n e y and b y such a variety o f routes, yet y o u have not dispelled the g l o o m y heaviness within your heart? Y o u need a change o f soul, not of clime . . . It serves no purpose to be unsettled in this w a y . Y o u ask w h y y o u find no relief in running off. It is because you are still with yourself.' B u t the commonest f o r m o f symbolism was the allegory of the Last J o u r n e y : as early as the Etruscans, the passage into the next life was thought of and represented as a journey, b y land or b y sea. S o m e R o m a n mosaic floors illustrate this crossing over to the Isles of the Blessed. For some sects, the j o u r n e y took place through the air, as in the case of the Pythagoreans, w h o believed in the identity o f the soul and the stars. This explains one important aspect o f the symbolism associated with funerary monuments: the deceased is seated in a vehicle (we must not forget, after all, the importance attributed to the horse as a conductor of souls) and is going past a milestone on which the figure was that o f the number o f years he had lived. This symbol was also used b y the Christian apologists, like Tertullian (De resurrectione carnis, 43), for w h o m the b o d y is compared to a traveller w h o finally comes back to his o w n country. APPENDIX I

A famous text: Horace's journey to Brindisi15 W h e n I left mighty R o m e , I found bed and board in a modest inn at A r i c i a 1 6 . M y companion was the rhetorician H e l i o d o r u s 1 7 , far and a w a y the most learned of the Greeks. N e x t w e reached the Forum of A p p i u s 1 8 , swarming with sailors and knavish tavern-keepers. W e felt lazy enough to cut this stretch in t w o , whereas travellers w h o gird up their loins take it in one g o : the Appian W a y 1 9 is less tiring to those w h o are not in a hurry. There, by reason of the water, which was frightful, m y stomach and I were on hostile terms and I waited with some impatience for m y companions, w h o were dining.

Evidence from

literature and

inscriptions

25

Already the shadows of night were beginning to creep over the earth and stars were pinpointing the heavens 20 . Then slaves bellowed at boatmen and boatmen at slaves: 'Heave to, here', 'You've got three hundred in'! 'Stop! that's enough.' After the fares had been collected and the mule harnessed, a whole hour went by. The cursed mosquitoes, the frogs of the marshland drove off" sleep. Whereupon, after their fill of poor wine, a boatman and a passenger 21 vied in singing of the girl each of them had left behind. At length, the weary passenger fell asleep and the lazy boatman unharnessed his mule and let it out to graze, tying its leading rein to a stone. Then he too started to snore away on his back. And it was already daylight when we found that the boat was not yet under way. So one of us, hot-headed fellow, jumped ashore and, with a stick cut from a willow, belaboured head and back of both mule and boatman. At last, upon the fourth hour 2 2 , no earlier, we came ashore. W e washed face and hands in thy waters, O Feronia 23 ! Then, after, having had breakfast, we struggled along for three miles to reach the foot of Anxur 2 4 , perched on its white rocks gleaming from afar. Here the gallant Maecenas and Cocceius were to meet us, both sent as envoys to deal with important matters and both accustomed to bringing together divided friends. Here, too, I made use of a black ointment on my sore eyes. Meantime up came Maecenas and Cocceius 25 , together with Fonteius Capito, a man of matchless perfection, so that Antony can have no closer friend 2 6 . W e were not sorry to leave Fundi 27 and its praetor Aufidius Luscus, laughing at the knick-knacks with which that old clerk bedecked himself in his crazy vanity, laughing too at his bordered robe and his broad stripe and his pan of charcoal. Then, weary indeed, we made a long halt in the town of the Mamurrae 2 8 , where Murena 2 9 provided us with lodging and Capito with board. Dawn the next day was most delightful: for at Sinuessa 30 we were joined by Plotius, Varius and Virgil. Earth has never borne purer souls, there is no man more attached to them than I. O h ! the embraces and the j o y ! So long as I have my senses, nothing will compare for me to the delight of friendship. The little post-house close to the Campanian bridge 3 1 offered us the shelter of its roof, while the state purveyors dutifully provided wood and salt. Next, at the appointed time, our mules laid aside their panniers at Capua. Maecenas went off for exercise, while Virgil and I retired to sleep, for sport is the enemy of weary eyes and sore stomachs. O u r next stop was at the richly furnished villa of Cocceius, situated above the inns of Caudium. N o w come to my aid, Muse; come and help me to recall briefly the contest of wits between the clown Sarmentus and Messius Cicirrus 32 and to give the ancestry of these two adversaries. Messius has a well-known family in Osca; the former mistress of Sarmentus is still alive 33 . With such forebears they came to do battle. First of all Sarmentus: 'You, I must say, are like a wild horse 34 .' W e laughed. Messius for his part said: 'Agreed' and tossed his head. 'Oh!', said the other man, 'if your forehead had not had its horn removed, what would you not be doing, since you are making such threats, even without it?' The truth is that on the left side of his face a scar had made a nasty gash disfiguring Messius' tousled

26

Evidence from literature and inscriptions b r o w 3 5 . A f t e r many a quip on his Campanian disease and his face, Sarmentus begged him to dance the Cyclops shepherd's dance, saying he w o u l d not need a frightful mask nor the buskins of tragedy. Cicirrus was not behind in this: he asked Sermentus if, when he had acquitted himself of his v o w , he had given his chain as an offering to the Lares? Simply because he was n o w a scribe, the rights o f his mistress had in no wise been taken a w a y 3 6 ; and, finally, w h y had he ever run off, since all he required was a pound of corn, skinny, puny creature that he was ?

A merry company did w e make over that meal 3 7 . Thence w e made our w a y straight to B e n e v e n t u m 3 8 , where, in his eagerness to please, our host, was nearly burned up 3 9 while turning some lean thrushes over the fire. For Vulcan slipped out and the wandering flame spread through the old kitchen, rising to lick the very roof. Then you would have seen the starving guests and the trembling slaves grabbing the meal, and afterwards doing their best to put out the blaze. Thenceforward A p u l i a 4 0 was beginning to reveal its mountains, which I k n o w so well, scorched by the Altino 4 1 and over which w e w o u l d never have been able to scramble, had w e not found a welcome at a country house in the neighbourhood of T r i v i c u m 4 2 , not without smoke that drew tears 43 , for in the hearth were burning damp branches, leaves and all. There, in m y complete stupidity, I waited until the middle of the night for a lying w e n c h ; however, sleep overcame me, impassioned as I still was; and, while I lay on m y back, lustful dreams soiled me. F r o m that point w e were taken 24 miles in Gallic carriages, to halt in a small town whose name cannot be put into verse, though it is easy enough to give some clues 4 4 : the commonest of things, water, is sold there 45 , but the local bread is quite magnificent, so much so that the wise traveller takes a quantity with him as he goes on his w a y , for at Carnusium 4 6 the bread is as hard as stone 4 , while there is not a j u g f u l of water to be had in this t o w n founded in b y g o n e days by brave Diomedes 4 8 . Here Varius sorrowfully took leave of his weeping friends. N e x t w e came to R u b i 4 9 , very weary indeed, for w e had gone a very long stage, with conditions made worse furthermore by the rain. T h e f o l l o w i n g day, the weather improved, but the track was even viler, until w e reached the walls of the little fishing-port of Barium. Then Gnathia, a t o w n built despite the waternymphs 5 0 , gave us occasion to laugh and make pleasantry: for there people w o u l d have it that incense placed on the temple threshold melts without the need of fire. T h e J e w Apella 5 1 may believe it; I do not, for I have learnt that the gods spend their time eternally at repose and that, if there are extraordinary phenomena produced by nature, it is not they w h o , to relieve their boredom, send these things d o w n f r o m their heavenly abode on high 5 2 . Brundisium 5 3 is the end of this long tale and this long journey. (Horace, Satires, 1, v)

Evidence from literature and inscriptions

27

APPENDIX 2

The mountain-passes Augustan Age

across the Alps,

according to Strabo,

the Greek geographer

of the

(iv, 1, 6.) Sextius [conqueror of the Salyes in 1 2 3 - 2 2 BC] drove the barbarians from the whole coastline leading from Massalia into Italy . . . He forced them to withdraw a distance of 12 stadia [2-22km] from the sea, along where the coast offers good harbours, and up to 8 stadia [i-48km] at points where it is rocky, but he presented the people of Massalia with the land thus freed. (iv, 1, 2.) The road [from Iberia to Italy] has one alternative leading directly to the Alps and a shorter one through the land of the Voconces. The coastal route, at first in the territory of Massalia, then in that of the Ligusti, is longer, but the passes that it takes to enter Italy are easier since the mountains are already lower in that area. (iv, 6, 2.) The whole coastline, from the port of Monoecus down to the Tyrrhenian Sea, is exposed to storms and short of harbours, except for shallow moorings and anchorages. Owing to the exceptionally sheer face that the mountains present to the sea, there is only a narrow strip for coastal movement. (iv, 6, 3.) The Salyes were the first of the Transalpine Celts to come under the domination of the Romans, who fought a long war against both this tribe and the Ligyes, for these two peoples barred the way that led to Iberia along the coast. They were indeed so successful with their brigandry on land and sea that this road was scarcely passable even to large armies. Finally, after 80 years of warfare, the Romans established with difficulty free transit over a width of 12 stadia for travellers on official business 54 . (iv, 6, 6.) Formerly Italy was dominated by a number of small tribes, povertystricken and thievish in their ways. Today they have been either wiped out or so completely mastered that the passes leading through their territory, which were formerly few in number and difficult to cross, have now been multiplied and the inhabitants can be regarded as trustworthy, while the roads, thanks to engineering works, are as passable as maybe. Indeed, Caesar Augustus finally suppressed brigandry through road-building . . . In some places, the road is so narrow that it causes giddiness [description of avalanches follows] . . . (iv, 6, 7.) For the traveller coming from Italy, the road first follows the valley of the Salassi. Then it divides into two [at Aosta], one branch going over the Poeninus as it is called, impassable for vehicles nearer the Alpine crest, the other [the Petit-St-Bernard] crossing more to the west through the territory of the Ceutrones . . . The Salassi used to wreak havoc with their brigandry on travellers going through their land in order to cross the mountains. Thus, when Decimus Brutus fled from Modena, they made him pay one drachma per head on behalf of himself and his men, and, when Messala had his winter quarters close by [in 34 BC], he had to pay cash for firewood and for the elm-wood needed to make javelins and practice weapons. These folk even went so far one day as to steal money

28

Evidence from literature and inscriptions

belonging to Caesar Augustus and roll boulders down onto columns o f soldiers, with the excuse that they were constructing a road or building bridges over the torrents. However, to end the matter, Augustus subjugated them completely and had them all sold off as warbooty. He then sent in 3,000 Romans who founded the town o f Augusta . . . and today there is peace in all the surrounding district as far as the top o f the Alpine passes. (iv, 6, 9.) Tiberius and his brother Drusus took one summer [15 BC] to put an end to the frenzied raids o f tribes from beyond the Adriatic. (iv, 6, 10.) In the exchange o f goods between Aquileia and ¡.1«. l^wh called Nauportum the mountain massif o f the Carni is crossed in covered waggons. From that point transport is by water. (iv, 6, 11.) One o f the passes which allows movement from Italy into Transalpine and Northern Celtica is the one leading to Lugdunum through the land o f Salassi. T w o routes are possible, one passable for vehicles over most o f its length, the other going by the Poeninus, steep and narrow, but short (the Little and the Great St Bernard). Lugdunum lies in the centre o f Celtica, o f which it forms in a way the citadel, since it lies at the confluence o f the rivers and is conveniently placed for the different parts o f the country. Agrippa, therefore, made it the point o f departure from the main roads: the first crossing the Cevennes Mountains and ending in the land o f the Santones in Aquitania, then the one down the Rhine and the one to the Ocean, which is the third and which leads to the Bellovaci and the Ambiani, finally the one going into Narbonensis and to the coast o f Massalia, the fourth. But one can also, by leaving on the left Lugdunum and the lands upstream from that town, fork off into the Poeninus itself, cross the Rhone or Lake Lemannus in order to reach the plains o f the Helvetii and thence, by a pass which goes over the Jura mountain, reach the land o f the Sequani and the Lingones, where the road divides into two, one fork leading to the Rhine, the other to the Ocean. (iv, 6, 12.) Polybius names only four ways through: the first through the territory o f the Ligieni, quite close to the Tyrrhenian Sea, then the one traversing the land o f the Taurini and used by Hannibal, next the one taking in the territory o f the Salassi, finally the fourth through that o f the Rhaetii, all four being steep. (Strabo, Geography iv)

3 THE ROAD THE Peutinger

Table

BOOKS

AND THE QUEST FOR THE OLD ROADS

T h e Peutinger Table, a medieval copy o f an ancient map, found at W o r m s at the end o f the fifteenth century, and given in 1508 to Conrad Peutinger o f Augsburg, after w h o m it is named, is today conserved in the Library o f Vienna, Austria. At

Evidence from literature and inscriptions

29

first sight it is something of a disappointment. It would appear to be a late compilation 55 , with endless discussion on its ancestry. There has been an attempt to connect it, in a pedigree linking all the Roman road books, with an official document of the first half of the third century, contemporary with the Forma Urbis in the Capitol, which itself is derived from the map drawn up under the direction of Agrippa and painted after his death under the Portico of Polla in Rome for propaganda purposes 56 . What is immediately noticeable is the geographical distortion affecting the appearance of the rivers and mountains. It is not enough for people to attribute this to the shape of the material on which the map was drawn. It is claimed that the elongation of the map was caused by the long narrow shape of the parchment roll (6.80 by 0.34 m), a format similar furthermore to that of the version in the Portico. The fact is, the road routes themselves are looked upon as being more or less rectilinear and are drawn independently of each other. There is no guide to the orientation of the roads nor to the significance of their bends and twists, so the road stations are not sited in their correct relative positions. Here we must realise once more that the Romans, whose utilitarian outlook is well known, were unconcerned with the scientific representation of land-shapes 57 and simply required their maps to give practical information: road distances, the best way of moving from one place to another, amenities (posting-stations, baths hence the frequency of the symbol indicating bath establishments: namely four buildings in the form of a square, alongside the little illustrations of cities: two towers with pointed roofs). When he has come to terms with these arrangements, the investigator will more willingly accept the Table as it is, without thinking that he is justified from a first view of the map in carving up the information. For, strangely enough, despite its appearance of having been compiled at a late date, the Table enjoys such a reputation that for a province to see one or other of its ancient roads depicted on it is like laying claim to nobility. Hence the many efforts of wasted erudition (we could call them cooked-up, pseudo-philology), aimed at making the Table agree with road details garnered elsewhere that fitted more particularly with fanciful etymologies, inspired, one can only say, by local patriotism. It is certainly possible to accept in the last resort errors in manuscript (examples of figures written one above the other; edges of map portions which have been damaged or where there are omissions at the joins; confusion between different sigla: V having become II through separation of the verticals or, conversely, II transformed into V by bringing together the two strokes, confusion between X and V ; omission or duplication of a figure, with mistakes having been possibly superimposed). Present-day publishers and especially scholars have been too inclined to make corrections, by assuming that lines have been omitted or displaced and that figures have shifted, so that one conjectural mistake has brought about another 58 . W e find one of the most healthy reactions to this attitude expressed by Colonel J . Baradez, whose knowledge of topography and capacity as a field-worker can hardly be in question 59 . He has written somewhat paradoxically : 'Even the dia-

Evidence from literature and inscriptions

3°

orifRiB««

kuta.-xiiut.

TBi«^1

[«Urcom^«3 Hcomtnu"1

toufru*jgK1

L-5farponnd

—T*j -\JUV

¿lb Ucct«*

"Ojuo Durrmebto. 0

ÄUtRtwH»?\ ¿A ill

-«jSt^ohcibtD.^r XJ i

¿«enticum

topofag^ ¿buroduno-

\1tdubU»'-

Ht1ci»«*um

1

\\.vnotH»m xn" I -«OtroTnAguA.

.folortu-. rqi

Littn'^" :

mione.

1—mrrtm. — *•*»«

.t-tfpw. ft)obd iftumnpi •cim

Chv

21M15E5 SCHI'"

Pna o\i'. e ¿AnufWd,

-SdluWtlA.

Liimbtruti

i

Extract f r o m the Peulinger Table

grammatic style of presentation in the Table, linear as it is, gives me entire satisfaction as a former air force navigator with exacting requirements in the standards of maps.' It must be evident to anyone having an overall view of the whole of a country's road network that the Itineraries 'did not aim at showing in what way one travelled from this point to that, but rather that they chose from amongst the many

Evidence from literature and inscriptions

3i

stretches of road those which, at a particular time and for a particular purpose, allowed movement through particular regions'. Let us not forget that there always have been between two points supplementary routes (as for example, the doubling of tracks close to difficult sections: a steep hill or a ford, sometimes with traffic in one direction only) 60 . Thus the road guides indicate only one of the possible methods of reaching Y from X , since an itinerary may be reckoned to have included different sections of several roads. The difficulties often encountered in working out distances may be explained by the fact that we are dealing with texts compiled at different dates and also because in the course of time changes of line took place. A road may be compared to a vibrating cord, which is taut when the traffic is heavy (with straightening due to short cuts), but is loose and sinuous at slack periods. T o become aware of the secret rhythm in the inner life of the countryside, it is enough to take notice of the present-day down-grading of many minor roads. 'The Peutinger Table, therefore, describes particular routes involving one after another a series of stretches from the existing network of roads and of connections between these roads. It does not depict any single road. It does not depict either all the roads' (J. Baradez). The normal scale of the archaeological maps that have been published (1/500,000 or even 1/1,000,000) is anyway deceptive. It is not enough to state that a road goes through two points A and B and then to join them up to obtain an exact line on the ground. Drawing a map of the Roman roads of an area has too often been simply a matter of linking the cantonal capitals, which results in beautiful star patterns. Hence the reputation of Roman roads for being straight, which is correct only in flat country and in the case of the great strategic trunk routes. Elsewhere the straight line very often breaks down. Much fruitless discussion about the distances shown on the Table - and these are missing for many cross-routes - arises therefore from a misunderstanding of the actual course of the road on the ground. Let us not forget either that some sections of the Itineraries may correspond to movement by water and that the intervals between stations are given to the nearest mile, so that an error of up to half a mile is possible. It may be that the map is showing the milestone and not the actual distances: Goodchild 61 is of the opinion that lists of milestones were used in compiling the road routes. As regards the identification of the stations 62 , one of Colonel Baradez's revealing suggestions is that the Table 'does not always indicate the stations or the population centres that are passed through, these lying frequently off the road, but rather the crossroads and forks 63 where uncertainty or an error in direction were to be feared'. It is from these cross-roads, which the Table brings out well, and not from the point of arrival or departure as such, that some distances should be calculated. There is confirmation of this idea in that the symbols do not always correspond to important places: the network that is depicted does not show the complete network of the Roman roads in the Empire, but those used by the Imperial Post 64 . Audin has noted that the Itineraries are set out so as to avoid difficult slopes, that is, as an aid to heavy transport. Finally, the Table must not be studied on its own, but in conjunction with other

32

Evidence from literature and inscriptions

road lists, of varying date and purpose, particularly the Antonine Itinerary, a more accurate compilation, it seems. Most important, all possible resources must be brought to bear on the problem: literary texts, inscriptions, archaeological discoveries, place names and - this is something that must be emphasised - examination o f the terrain, as it was in the past and as it is today (maps and air photographs), not forgetting fieldwork. A recent publication, which looks at the Peutinger Table against the wider background of the map w o r k of antiquity 6 5 , has given a detailed report on the 55$ symbolic illustrations o f the road amenities. These are: in particular, a double tower o f which only the front is visible (429, with 79 variants and 4 main types); sometimes a half-circle seems to suggest a surrounding ditch; a building in the f o r m of a temple, o f which only the front and one o f the sides can be seen (44). Must w e take it that this representation indicates the existence of a real temple?

Evidence from literature and inscriptions

33

(a) 2 towers linked b y a central block (b) independent towers lacking a superstructure (c) twin towers (d) three towers (e) temple-like buildings (f) bath buildings (g) horrea (h) t o w n o f Aquileia (i) t o w n o f Ravenna (j) R o m e and St Peter (k) port o f Claudius at Ostia (1) pharos (Bosphorus Alexandria)

'spas', bath establishment (and pump-room?) of square plan, of which we have a bird's-eye view (52) and comprising a central open courtyard with buildings along the four sides; a composite building, also seen from above (10), made up of two or more parallel units, sometimes of differing length, symbolising horrea; city-walls of circular shape, seen in perspective (6); three larger illustrations, personifying the metropolises of Rome, Constantinople and Antioch; two ports, shown as semi-circles (Ostia, Fossae Marianae); lighthouses (Ostia, Alexandria); various: altars (at the ends of the known world), a tunnel between Naples and Pozzuoli (Crypta Neapolitana). W e also notice a few landscape details (mountains, trees). Some of the little pictures are clearly additions, made in the course of later editing: at one time, there seems to have been a shift in emphasis towards the East. The authors think that this can help with the dating. There appear to have been two phases earlier than the copy as we see it: one belongs to the beginning of the third century and was edited in the West:

34

Evidence from literature and inscriptions the details are s h o w n o n a larger scale because they w e r e closer to the p e n m a n 6 6 ; the o t h e r is dated to the end o f the f o u r t h c e n t u r y or the b e g i n n i n g o f the fifth, to w h i c h p e r i o d can b e assigned the six w a l l e d t o w n s , the illustrations o f C o n s t a n t i n o p l e and o f A n t i o c h , as w e l l as o f the c h u r c h o f St Peter, and a f e w Christian entries. T h e l a w s o f the Theodosian Code bear witness to the great interest then taken b y the E m p e r o r s in the cursus publicus. T h i s second stage o f the m a p p r o b a b l y dates f r o m T h e o d o s i u s n, w h o is t h o u g h t to h a v e carried o u t terrestrial m e a s u r e m e n t . T h e g o o d relations that the Eastern R o m a n E m p i r e e n j o y e d at the b e g i n n i n g o f the f i f t h c e n t u r y w i t h Persia appears to e x p l a i n the inclusion o f certain illustrations d e p i c t i n g stations in the East. T h e m a p certainly reflects the e x t e n t o f g e o g r a p h i c a l k n o w l e d g e and the administrative organisation o f the E m p i r e .

T h e r e d o n o t seem to h a v e b e e n a n y m e d i e v a l alterations: f o r in that case, o n e w o u l d n o t i c e typical additions such as monsters, m o r e mistakes in the n a m e s and m o r e f a n c i f u l t r e a t m e n t o f the illustrations. A f e w details h a v e b e e n c h a n g e d in respect o f the three great cities o f R o m e , C o n s t a n t i n o p l e and A n t i o c h , the s y m b o l s f o r w h i c h r e m i n d us o f t h i r t e e n t h - c e n t u r y miniatures (the C o l m a r m o n k ?).

A ROAD BOOK: THE Itinerarium

Antonini

Augusti

F o r each r o u t e q u o t e d , this r o a d b o o k g i v e s a list o f mansiones w i t h their distances 6 7 . T h e w o r d ' A n t o n i n e ' seems to refer to C a r a c a l l a ( 2 1 2 - 1 7 ) , a date c o n f i r m e d b y g r o u n d e v i d e n c e f r o m the roads o f B r i t a i n and the l e g i o n a r y fortresses. B u t the w h o l e w o r k p r o b a b l y dates f r o m the end o f the c e n t u r y , j u d g i n g b y the m e n t i o n o f titular offices o f the t i m e o f D i o c l e t i a n and M a x i m i a n (284-305). S o m e scholars c l a i m to h a v e f o u n d in the Itinerary references to the raids o f 256 and 275 in G a u l 6 8 . T h e t e x t also includes later additions. In this respect w e c o u l d m a k e the same c o m m e n t s as w e r e m a d e f o r the Peutinger Table. O f course, there are m a n y mistakes, especially in the figures (and the successive use o f entries in miles and leagues has n o t y e t exhausted the i n g e n u i t y o f c o m m e n t a t o r s ) , b u t those w h o use the t e x t w i t h care c o n t i n u e to c o n f i r m the data in the Itinerary. B e f o r e c o r r e c t i n g the t e x t because w e can see n o t h i n g b u t inaccuracies, contradictions, repetitions, and gaps, w e should start b y c o n s i d e r i n g it in its o r i g i n a l state b y p l a c i n g it a l o n g s i d e the i n f o r m a t i o n f r o m the o t h e r r o a d sources. It m u s t n o t b e f o r g o t t e n that l i g h t w i l l d a w n m o s t o f t e n f o l l o w i n g a c o m p a r i s o n w i t h m a p s o l d and n e w , w i t h air c o v e r and w i t h the ground itself69. A c c o r d i n g to K u b i t s c h e k , the Itinerary is based o n a w a l l - m a p o f the t i m e o f Caracalla and so w o u l d seem to be a c l u m s y c o m p i l a t i o n d r a w n u p in the first years o f the reign o f D i o c l e t i a n , b e t w e e n 280 and 290, w h e n he w a s r e o r g a n i s i n g the annona. V a n B e r c h e m ' s v i e w 7 0 is that the Antonine Itinerary, w h i c h is a p p a r e n t l y m o r e c o n c e r n e d w i t h listing the greatest possible n u m b e r o f places than w i t h satisfying the real needs o f travellers, is in truth a list o f mansiones c o n n e c t e d w i t h the

Evidence from literature and

inscriptions

35

ITER AB AQVILEIA BONONIAM M. P. CCXVI Sic CONCORDIA ALTINO PATAVIS ATESTE ANNEIANO VICO VARIANO VICO SERNINO MVTINA BONONIA A VERONA BONONIA HOSTILIA COLICARIA

M. M. M. M. M. M. M. M. M. M. M. M.

P. P. P. P. P. P. P. P. P. P. P. P.

XXXI XXI XXXII XXV XX XVIII XX XXIII XVIII CVSic XXX XXV

MVTINA M. P. XXV BONONIA M. P. XXV A VERCELLIS LAVDE M. P. LXX Sic LAVMELLO M. P. XXV TICINO M. P. XXII LAVDE M. P. XXIII A CREMONA BONONIA M. P. CXII Sic BRIXELLO M. P. XXX REGIO M. P. XL MVTINA M. P. XXVII BONONIA M. P. XXV

ITER A FAVENTIA LVCAM M. P. CXX Sic IN CASTELLO ANNEIANO FLORENTIA

M. P. XXV M. P. XXV M. P. XX

PISTORIS LVCA

M. P. XXV M. P. XXV

ITER A PARMA LVCAM M. P. C VIA

CLODIA

ITER A LVCA ROMAM PER CLODIAM M. P. CCXXXVIII Sic PISTORIS FLORENTIA AD FINES SIVE CASAS CAESARIANAS ARRETIO AS STATVAS CLVSIO

M. P. XXV M. P. XXV M. M. M. M.

P. P. P. P.

XXV XXV XXV XII

VVLSINIS FORVM CASSI SVTRIO BACCANAS ROMA

M. M. M. M. M.

P. P. P. P. P.

XXX XXVIII XI XII XXI

ITER A ROMA FORO CLODI M. P. XXXII ITER AB ARI MINO DERTONAM M. P. CCXXIX Sic CVRVA CAESENA FORO LIVI FAVENTIA FORO CORNELI CLATERNA BONONIA MVTINA REGIO

M. M. M. M. M. M. M. M. 3

P. P. P. P. P. P. P. P.

XX XIII X X XIII X XXV XVIII

TANNETO PARMA FIDENTIA FLORENTIA PLACENTIA COMILLOMAGO IRIA DERTONA

Lists o f stations taken f r o m the Antonine

M. M. M. M. M. M. M. M.

P. P. P. P. P. P. P. P.

Itinerary

X IX XV X XV XXV XVI X

Evidence from literature and inscriptions

36

l e v y i n g o f the annona. A m o n g the i m p o r t a n t roads that he quotes (some often i n v o l v i n g detours, others being repeated or confused) there is a v e r y l o n g one g o i n g f r o m R o m e to E g y p t d o w n the D a n u b e and across Asia M i n o r , reminiscent o f a military expedition, in this case the j o u r n e y to the East m a d e b y Caracalla (214-1$). T h e route in question w o u l d correspond, furthermore, not to the itinerary actually f o l l o w e d , but to the one that had been prepared. Such m o v e m e n t s had indeed to be m a d e ready w e l l in advance and edicts sent to the provincial

4

M a i n roads in Gaul, a c c o r d i n g to the Antonine Gaule,

Itinerary (after E. Desjardins, Géographie

de la

v o l . i v , pl. ii.)

g o v e r n o r s w o u l d f i x the successive road stages, indicate the stay at each station and list orders for the repair o f roads (cf. the discovery o f milestones along this route) and for the victualling at the mansiones o f the Imperial court and the troops. W e are aware o f this t h r o u g h the Life o f Septimius Severus in the Historia Augusta (45-47), w h i c h can be set against a text o f D i o Cassius ( L X X V I I , 9): ' W e w e r e obliged, w h e n ever he left R o m e , to m a k e ready for the E m p e r o r , at our o w n expense, a c c o m m o dation o f all kinds and sumptuous resting-places m i d - w a y , even o n short j o u r n e y s , at places w h e r e he never stopped and even at s o m e w h i c h he w a s n o t g o i n g to see. In addition, w e used to build amphitheatres and circuses w h e r e v e r he w a s spending or h o p e d to spend the winter, w i t h o u t receiving f r o m h i m any compensation.'

Evidence from literature and inscriptions

37

The Itinerary from Rome to Egypt on this argument was part of an edict issued by Caracalla on the eve of his campaign and the Antonine Itinerary a collection of such edicts, of varying date and importance 7 1 , but all governed by events. According to the names borne by some towns, the compilation was probably made around 280 to 290, that is, during the years when Diocletian reorganised the annona. Advantage was taken of the experience gained in raising this tax and, by making use of the files in the office of the praetor, itineraries were grouped together, that is, lists of mansiones that had been made ready. They were probably classified in a certain order of merit and to round matters off, for those provinces which until then had not been involved in troop movements, new lists would have been prepared of mansiones already in service or that needed to be built. Finally, to give the collection a title, it received the heading of a famous edict of Caracalla. Such is the most recent and most attractive theory of the origin of the Itinerary. Itinerarium Burdigalense sive Hierosolymitanum (ITINERARY FROM BORDEAUX TO JERUSALEM OR HIEROSOLYMITAN)

This record of a pilgrimage 72 to the Holy Places occurred during the consulate of Dalmatius and Zenophilus (333) and mentions the stations that were passed through, no doubt according to a map or an itinerary. The section in Gaul comprises 371 miles, 30 changes or posting-stations, 1 1 mansiones or staging-points: Bordeaux, Bazas, Auch, Toulouse, Narbonne, Via Domitia, Aries, Valence, Aosta, Die, Luc-en-Diois, Mont-Saleon, Gap, Suse . . . Jullian 73 has pointed out that the road going through Aquitaine abandons the old way along the valley in order to pass along the heights, free from the danger of floods and encounters, and corresponds to the 'chemin Gallien', which possibly dates back to the Emperor of the same name (254-68). The Hierosolymitan Itinerary gives a detailed account of the crossing of the Alps and so allows correction of both the Peutinger Table and the Antonine Itinerary. There has been discussion w h y the route makes detours in places and perhaps this was to allow visits to local chapels containing holy relics. The places involved might not be mere relaying-inns, but the homes, either contemporary or earlier, of fellow-believers. THE

Peregrinatio Aetheriae

A N D THE FIRST A C C O U N T S OF PILGRIMAGES

The Journey of Aetheria is the account, discovered in 1884 at an Arezzo convent, of a pilgrimage to the Holy Places, undertaken around the year AD 400, by a woman of rank, probably a nun. She writes to her 'sisters' in a simple, affectionate and lively manner, but lacks both literary knowledge and artistic sense. The surviving fragment is concerned with Sinai, Jerusalem, Mount Nebo, Mesopotamia, Antioch, Cilicia, Cappadocia, Galatia, Bithynia and Constantinople, where the authoress did her writing, displaying interest in the places mentioned in the Scriptures and in legend. She thereby gives us much topographical information, as well as the account

38

Evidence from literature and inscriptions

of long marches across the desert, encounters with hermits, conversations with bishops, impressive ceremonies. The above two texts may be considered to be the beginning of a long series74 that we shall meet again in dealing with medieval sources.

4 GEOGRAPHICAL PTOLEMY'S

WORKS

Geography

Claudius Ptolemeius was a mathematician and astronomer of Alexandria (first half of second century AD) who drew up general maps covering the most important towns, rivers and mountains and their astronomical coordinates. A Geography bearing his name has come down to us, a Byzantine compilation of the tenth or eleventh centuries, including material derived from Ptolemy and other geographers, together with maps which are probably of later date. The most ancient copies are those of Vatopedi (the monastery of Mount Athos), n. 754, belonging to the end of the twelfth century or the beginning of the thirteenth; and the later Vaticanus Urbinas Graecus 82. On the medieval maps revealed by these documents, the names of the towns are enclosed within an irregular quadrilateral having embattled walls, sometimes with towers, a genuine symbol illustrating ramparts and going back to the Roman period. The size of these symbols varies and corresponds to the classification of towns made by Ptolemy according to how far he was certain of their geographical position. The maps that accompany modern editions show us, not the roads (Ptolemy's view is that these are the province of the geographer proper), but some topographical information, the political and administrative divisions and a detailed list of town names. The maps have been drawn by Mercator, the Dutch geographer, mathematician and engraver (1512-94), and are based on lost versions going back maybe to the Greek Agathodaimon (fifth century). Ptolemy himself used road books and descriptions of itineraries: it is perfectly possible to argue from his information where data occur in a significant line, but the geographer has missed out details that would enable the interrelationship of itineraries to be established. THE Ravenna Cosmography This compilation, made by a monk in the eleventh century - he called the via Aemilia 'Imperialis Estrata' - summarises documents that certain place names suggest date back to the fifth century. The fifth century copy was itself lifted from a road-map, a more complete one than the Peutinger Table , but these documents are very corrupt (some names have been arbitrarily cut out) although no earlier than the Late Empire. Besides the lists of stations - shortened at some point in the transcription - the Cosmography provides river names 76 and a few landscape details. The

Evidence from literature and inscriptions

39

information f r o m the road-map appears in t w o f o r m s : names o f the stations listed one after another in the same order as f o r the itineraries, which are carefully identified; and names inserted into the lists o f towns, amongst those coming f r o m other sources. C o u l d one of these sources have been signposts erected where travellers left towns and listing all the towns ahead, just as today, on the w a y out o f a built-up area, destination boards display a number o f place names en bloc without showing that it will be necessary to take different forks in order to reach them? Set against the Table and the Antonine Itinerary, the Cosmography enables us to complete some of the identifications.

5 MILESTONES T h e w o r d for milestone, milliarium, came f r o m mille11, a thousand, the unit o f distance being a thousand R o m a n paces, 1,481-5 metres or 1 , 6 1 1 English yards. T h e R o m a n pace or double-step {passus\gradus= step) was equal to i-48m, that is almost 5ft, but in the beginning there was apparently an attempt to establish the equivalents o f Greek measures: the mile corresponded to ten Greek stadia, which m a y have been the case o f the via Appia, then to eight stadia 78 for the Greek settlements in Southern Italy, which may have affected the via Domitia. W e shall consider later the problems raised b y the Gallic and R o m a n leagues and h o w it is possible to calculate the position o f vanished milestones, with the additional help o f place names. It is sufficient here to notice that in medieval documents milestones m a y be quoted as marking the boundaries of estates or o f administrative areas. Lebel has shown that along Agrippa's road f r o m Lyons to Langres there are striking coincidences between the siting of milestones and lanes f o r m i n g the boundaries o f registered land. As far as possible, milestones were placed in prominent positions, such as hill-tops and bridges, which have f o r long served as boundary marks. It is c o m m o n knowledge that the stones take the shape o f cylindrical or ovalsectioned columns, occasionally parallelepipeds, and are made o f limestone, sandstone, granite or basalt. T h e y stand on a square projecting base to allow a firm setting. T h e height varies f r o m 2 to 4m and the diameter f r o m 0-5 to o-8m, the smaller diameters seeming to be earlier. Some milestones, erected against a rock or a building, have their rear face shaped. Milestones were illustrated in antiquity - for example, there are three Gallic relief sculptures showing a vehicle passing in front o f a milestone 7 9 - and in addition they are mentioned in literature 8 0 . A typological comparison could be made between R o m a n milestones and those o f recent origin. There were also lapides tabularii, stadia stones 8 1 , the equivalent o f modern French 1 0 0 m stones, which used to bear letters in place of figures. A connection with the w o r d meaning messenger (tabellarius) has led to these being interpreted as mounting stones for horsemen 8 2 . T h e number o f milestones inscribed in Latin amounts to some 4,000, with as many

40

Evidence from literature and inscriptions

in Greek, and more than 2,300 o f these are in Africa, 600 in Italy, the same number in Gaul and Germany, about 100 in the Balkans and 95 in Britain 8 3 . Confronted b y this mass of valuable evidence the publication of which is either inadequate or dispersed (until n o w the reports have been scattered through the different volumes o f the C I L under the heading 'via publicae') the Institute o f Ancient History and Epigraphy in the University o f Berne, in cooperation with the A c a d e m y o f Sciences in Berlin, has undertaken the publication of sections o f V o l u m e x v n o f the C I L , which will bring together all the milestones. Furthermore n e w discoveries have added something like 30 per cent to the corpus and many studies made in the past prove inaccurate when the stones are re-examined. A card index has been prepared, using various types o f on-the-spot survey, always a tricky matter with curved surfaces: photographs, moulds in special papier mâché or in latex, hand tracings. Progress is being made with distribution maps to bring out dating differences, which will provide historical information o f the greatest importance on the phases of road construction, f r o m 252 BC, the date o f the earliest k n o w n milestone 8 .

5 A milestone from northern Italy C I L HI, 5708 = ILS, 464, S. Lorenzo di Sabato (photo: R . Chevallier)

A t the same time, a series of summaries is being edited, under the title o f Itinera romana, the first of which deals with Swiss milestones, the second with the road over the Alpis Poenina. B y 1968 inventories had been completed for Lusitania, Gallia

Evidence from literature and inscriptions

4i

Narbonensis, the Three Gauls, and the Germanies and Norica, as well as for part o f northern and central Italy, Britain and Pannonia. T h e inscriptions on the milestones m a y have been engraved directly onto the rounded part o f the column or else within a specially flattened space, which is depressed or raised. S o m e stones are not inscribed, these being earlier or examples which there had been no time to engrave (many kilometre-stones along French roads have been waiting to be inscribed for years). It is also v e r y likely that some w o r d i n g was painted on, especially since many signposts w e r e o f w o o d , while besides the official stones there were private ones. T h e normal inscription includes the name of the builder or restorer of the road, the t w o operations being sometimes purposely confused. This name is in the n o m i native case, where the person responsible himself carried out the w o r k or in the dative case more and more, when there is a dedication b y the civites, or in the ablative for the date, the name o f the executor of the w o r k being added, possibly an imperial legate {per talem or curante so-and-so). There are regional distinctions. T h e dependent genitive is a late feature. T h e w o r d i n g o f the titles usually provides an accurate date, with restoration being possible where inscriptions are incomplete or difficult to decipher 8 5 . In m a n y cases all w e have are fragments and old drawings. T h e dating evidence is a little less certain w h e n w e examine the positioning o f the inscription, the size o f the space on which it is engraved, the shape o f the lettering and the forms o f address used. In the later milestones where the Emperor's titles are shortened, there is a dedication in the dative (and not a date in the ablative). It ought to be possible to identify different workshops according to the quarries that w e r e worked. T h e stones display the distance between the spot at which they were set up (which is never mentioned) and the point of departure or arrival as it m a y be, the latter being possibly the centre of the town (in R o m e , the 'golden milestone'), the forum or, most often, a gate 8 6 . T h e statio may also have been at the city limits or at a place where a branch road left the main h i g h w a y . T h e place to or f r o m which the distance is measured is often a t o w n , as in the case o f the Lutetia milestone, which was found alongside a road leaving Paris, perhaps the one to Sens, and which, in its inscription o f the time o f M a x i m u s Daia (305-9), mentions the civitas Parisiorum, the first appearance of the name of Paris. Distance might also be measured f r o m the boundary o f a civitas (with the w o r d i n g ad fines81) or f r o m a geographical feature like a river or a pass. T h e distances shown are sometimes multiple, expressed in miles (MP) or leagues (L) 88 . Calculation makes it clear that the latter unit had t w o values, representing either the 'Gallic league' (with values of 2 , 3 3 8 - 2 , 4 0 0 - 2 , 4 1 0 - 2 , 4 3 0 and 2,475m) or the ' R o m a n league (2,208 to 2,220m). It is understandable to find in Gaul a local unit of measurement 8 9 used in unromanised districts and such use o f the league along local roads, possibly involving native entrepreneurs, could be associated with the revival, in the time o f Trajan and above all the Gallic emperors 9 0 , o f native customs as well as local cults and pottery types. T h e league o f 2 , 2 2 2 m is equivalent to R o m a n miles

42

Evidence from literature and inscriptions

(3,000 single paces) 9 1 . It was made official b y Caracalla. T h e r e w e r e also a f e w other local units o f distance, an A f r i c a n mile o f 1,625 m in the n e i g h b o u r h o o d o f Gabès and a milia in Spain o f v a r y i n g length. A selection o f inscriptions f r o m milestones is n o w given b y w a y o f example : T h e Sacquenay stone 43 bc ( C I L , xiii, 9044) Tiberius Claudius Drusi filius Caesar Augustus Germanicus, P o n t i f e x M a x i m u s , Tribuniciae Potestatis III, Imperator III, Pater Patriae, C o n s u l III, Consul designatus I V , A n d e m a n t u n n o [Langres] millia passuum X X I I ( = 32-560km, the correct distance). A milestone f o u n d 2 0 k m f r o m the Porta N i g r a at T r ê v e s ( C I L , xiii, 9 1 2 9 , dated to 212: [Imp(eratori) Caes(DW) M(îjîto) Aurelio S]evero [Antonino P]io Felici [Aug(«sio) Arjabico, Adiabenico, Parthico max(i'mo), Brit(annico) max(i'mo), pontifi(ia) max(imo), trib(unicia) pot(estate) X V , imp(eratori) II, co(m)s(m/j) III, p(atri) p(atrige). A b Augusta T r e v (erorum) leug(iis) n o v e m . A stone f r o m S i r m i u m (date 3 5 4 - 5 5 ) , I L S , 7 3 2 ( C I L , m, 3705) : m . p . V . I - Imp. Caes. Fia. Iul. / Constantius pius fel. / A u g . victor m a x i m u s / trimfator aeternus, / divi Constantini optimi / m a x i m i q u e principis [filius understood], divo/rum M a x i m i a n i et / Constanti nepos, divi / Claudi pronepos, ponti/fex m a x i m u s , Germanicus / A l a m a m n i c u s [sic] m a x i m u s , / G e r m , m a x . , Gothicus I m a x i m u s , Adiabin. [sic] m a x . , / tribuniciae potestatis / X X X I I , imp. X X X , consuli V I I , I p.p., proconsuli, viis muni/tis, pontibus refecti [sic], / recuperata re publica, / quinarios lapides per Il/lyricum fecit, / ab Atrante ad flumen / S a v u m milia passus / C C C X L V I T h e E m p e r o r Caesar Flavius Julius Constantius, h o l y , fortunate, Augustus, v e r y great victor, eternal triumpher, son o f the divine Constantine, v e r y great and noble first citizen, grandson o f the divine M a x i m i a n and Constantius, greatgrandson o f the divine Claudius [11 Gothicus], v e r y great pontiff, Germanicus, v e r y great Alamannicus, v e r y great Germanicus, v e r y great Gothicus, v e r y great Adiabenicus, in his 32nd tribunician p o w e r , greeted 30 times as emperor, consul 7 times, father o f his country, proconsul, after having built the roads, remade the bridges, restored the c o m m o n weal, he has set up across Illyria milestones every 5 miles f r o m the river Atras to the river Sava 364 miles. S o m e o f the details make this an interesting inscription: the complete list o f all the E m p e r o r ' s titles, the embellishments o f the traditional f o r m o f address and the craftsman's errors, hardly a surprise at that late date and in so remote a province. T h e formulae used v a r y and m a y be m o r e or less suitable f o r inscription : refecit (et restituit) ; v i a m faciendam curavit ; v i a m ante hac lapide albo inutiliter stratam et corruptam silice n o v o quo f i r m i o r

Evidence from literature and inscriptions

43

commeantibus esset, per milia passuum X sua pecunia fecit (ILS, 5822, year 216) viam Juliam Aug(ustam) a ilumine Trebia quae vetustate interciderat sua pecunia restituit (ILS, 5823, year 125); silice sua pecunia stravit; viam Aemiliam vetustate dilapsam operib(Ms) ampliatis restituendam cur(avit) (ILS, 5824, year 143). Elsewhere one reads: viam diutina incuria prorsus corruptam, (ILS, 5876) or amissam (ILS, 5875); viam munit per millia passuum X (ILS, 5829) ; redactam in formam provinciae Arabia viam novam a finibus Syriae usque ad mare Rubrum aperuit et stravit per C . Claudium Severum leg.Aufg. pr.pr.] (ILS, 5834, year 1 1 1 ) ; miliaria vetustate dilapsa [or conlapsa] restituit (ILS, 5849); miliaria novae praetenturae poni iusserunt (ILS 5850) ; miliaria) commeantibus innovavit (ILS, 5853); miliaria orbis sui restituit (ILS, 5855) ; viam stratam novam instituerunt (ILS, 5861); substructionem contra labem montis fecit (ILS, 5856) ; vias exaustas restituit ac novis munitionibus dilatavit (ILS, 5862). Some inscriptions from other sources, like bases or cippi, nevertheless come close to milestones and signposts in their wording : viam prolapsam nova substructione restituit (on a base, ILS, 5827) ; viam diu neglectam . . . usui publico reddidit (ILS, 5888, in a vault). W e may refer to a stone from the road from Capua to Reggio, known as the milestone of Popilius Laenas, consul in 132 BC, but attributed by Bracco to the consul of 153, T . Annius Luscus 92 . It is in fact a limestone tablet, bearing what amounts to an elogium, in language that is still archaic: viam fecei ab Regio ad Capuam et in ea via ponteis omneis, miliarios, tabelariosque poseivei. Hince sunt Nouceriam meilia LI Capuam X X C I I I Muranum L X X I V Consentiam C X X I I I Valentiam C L X X X ad Fretum ad Statuam C L X X X I Regium C C X X X V I I . Summa af Capua Regium meilia C C C X X I . Et eidem praetor in Sicilia fugiteivos Italicorum conquaeisivei, redideique homines D C C C C X V I I . Eidemque primus fecei ut de agro poblico aratoribus cederent paastores. Forum aedisque poplicas heic fecei. The man responsible for the work takes pride in having laid out the road, built a forum and handed back the pastures of the ager publicus to the tillers of the soil. The terminology used is an excellent illustration of the economic importance of the main highways, acting as tangible axes in the settlement and development of the land. It is also worth referring to CIL, xii, 1524, which commemorates the opening of a road by the praetorial prefect for the Gauls, Dardanus, w h o was a correspondent of St Augustine, in whose honour the name of Theopolis was given to the Chardavon basin in the Basses-Alpes. Other inscriptions provide us with valuable information on the roads, the

44

Evidence from literature and inscriptions

engineering works and the buildings alongside. T h e following are some typical examples, chosen f r o m the Inscriptiones Latinae Selectae by Dessau. A m o n g engineering works, bridges take first place : ad mil(iarium) X X X V (de la via Caecilia) pontem ; viam et pontes pecunia sua ($866) ; pontes fecit veteres restituii (5887) ; viam imbribus et vetustate conlapsam cum pontibus restituit (5869) ; pontem peilas faciundum coiravere (5896); pontem faciendum locarunt eidemque probarunt (5897) ; pontem lapideum de suo f(aciendum) c(uravit) (5898); pontem fluvi a solo restituerunt et transitum reddiderunt (5899) ; pontem ligneum qui per multo tempore vetustate conlapsus adque destitutus fuerat, per quo nullus h o m i n u m iter facere potuerat, providit, fecit, dedicavitque (5902) ; pontem vetustate corruptum in usum cursus publici restitui aptarique iusserunt (5903)O t h e r inscriptions mention schemes protecting roads against flood damage, a matter of vital concern and rightly so, to the R o m a n engineers. viam inundatione aque (sic) interruptam restituit (5859); viam Anniam longa incuria neglectam influentibus palustrib(w.s) aquis eververatam sic et commeant(i'ÒMs) in viam . . . restituit ($860); vias per fines C e u t r o n u m vi torrentium eversas, exclusis fluminibus et in naturalem alveum reductis molibus pluribus locis positis (5868) ; viam torrentib(w5) exhaustas restituit ac novis munitionibus dilatavit (5871, cf. also 5887). Similar to these raised causeways are the built-up embankments : substructionem contra labem montis fecit (5856) ; m u r u m ad defension(em) viae vetustate conlabsum restituerunt (5858). M a n y inscriptions are found carved on rock-faces left after engineering w o r k : montibus excisis anconibus sublatis viam fecit (5863) ; viam fluminis vi abruptam interciso m o n t e restituerunt (5864) ; montibus inminentibus Lyco flumini caesis viam delatavit (5865) ; m o n t e m cecidit et viam et sessionem d(e) s(ua) p(ecunia) f(ecit) (5883); hoc iter ubi homines et ammalia c u m periculo commeabant apertum est (5885) ; titulum i m m a n e m , m o n t e m Alpinum ingentem litteris inscripsit (5886). A series of inscriptions recalls the building ofstationes, stabula or horrea. T h e following are examples : stationem a solo fecerunt (5905) ; stabulum ne animalia cursus publici longi itineris labore diutius deperirent, providit, constituit, aedificavit adque dedicavit (5906) ;

Evidence from literature and inscriptions

45

locum ruinis obsitum, qui antea squalore et sordibus foedebatur, ad necessarium usum et ad peregrinorum hospitalitatem in meliorem (aditum) et aspectum propria pecunia reformavit (5907) ; horrea ad securitatem populi Romani pariter ac provincialium constructa omni maturitate dedicavit ($910); horrea fortia et felicia (5912) ; horrea compendiaria armaria (5914). B u t also at the roadside w e r e to be found : altars: Triviis Quadriviis [goddesses at crossroads with three or four forks] ceterisque dibus aram (5886) ; temples and bath-houses: tempia et balneas pecunia sua restituit (5868). S o m e inscriptions, not the least interesting, mention the authorities w h o w e r e responsible for the construction w o r k , gifts b y public or private bodies and other details o f financing : curante . . . legato pro praetore (5828); per talem legatum Augusti pro praetore (5834) ; per legionem III Augustam (5 83 5) : paving o f the road f r o m Carthage to Theveste in 1 2 3 BC) ; per legionem III Gallicam (5865) ; Curante . . . legato Augustorum (5847); per talem proc(uratorem) suum (5851); Centurio legionis qui operi institit (5864A) ; curator viarum sternendarum viam sua pecunia fecit (5881) ; curator viarum et pontium U m b r i a e et Picini (5891) ; curante quaestore (5886); hanc viam derectam per centurionem post sententiam dictam ab legato . . . restituit . . . primipilaris (5888); award o f contract and 'manceps' (5799) ; w o r k done b y roadside dwellers : via nova strata per possessores territori Cirtensium (5873); H S . . . quae possessores agrorum contulerunt ; or a m o n g private individuals, a mother in honour o f her son: ob honorem eius in via ducente Herdoniastria milia passuum ex d(ecreto) d(ecurionum) intra lustrum honoris eius repraesentata pecunia stravit (5878); use o f taxes or sums of m o n e y : via munita de vectigali rotari (5874) ; reddito ordinario vectigali (5876); sums given ex decreto decurionum (5882, cf. 5878). Sometimes there was an address o f w e l c o m e to travellers : ' M a y he be invigorated in mind and b o d y w h o , across thee, O most venerable Mustis, makes a pleasant j o u r n e y , through thy ever-varied landscape! 9 3 .'

46

Evidence from literature and inscriptions

T h e distribution and dating o f milestones clearly t h r o w m u c h l i g h t o n general h i s t o r y , f o r their inscriptions, w h i c h it is interesting to c o m p a r e w i t h the l e g e n d s o n coins, are v e r y i n f o r m a t i v e o n I m p e r i a l p o l i c y and still m o r e so o n the political a t m o s p h e r e o f the times 9 4 . W e m u s t bear in m i n d that s o m e w e r e r e c a r v e d , rew o r k e d or replastered. In the first place an inscribed stone bears witness to the d e v o t i o n and l o y a l t y o f the local p e o p l e b e f o r e it c o m m e m o r a t e s r o a d - b u i l d i n g o r repair, unless this is specially m e n t i o n e d . T h i s explains the clusters o f milestones, constantly added to and s e r v i n g a d o u b l e p u r p o s e , especially at land boundaries. A n o t h e r aspect o f milestones has r e c e i v e d t o o little a t t e n t i o n : the dates w h e n the quarries w e r e in use. T h u s the milestone f r o m Solaise, to the south-east o f L y o n s (AD 43), m a r k s the b e g i n n i n g o f the e x p l o i t a t i o n o f the ' c h o i n ' stone f r o m F a y , a q u a r r y l y i n g close to the R h o n e , s o m e 6 0 k m u p s t r e a m , w h e r e a s , until then, use h a d b e e n m a d e in L y o n s o f Seyssel stone (the material f o r the A u g u s t a n t o m b s at C h o u lans). B u t the fire at L y o n s s h o w e d that this f i n e limestone, v e r y suitable f o r the m o u l d i n g s o n c o l u m n s , w a s unsuitable f o r structural purposes. T h e F a y stone w a s m o r e durable and h a d the a d v a n t a g e o f b e i n g closer to L u g d u n u m . A host o f i n f o r m a t i o n m a y b e f o r t h c o m i n g f r o m a n e w l y d i s c o v e r e d milestone. F o r e x a m p l e , the C o n s t a n t i n e milestone f r o m M o n t Gaussier 9 5 establishes the line o f the r o a d f r o m A i x to the R h o n e t h r o u g h G l a n u m , c o n f i r m s that that c i t y w a s d e s t r o y e d earlier than the f o u r t h c e n t u r y and that it w a s transferred to the site at St R e m y , supports the siting o f T e r i c i a e at M o u r i e s , and finally adds useful d a t i n g to the Peutinger Table. S o m e t i m e s milestones h a v e r e m a i n e d in situ, m a r k i n g s y m b o l i c a l l y the scene o f open-air courts o r used as p r o p e r t y boundaries, h e n c e place n a m e s such as ' T h r e e L o r d s ' stone' o r ' T h r e e B i s h o p s ' stone', and ' T o l l stone', and hence, t o o , the f r e q u e n t c o n n e c t i o n w i t h c o m m u n e boundaries. B e t w e e n C h a v a n n e - l ' E t a n g and M o n t r e v i e u x ( H a u t - R h i n ) a milestone m a r k e d the f r o n t i e r b e t w e e n France and G e r m a n y d u r i n g the last w a r . T h e ' H i g h - S t o n e ' at F o n t a i n e s - s u r - M a r n e stands w h e r e the H a u t e - M a r n e meets the M e u s e d e p a r t m e n t . S o m e milestones h a v e b e e n t u r n e d to C h r i s t i a n use and t o p p e d w i t h a cross ( a l t h o u g h the siting m a y v a r y ) , because t h e y used to attract local respect, a r e m o t e s u r v i v a l o f the cult o f the g o d T e r m i n u s 9 6 . T h e r e w a s , f o r e x a m p l e , the local n a m e f o r the S a c q u e n a y stone, ' T h e B e a u t i f u l S t o n e ' , the m o v i n g o f w h i c h caused great c o n c e r n a m o n g s t the p e o p l e o f the n e i g h b o u r h o o d . O c c a s i o n a l l y there is c o n f u s i o n w i t h m e g a l i t h s ( p r o b a b l y themselves used as b o u n d a r y stones), w h i l e local f o l k - l o r e m a y arise, as in the case o f the ' P e i r o dei N o v i ' , the stone o f the b e t r o t h e d 9 7 : ' T h e p e o p l e o f j o n q u i e r e s , w h o d i d n o t h a v e a n o t a r y in their v i l l a g e , used to g o as far as that b o u n d a r y stone to sign their m a r r i a g e contracts in f r o n t o f the n o t a r y f r o m Redessan [Gard], w h o , f o r his part, h a d n o r i g h t to g o b e y o n d it.' T h e m a j o r i t y o f milestones h a v e c o m e to light, n o t in their original l o c a t i o n , b u t in s o m e reuse, f o r e x a m p l e in churches as c o l u m n s (there are cases w h e r e c o n v e r s e l y c o l u m n s h a v e been e m p l o y e d as milestones) o r as h o l y - w a t e r stoups. T h e y h a v e b e e n f o u n d , again, as thrashing-stones at the v i l l a g e w a s h - p l a c e , as h e a v y rollers, as h o n e s

Evidence from literature and inscriptions

47

for woodcutters' axes, as road ballast, as headers in walls, and often hollowed out to serve as drinking-troughs or sarcophagi (milestones f r o m Paris and Tours) - the converse m a y apply. W h e n a milestone has been moved, it is difficult to be sure of its original siting 98 , since the distances can be given only approximate equivalents in metric terms. N o fractions were given and, while w e find the w o r d i n g m[ilia] p[lus] m[inus], is the higher or the lower figure meant? Were the distances calculated as the c r o w flies or did they take into reckoning ultimate detours? Sometimes there w e r e mistakes in carving (addition or omission o f an X , confusion between V and X or II, figures reversed, carelessness over ligatures). It follows that metric data simply offer a range o f probability. In order to check the calculations that w e make with road itineraries, w e must w o r k f r o m both directions. There m a y be a point at which milestones are crowded or where there is a change in the system o f road measurement and this m a y assist in locating the boundary between one civitas and another. In the same manner today, a change in the road surface often betrays the fact that w e are passing f r o m one administrative district to another.

6 OTHER EPIGRAPHIC

EVIDENCE

THE VICARELLO GOBLETS 9 9

These four silver goblets, n o w at the National M u s e u m of the Thermae in R o m e , were discovered amongst votive offerings at A q u a e Apollinares (Lake Bracciano). T h e y are shaped like pillars with the names o f mansiones and their distances apart arranged on them in vertical lists. Their heights vary f r o m 9-5 to 1 5 - 3 0 1 1 and the technique is uneven. T h e goblets may be patterns f r o m one workshop. T h e y seem to belong to the same period, but the fashion that inspired them probably lasted some time. T w o facts suggest that this information was not a guide to help pilgrims. First, there is no reference to A p o l l o or to the N y m p h s , and second, the route taken to R o m e does not divert (the Flaminia, which does not pass b y Lake Bracciano). Perhaps the owner, f r o m Cadiz, made a gift o f them to A p o l l o the Healer w h e n visiting R o m e . T h e lists m a y reflect the interest that was aroused b y construction or repair of the road f r o m R o m e to Cadiz and appear to correspond to routes described b y the Augustan geographer, Strabo. D i o n , in a lecture at the Collège de France 1 0 0 , has clearly shown h o w the selfinterest and ambitions that took Mediterranean peoples f r o m the earliest times as far as the Atlantic coastline (the Greek Hercules being in rivalry with the Phoenician Melqart) might in a general w a y be summed up in a passage f r o m the Res Gestae o f Augustus (26) : 'I have pacified the w h o l e Atlantic coast f r o m Cadiz to the estuary o f the Elbe.' N a m i n g Cadiz in connection with operations that w e r e conducted on the western and northern limits of the k n o w n w o r l d , says Dion,

48

Evidence from literature and inscriptions

'. . . w a s o n e w a y o f creating the highest p u b l i c respect f o r such o c e a n i c feats. F o r C a d i z w a s , in the eyes o f the A n c i e n t s , o n e o f the places in the w o r l d w h e r e there w a s the clearest revelation n o t o n l y o f the i m p e r i a l mission h a n d e d d o w n b y the g o d s to the R o m a n p e o p l e , b u t also o f that b o n d o f tradition l i n k i n g that mission w i t h the deeds o f A l e x a n d e r . [ T h e latter] s o u g h t t o w r e s t f r o m P h o e n i c i a n d o m i n a t i o n the f a m o u s c o l u m n s that he t h o u g h t his ancestor H e r c u l e s h a d set u p at C a d i z . ' T h e r e is c o n f i r m a t i o n f o r this p o i n t o f v i e w in the milestones that lie a l o n g the r o a d to Spain, the via Augusta, Oceanum1

m a r k i n g the w a y to a r e n o w n e d destination: ad

1.

——-m—— /V ¿li 1 XV S A C . V N r0 X'llli Aorojuv XVI — A^iRvSto tyMAvSo xvi XV HAS TA XClvOVA Ve.fx.No KlMll Ubv vClAf X VI xmi" OXJPPC Tt/IiflvwrHOBANl » INTUlLl «furo S A tlANO XI niSFAU vim X*Vlt SVASAITV CA8ELL10NE >&HlSPALK0K[>r8>,E XII XXXVI» w,.«» ; \ W 0 NE x»il X/ll AFTA.viia ./^/ACOMt XXV covcixe xx Al/ CATVIAC'A ASTICI XV AI.AVMVJ» XVI *vr s>iu> SfCViTf HO.V 'F Xltl AcTihU >CoRDVBAf a c o v> * t A a.« A c o wtAfcAACom XVI VA80NTÍ XVII VArflNGVO Xvi[í x>r f , « ^ La CATVfMCO^CO XVII XII jiv/ii xvi VCItNSE AVli tBon.ocvfvo dm At>MOVOI*S xv CfRVWO/v -xi.il *A»aa Avll VoAiAS, . All JCVÜÍ •f'NN^MA. XV,V At»AC JWVi *vmi X IVWfKIA AOEVISOIARIA Bi,VAf«TH> vt AX Xn itmo L xvi V RvsciMor«/ is« Xxnu O-MPfemiA xjw A D C O A \ X A V . $ T A XXtl M l X*lll A O MART 15 MXX^one /amisi x*».'llf XVH A0'lN4

16r Refmiers, Hojpitalfainti Antoine v. i P.

A Turin,ville capitale de Piedmont.

Riuole v. ch. T u r i n v. ch.

S a i n d Iean de Morienne,cy dejfiu. Sainû lulian Sainft Andre v. ch. ut Bregarre

i P. P.

Bourget,d maingiucht Brafme Sollieres Trcfmignon Lafnebourg l e mont Senys

R.

I P.

i i

p.

?. i r.

i P. «i P. g.

Monte roide.

i r.

La Ramifie

Nojire Dame des Keiges,d gauche. LaChappelledes tranfiz i P. D. AH milieu de U place fur le mont.

La Tauernctte L'hoipital La Ferriere b. La Noualefe v. Suze v. Borfeling Sainô George v. Sainft Ambroife v. Sainû Michel Viglanne v.

8

i f. R. i a

p. i i d P. ii ii i

p.

i r.

P.

g.

P. P. R. P. P.

g.

vn.

ii

ii P. P. R.

Ville capitale, Parlement, (y Archeuefchc, difie T2urinum,acaufedumontTiunis,qin ejt l'idfe de cejt endroit.

Le Daulphinè,&fes dependences. Le Daulphine'cômence a la Guillotiere, faulxbourgi de Lyon,bornant la riuiere du Rofne en amont, le long duquel eft fepare delà Brefle,pres Quineulx:adhere au Mar quifatde Saluces & a la Prouëce,veri Cyfferon, au conte' de Venife a Montelimart, au PiedmSt, a f a i n û Ambrois,a la Sauoye, a Pierre Chaftel,& au pont Beauuoifin. Ce pays eft fitué,partie enmôtaigne, partie en p l a i n e , m a i s ne font les montaignes de Daulphibé tant înfcrtiles.que celles de Sauoye, en tefmoignage de la montaigne de Chalemont,dont viennêt les bons vins du coftede la montaigne deVauloire & Vaul o i r o n j a Vache, Se de N ybe, Se encor de la Vaulpute,qui font endroits fertilei en toutes chofes que Ion feauroit foubhaiter. l.lii.

Extract from La Guide des Chemins de France by Charles Estienne (after J . Bonnerot, La Guide des Chemins de France, 1553, par Charles Estienne, vol. ii, Paris, H. Champion, 1936)

It also takes into account details t o u c h i n g o n politics (local boundaries), economics (fairs and markets, tolls, messengers, relays, fares), t o u r i s m (sights to see, historic places, pilgrimages, landscapes), history (battles, disasters, legends, folk origins

Evidence from literature and inscriptions

59

and settlements, visits by famous people) and archaeology (identification o f sites and description o f ancient remains: defences, tombs, springs, aqueducts, amphitheatres, columns, harbours, inscriptions). T o w n s that are quoted as possessing commanderies, leper hospitals and R o m a n roads include: Aiguebelette in the Alps, Argenton, Auxerre, Bourges, Chartres, Châtillon-sur-Seine, D r e u x , Dun-sur-Auron, La-Ferté-St-Aubin, Houdan, Ingrandes, L e v r o u x , Lorris, Lyons, Metz, M o n t f o r t l ' A m a u r y , Montreuil-Boulogne, Orléans, the P u y - d e - D ô m e , St-Marcel-d'Estréaux, Saintes, T o u l and Tournus. It is important to understand that the Guide was not drawn up by starting with the examination o f a map. As in the case o f Pliny, with his geographical lists, references have been mixed, which explains w h y there are facts transposed or omitted. Confusion has also arisen between place names of different types (towns, districts, rivers). In fact, it was the Guide that provided a basis f o r map-making (cf. infra f o r the example o f Ptolemy). T h e first maps (Oronce Finé, 1536) w e r e local ones; progress was made by assembling these in groups. It was only in 1646 that Boisseau completed the map w o r k for the Guide, the earliest road-map proper having been completed by Tavernier (1632, to show mail services). There is an unending list o f guide books. If one wishes, links can be traced between them and a study made o f their improvement with t i m e 1 3 4 . F r o m them it is possible to glean much valuable information about the old road lists and ancient sites. As an example o f the material available for research into early roads through travel accounts o f more recent date, w e m a y take Montaigne's Journal de voyage1 5 : Travelling conditions, which had at that time hardly changed since the earliest days, were difficult and it is understandable that there are many passages that m a y preserve for us information about roads often o f great value in interpreting the ancient itineraries. General details m a y come to light concerning not only the various alternatives, but also the survival o f the ancient names f o r the roads, the spaciog o f the stations and the operation o f the relay system. F r o m Fossombrone to R o m e (p. 267) the via Flaminia had kept its name. M a n y important connections were made partly b y land and partly b y water (p. 262) ' F r o m Loreto one can g o along the coast, in a week's short stages, as far as Naples. O n e has to go to Pescara and the city o f Chieti, whence a mail-coach leaves every Sunday for Naples.' Montaigne had hoped to reach the latter town. It is a pity he did not manage to carry out his plan. T h e situation in Liguria did not seem to have changed since R o m a n times, or rather it had reverted to a pre-Augustan state o f affairs. W e m a y j u d g e f r o m the following (p. 428) : There were ' t w o roads f r o m Sarrezana, close to Luni, leading to Genoa . . .' (where Montaigne did not go, since 'the roads f r o m Genoa to Milan are infested with thieves') '. . . one takes three days'journeying f r o m Sarrezana and has 40 miles o f dreadful track, very hilly, with boulders, precipices, and foul inns and v e r y f e w travellers. T h e other road passes through Lerici, three miles distant f r o m Sarrezana. There y o u embark and in 1 2 hours y o u are in Genoa.' It is k n o w n that the R o m a n conquest, in Liguria just as much as in Provence, was

6o

Evidence from literature and inscriptions

not very successful with its occupation of the coastal belt, through which ran the road to the province of Spain: 'The whole coastline', wrote the geographer Strabo (iv, 6, 2), 'is dominated by unusually high mountainous escarpments, leaving between them and the sea only a narrow w a y . . . After eighty years of warfare, the Romans managed to ensure with great difficulty that there should be free passage over a width of 12 stadia for travellers on official business.' This text, though dealing with the Provencal coast, holds true for that of Liguria. All this explains the extreme importance, in the maintenance of Roman travel connections, of the ports of Pisa, Luna, Genoa, Marseilles and the early advance in the plain of the river Po, which, strangely enough, ensured the safest communication with Spain, thanks to the easy climb over the Mont-Genevre pass. Even in Montaigne's time the Ligurian pirates were still to be feared! His diary contains indeed many details about the distances of one road station from another (for example, p. 430: Fornovo was two stages from Parma), about road measurements (at Sienna, p. 193, the length of the mile was different), about the organisation of the relays (the hiring of horses, which worked well on the whole, cf. p. 198, and could not have changed since the Roman cursuspublicus) and about the costs of tolls and taxes (p. 430, the Apennine mountain-folk taxed travellers 136 and in northern Italy, tolls were levied at strongholds that lay athwart the roads: Montaigne noticed 'several enclosures that obstruct and close'). This would explain certain peculiar features notices along Roman roads, both in Italy and in France, bends that are not easily explained on flat ground, with the curve being completed on the opposite side by field-boundaries, hedges or lanes, so that the oval thus formed encloses the site of the burgus formerly watching over the road. The course of the highway before its diversion can sometimes be made out as the major axis of the ellipse, if the ground is not built u p o n 1 3 7 . Water tolls were paid during the crossing and not on landing (to avoid arguments !): it was a Roman and an Italian custom. Another ancient tradition can be quoted, where hosts led the way over dangerous stretches (p. 18 5, in the Apennines). This practice was recorded in early times for the Alps 1 3 8 . N o w some remarks about road conditions. The Alps were crossed with ease, in fact everywhere in a very picturesque manner: the climb up to the Mont Cenis pass was made in a sedan chair, the descent on a sledge (p. 445): 'I went up the Mont Cenis pass partly on horseback, partly in a chair carried by four men, with four others acting as relief bearers. I was borne on their shoulders. The ascent lasts two hours, it is stony and awkward for horses unaccustomed to it, but otherwise free from danger and difficulty.' The fact that the Alpine chain could be crossed so easily amazed the Ancients: 'There are many more passes . . . today', wrote Strabo 1 3 9 , 'and, thanks to engineering works, they are as practicable as may be. Indeed, Caesar Augustus completed the wiping out of the bandits by means of road construction' . . . and a little later 1 4 0 : 'One of the passes which allow movement from Italy into transalpine and northern Celtica leads to Lugdunum by the land of the Salassi. It gives two choices of route,

Evidence from literature and inscriptions

61

one being possible for vehicles over the greater part of its course' (the Little St Bernard). On the other hand, Montaigne found that crossing the Apennines was troublesome (except along the Magra road which proved easy) (p. 184): they were 'more difficult mountains than on any other part of this journey'. The course of the autostrada del sole must not make us lose sight of the problems facing the remaining road links, with their steep slopes and with the 'frane' (precipices) a constant obstacle to maintenance. Hannibal is known to have had more trouble crossing the Apennines than the Alps 1 4 1 . Montaigne made careful note of the difficulties of travelling by road, chiefly caused by rivers or marshy country: (p. 182) close to Ferrara: 'it is awkward going for foot-passengers in miry weather . . . as the roads have ditches on both sides, there is no means of walking clear of the mud'; (p. 193) the pot-holes of the Paglia: 'We crossed and recrossed a hundred times over a torrent rushing all the w a y down' the road; (p. 400) at San Chirico 'the beast of burden which was carrying our baggage having fallen into a little stream that w e were fording, all m y clothing and the books in particular were spoilt'. In the eighteenth century, Charles de Brasses was to relate a similar mishap in northern Italy. Finally, a number of observations about crossing rivers: bridges, fords and ferries. There were few bridges; they were kept in repair only at the frontiers of the various states (p. 196, for the Church) in order to maintain prestige. It is quite certain that the majority of the stone bridges that w e find mentioned dated from the Roman period. A number of the routes followed included sections travelled by water, especially in northern Italy, and this had already been the case in the Antonine Itinerary142. (p. 166) at Rovereto: 'They put their travel-chests on these zatte that in Germany are called floats, in order to bring them down to Verona along the said river Adige.' (p. 172) at Casa Fucina: 'an inn where one embarks in order to make one's w a y to Venice'. (p. 174) Horses tow the boat. (p. 176) Battaglia: 'a canal, which, although it is not very deep, is used nevertheless by some strange boats'. (p. 179) On the Adige: 'a floating bridge constructed on two small boats and capable of carrying fifteen to twenty horses, moving along a rope secured across the water 500 paces away, and to hold this rope up, a number of small boats moored between two more, which with the aid of stakes keep the rope out of the water'. (p. 240) 'They get three or four pairs of buffalo to tow their boats up the river Tiber.'143

62

Evidence from literature and inscriptions

(p. 436) 'We crossed the Po on a floating bridge resting on two boats and possessed of a little cabin. It is steered across with the aid of a long rope supported at several points on small vessels moored against each other in the river.' 1 4 (p. 440) The Naviglio: 'the canal is narrow, but so deep that it brings up quite large boats to Milan.' The Ticino was crossed by boat: 'at Chivassa travellers moved across several rivers and streams, sometimes by boat, sometimes on foot'. Fords are mentioned on many occasions. There are several details dealing with the build of the road, often going back to the Roman period. In northern Italy, there is frequent mention of embankments (hence the place name 'Levata') alongside waterways, cf. pp. 172, 177, 178: 'a canal with two roads raised on each side'. This landscape had changed little since ancient times, for drainage and irrigation works had been carried out within the areas marked out for centuriation by those responsible for highways; water-courses have turned out to be the boundaries of subdivided centuries. The old road in such cases had a ditch alongside, often filled up later. The role of cardo maximus was frequently played by a road running parallel with an artificial waterway or a stream that had been straightened. Montaigne noticed the paving of Roman roads: p. 198 (close to Rome): 'Some causewayed roads with very fine paving seemingly ancient'; he observed engineering works: p. 267: the via Flaminia was cut into living rock and possessed firm foundations: 'In that vicinity there is much to be seen of construction in masonry, built up from the bed of the stream, well below the road; immense crags that have been carved and smoothed; and all the way along, traces of their heavy pacing now mostly buried.' p. 420: an example of double superimposed 'bridges', like the Pont-du-Gard. A few more passages may be quoted about the old roads of Italy. 'The road from Ostia to Rome, which is the via Ostiensis, still gives many reminders of its former grandeur, causeways, ruins of aqueducts, almost the whole route strewn with building remains, and more than two-thirds of the said road still paved with that black quarrystone, with which they surfaced their h i g h w a y s . . . (p. 229) Amongst other ruins, we came across about half-way on our right hand the very fine tomb of a Roman praetor, on which the inscription can still be read in its entirety.' In the opposite direction, Montaigne went to Tivoli (p. 243): 'At the Lucan bridge, which still bears its ancient name, there are some old inscriptions, with the chief one very readable. There are also two or three Roman tombs along that road; there are no other traces of antiquities, and very little of the old paving; this is the via Tiburtina.' The 'Ponte Lucano' which crosses the Aniene not far from Bagni di Tivoli, owes its name to Lucanus Plautius. Quite close by stands the circular tomb of the Plautii, dating to the early Empire and bearing the priases of the members of that family. It was these inscriptions that Montaigne saw. Let us follow Montaigne on his travels through central Italy. He noticed (p. 250)

Evidence from literature and inscriptions

63

in the Tiber 'great masses of stone, the remains of the bridge that Augustus had built there to link the territory of the Sabines with that of the Falisci'. W e can identify this with the Ponte Felice, near Borghetto, on the via Flaminia ; it was indeed constructed by Augustus, but rebuilt by Sixtus v (1589, shortly after Montaigne's visit). Concerning the Furio pass, cf. infra p. 105. A bibliography of travellers' tales can be found in the following sources: the Bibliothèque universelle des voyages by Boucher de la Richarderie (1808); d'Ancona's commentary on the Journal de Voyage de Montaigne ; the catalogue of the FossatiBellani collection in the Ambrosiana of Milan; the catalogue under preparation of the Tursi collection of the Marciana in Venice ; the files of the Hertziana Library in Rome and the German Institute in Florence (much surprising information will come to light by consulting cards under headings in the different languages: journal, diary, reisen . . . and different countries, not forgetting anonymous w r i t e r s there were many in the seventeenth and eighteenth centuries - and by consulting original manuscripts). A compilation with a good bibliography has been made by Schudt: Italienreisen im 17. un 18. Jahrhunderten, Vienna, 1959; and there are a couple of good monographs : for Egypt: J . - M . Carré, Voyageurs français en Egypte, Cairo, 1932; for Sicily: H. Tuzet, La Sicile au XVIIIe siècle vue par les voyageurs étrangers, Strasburg, 1955 and Voyageurs français en Sicile au temps du romantisme, Paris, 1946, but this work has only a literary purpose.

OTHER WRITTEN SOURCES

From the eleventh century onward, with the extension of manorial tenure, those lords holding well-sited castles were able to claim the right to exact tolls or to provide guides or escorts. Land schedules may give the list of places where the toll sections began, that is to say the keypoints lying alongside the 'great toll roads' 1 4 5 , which were 10ft wide. Speaking generally, all archives whether religious or otherwise (charters, diplomas, wills, travellers' accounts, reports of military expeditions, monetary statements) may furnish valuable information about topography, especially where there are roads crossing or skirting estates (via romana, stratapublica, via, iter)146. T o these w e should add the results of official enquiries 147 . A trial investigation among the archives of the French government departments dealing with Roads and Bridges, Waterways and Forests, and Public Works has shown what possibilities there are in quantity surveys and specifications arising from civil engineering and maintenance. Already, when Sully was in charge, the Treasury was required to give reports to the King's Council on the overall condition of the roads, together with estimates of the repairs considered necessary. The eighteenth century, a time of great road development, provides many such documents. A

64

Evidence from literature and inscriptions

number of technical notes about the materials needed to metal highways mention how reuse was made of 'paving' from old roads, which were treated as quarries. These activities, which have left clear evidence on the surface, went on to help the early railways pay their way : several Roman roads in Berry were used as sources of ballast. Sometimes the destruction was only partial: one nineteenth-century document dealing with the straightening of an Imperial highway in Provence authorises the sale of half the Roman road, which was more than 8m wide. Only 3 m were retained, leaving the rest to fall into the hands of the owners of adjoining land. All the Roman provinces took part in this type of reuse. In his Lettres familières écrites d'Italie, Charles de Brosses noted: 'The very poorest people in the district have scaled the via Appia as though it were a carp.' Early maps are important, but difficult to use before the seventeenth and eighteenth centuries . W e shall deal below with the contribution to be made to our studies by maps.

2 The archaeology of Roman roads

i CLASSES O F R O A D O u r starting point must be a significant note on the classification of roads m a d e by Siculus Flaccus, a surveyor of the first century AD1 : 'Public highways, constructed at State expense, bear the names of their builders and they are under the charge of administrators (curatores viarum), w h o have the w o r k done by contractors; for some of these roads, the landowners in the area are required, too, f r o m time to time, to pay a fixed sum.' In addition to the viae publicae, usually mentioned in the itineraries, and which, being n a m e d after the magistrates building them, were also called praetoriae or consulares, w e find reference to viae militares2, which were roads possessing strategic importance, at any rate at a particular time, rather than roads specially designed for troops and financed f r o m military funds. These viae w e r e always publicae, but the converse is not the case. Inscriptions tell us w h a t the surveyor wanted us to k n o w about the sources of funds for the viae publicae: contractors were paid b y the curatores viarum with monies f r o m the aerarium, to which were added subsidies f r o m the imperial fiscus and contributions f r o m the inhabitants of each civitas and those living beside the road, with the governor of the province officially checking the execution of the contract. Siculus Flaccus went o n : There are in addition local roads [vicinal, f r o m vicus, a township, the chief t o w n in a pagus] which, after branching off f r o m the main highway, g o off across the country and, often, lead to other public ways. T h e y are built and maintained b y the pagi, that is to say b y the magistri of the pagi, w h o usually see that landowners provide the w o r k force, or rather hand to each landowner the j o b of looking after the stretch of road going over his land. At the ends of such sections, inscriptions m a y be seen of the n a m e of the territory and of the o w n e r in question, together with details of the piece of road that he has to keep serviceable. There is free m o v e m e n t along all these public w a y s . . . Finally there are ways leading across private estates that do not afford passage

65

66

The archaeology of Roman roads to everyone, but only to those w h o need to reach their fields. These w a y s lead o f f local roads. Sometimes, too, they fork o f f f r o m roads belonging jointly to t w o landowners, w h o have come to an agreement to take charge of them at the edges o f their estates and to share their upkeep. Finally, the public highways, the local roads and amongst the private w a y s , those jointly belonging to t w o landowners, all coincide with boundaries. H o w ever, these roads were not laid out to act as boundaries but to open up communications.

T h e local roads (actus) must have made up the greater part o f the network. T h e private roads (privatae, itinera) were linked to the great estates. W e find them mentioned on inscriptions (privatum iter). A sentence f r o m Pliny the Elder confirms that roads acted as boundaries, a function that was long-lasting. 'In Campania, the Leboriae is bounded on each side b y a consular road : one road goes f r o m Pozzuoli to Capua, the other f r o m C u m a e to Capua3.' T h e R o m a n classification m a y be compared with the French system o f m o t o r w a y s , and national, departmental and local roads. These are grouped broadly to serve as main links between important towns, then f o r community purposes, between communes, and finally, f o r everyday use, f r o m hamlets to communes. It will be a useful exercise to note h o w one category passes into another. T h e width o f R o m a n roads varied w i t h their class, although this is not an infallible guide : f r o m the law o f the T w e l v e Tables, fifth century BC4 : 'Military roads must be 8ft (o-295m) wide on straight sections and 16 on bends 5 '; f r o m the so-called Augustan law 6 : ' T h e major axes o f centuriated land, the decumanus and the cardo maximus [cf. b e l o w ] must be respectively 40 and 20ft, the other decumani and cardines 8ft. T h e same writer Hyginus gives elsewhere different figures 7 : 30, 1 5 , or 1 2 f t for the axes and 8 for the other ways. This information m a y be compared w i t h that given b y Pliny the Elder 8 : 'Vineyards must be bounded by a decumanus 1 8 f t w i d e so as to allow t w o w a g g o n s to pass each other. Cardines 10ft wide must be laid out through the middle o f the iugera, or, if the vineyards is extensive, these cardines will have the same width as the decumani.' This freedom o f movement, mentioned b y Siculus Flaccus, will be considered again with regard to centuriation and the regulation Iter Populo debetur. Further classification o f the roads will arise when the methods o f construction are studied. 2

TOWN

STREETS

THE E X A M P L E OF ROME

In his fine w o r k on the Tiber (Le Tibre, fleuve de Rome dans l'Antiquité), Le Gall has emphasised the importance o f the site o f R o m e , which lay at the lowest bridging point of the river, where the via Salaria crossed a road running f r o m Etruria d o w n to Campania. T h e first o f these roads was in fact f o r m e d o f t w o parts : initially, f r o m

The archaeology of Roman roads

67

the salt marshes north of the mouth of the Tiber as far as R o m e , it was the Campana. This was a very ancient w a y , as is shown by the existence of the sacred grove o f the Arvales at the fifth milestone where it emerged f r o m R o m a n territory. Its course was probably improved about 350 BC at the time when Ostia was founded. T h e road crossed the Tiber by the Sulpicius bridge and continued towards Sabine territory as the via Salaria. O n it, the town of Fidenae, which was in the hands of Veii, controlled the salt trade, so vital to the people of the mountains 1 0 . This direct route f r o m the Apennines to the sea is a reminder o f one of the great economic realities of Italy, as of Provence: the phenomenon of transhumance, the role of which in Roman history has been well described by Grenier 1 1 . O n the whole, the R o m a n network o f roads radiated f r o m the centre towards the neighbouring towns, which gave their names to the roads (Ostiensis, Tiburtina) and it was along these that the city grew. T h e gateways in the city walls often have the same name as the roads leading through them: porta Appia, Aurelia, Flaminia, Latina, Nomentana, Ostiensis, Portuensis, Praenestina, Salaria, Tiburtina12. 13 In the outskirts, w e find alongside the viae tombs and cemeteries (the tombs of the Scipios and o f Caecilia Metella on the Appia, and o f Bibulus on the Flaminia), catacombs (that of Domitilla on the Ardeatina), temples (temple of Mars on the Appia), altars and sacred groves. T h e aqueducts crossed over the roads on an ornate series o f arches: the one serving the aqua Virgo over the Flaminia celebrated Claudius' British triumph; the so-called Drusus arch over the Appia was used for the aqua Antoniana. T h e arches of the aqua Claudia et Aniene, which bestride the via Praenestina, later became a monumental gate (porta Maggiore). Within the city, which grew without let or hindrance on hilly terrain, the streets followed courses that were far f r o m straight, very narrow (4-5 to 5m, with 6m the maximum), often on a slope. There are several scalae, some of them shown on the Forma Urbis, like the Scalae Caci leading f r o m the Palatine to the Great Circus or the Scalae Gemoniae, the 'staircase of groans', at the side of the arx, where the bodies of executed criminals were exposed to view. T h e streets were lined by high buildings and the traffic problems are evoked in a colourful passage f r o m Juvenal: One needs to have a lot of money to sleep in this town . . . T h e vehicles moving d o w n the narrow, winding streets, the quarrelsome crowd refusing to m o v e on, all would rob even sea-calves o f sleep. T h e rich man, when called away on business, will have himself borne through the crowd, which opens to make w a y for h i m ; he will make swift progress over everyone's head in his vast Liburnian litter. As he goes, he will read, write, sleep within, for with windows closed, one can sleep the sleep of the just. A n d for all that he will arrive before us. In m y case, the human tide in front o f me prevents me f r o m hurrying: the hastening throng behind me is thrusting into m y back. Someone shoves an elbow into m e ; another man gives me a nasty jolt with a long beam. Here's a fellow also set on giving m y head a whack with his joist and yet another with a mighty cask. M y legs are all over mud. D o w n comes a large boot on m y foot, leaving a military stud stuck in m y toe. Can you see the crush of people and the smoke where the sportula is

68

The archaeology of Roman roads

Porta Flaminia

jstinae_

spy Aedes Aedes Cgstorur ^•Vvicus X.^ O / y T Vestae ^•Vvicus lluuggoa rr ii u úT ? 77A porta r.mCino^o/Zr p o r t.a . . . __] Carmentalis

///

Horti

I \0omus' LJ*urear

Porta fcPraenestina

^Thermae

j^^aPraenestiria Horti Torquatiani

—Porta ' f/r 1 Flumçntana / / q .

6

Minerva Medica..

oT M M a t n i

" "FockJm, Boonun o T Herculis 4 Victons

M. / S i

\ T Divi Claudi! '/Curiae* /Novae

Domus

(TLunoe

^ ThermaeV) )) Xzr Helenianae ¡/ r Palatium Castra SessorianumPriora Equitum (ft Singularium itanus \\ Amphitheatrum Castro Novo Castrense \ Equitum ^orA S i n g u l a r i u m ^

Isium \ Metel I m u m Aro FortunOe" . C a m p u s Martialis

Ludus MatutmuS:

iomus »htlippi A^fcaput p[ T. Herculis f ¡(Africae / Victons \ Basilica iupanaria Hilopana,

iPosterula Templum \ Mercurn \ \ T Solis etedi Lunae Porta Metrovia DECENNIAE

Therrnae\ Decianae ^ ¡ ^ V £ / m îrnalis^^v/P^ta

W

«. / q \ D o m u s £ / / c ? \Cornificioe ePff£ \ Privata q / / â iHadnam

" \

. /

PortoN Naevia 'Porta Aotina

•y Thermae x\ vCommodianae Thermae Severianae 7 Porta ^ Ardeabna/ Via Ardeotinoi

9

Aqua Antonipiana NS-Posterula

.

Porta I Appia

Plan o f R o m e (after Atlas of the Classical World, Elsevier, 1 9 5 9 , p. 145)

Amsterdam-Brussels,

Circus Vonanus

70

The archaeology of Roman roads

being handed out? See, m y tunic's torn and it has only just been mended. A waggon is coming forward with a great baulk of timber swaying about on it; a second is loaded with a pine trunk. These are threatening the crowd as they swing in the air. W h a t if the axle bearing Ligurian marbles happens to break and, losing its balance the whole mass tips over onto the passers-by? H o w m u c h then would be left of the bodies? . . . Just consider n o w the dangers of another kind to which one is exposed at night, the great height f r o m the ground of the roofs perched up in the air, sending d o w n a tile to give you a crack on the head . . . There are also some folk w h o are sent to sleep by the noise of a rousing quarrel . . . A host of other mishaps m a y be feared. There w o n ' t be a shortage of people to strip you, once the houses are closed up . . . T h e odd cut-throat m a y approach unexpectedly, waving his knife around 4 . These famous lines are echoed by another quotation, not so well k n o w n , f r o m Martial: In t o w n . . . it is quite impossible for a poor wretch . . . to get any rest. W e are refused the right to live, in the morning by schoolmasters, at night b y bakers and all day long by tinkers hammering away; over here, the money-changer with no customers is jangling his stock of Neronian denarii up and d o w n on his hollowtopped table, and there, the Spanish gold-beater is striking his w o r n stone with his polished mallet; and there's no rest f r o m the crowd truly possessed by Bellona, first the shipwrecked sailor with his bandaged body and his gossip, then the Jewish child trained by his mother to go begging and the rheumy-eyed hawker with his sulphur m a t c h e s . . . W e are w o k e n by the laughter of the passing crowd, and R o m e is at our very bedside 1 5 . In only t w o instances were the city throughfares merely given the names of via: the Sacra Via, which began close to the Colosseum, then crossed the Velia, the Fornix Fabianus, and the arch of Augustus to be continued as the Clivus Capitolinus. Its course was doubled in width after the temple of Caesar was built. Along it passed the religious processions which used to go up to the Temple of Jupiter on the Capitol. Piganiol thinks it was the decumanus of the Etruscan Roma quadrata. It is impossible to speak of the Sacred W a y without calling to mind the p o m p and circumstance of the R o m a n triumphs which were the means by which the sovereign people in one of the most impressive spectacles of ancient R o m e gave itself a display of its o w n greatness. There, one could see, in addition to war booty, the trophies seized by the enemy f r o m neighbouring nations, captive chiefs and their families and tableaux of mountains, rivers and battles. It was as if the spoil of a whole province was being dragged into captivity 1 6 ; the Nova Via, on the n o r t h summa, to the north-west, Caracalla, lay parallel to the indicated a flat district to the gave its name to the seventh

slope of the Palatine Hill (to the east, it was called infima). Another via Nova, close to the baths of first section of the Appia, while the via Lata, which east of the via Flaminia originally called the via Lata, precinct of Rome.

The archaeology of Roman roads

71

Semita (alta semita, t o d a y 'viale di x x setiembre') stood f o r a m i n o r street, angiportus a lane o r a dead-end street and clivus á street o n a slope, as f o r e x a m p l e : the clivus Argentarius,

l i n k i n g the F o r u m (passing b e t w e e n the T e m p l e o f C o n -

cordia and the A r c h o f Septimius Severus, w i t h a section still visible b e t w e e n the arx and the Forum Julium) w i t h the Field o f M a r s and the the clivus Capitolinus,

Flaminia;

the o n l y street w h i c h w e n t u p to the C a p i t o l in ancient

times, starting f r o m the A r c h o f Tiberius. T h e part still to b e seen dates f r o m the end o f the E m p i r e ; the clivus Palatinus, leading o f f the Sacra Via to the n o r t h o f the A r c h o f T i t u s ; the clivus Scauri, rising f r o m the v a l l e y b e t w e e n the Palatine and C a e l i u s Hills; the clivus Suburanus, c l i m b i n g u p , b e h i n d the Imperial/ora, b e t w e e n the O p p i u s and Cispius Hills in the direction o f the porta Esquilina t h r o u g h the Servian w a l l curcuit (the A r g i l e t u m , the district l y i n g b e t w e e n the R o m a n F o r u m and S u b u r a n e x t b e c o m e s the street e m e r g i n g b e t w e e n the C u r i a and the A e m i l i a Basilica); the clivus Victoriae, o n the n o r t h side o f the Palatine. A city street w a s n o r m a l l y called vicus, such as: the vicus Capitis Africae the vicus Iugarius, part o f a business street leading f r o m the T i b e r to Sabina and, inside R o m e , l i n k i n g the F o r u m ( w i t h an arch at its entrance) to the porta Carmentalis in the Servian walls. W a s this the 'street o f the y o k e - m a k e r s ' ? the vicus Tuscus, also g o i n g f r o m the F o r u m to the T i b e r (Forum Boarium and Circus Maximus), the 'Etruscan street' w h e r e traders and artisans o f Etrurian o r i g i n had b e e n settled since p r e - R e p u b l i c a n times. It h a d a b a d reputation because o f the slave m a r k e t that w a s h e l d there. T h e vici w e r e n a m e d after details o f the local t o p o g r a p h y , buildings, o r statues standing alongside t h e m (vicus longus, Apollinis, delphini), trades (argentarius), gens names o r cognomina (Aurelius, G e m i n i ) . T h e date o f the first p a v i n g in the t o w n w a s as late as 238 BC (clivus Publicius); in 1 7 4 BC its use w a s m a d e m o r e general b y the censors ( w i t h sidewalks o r margines). It w a s skilfully laid, e m p l o y i n g an o v e r l a p p i n g technique. W e h a v e n o t e d incidentally the part p l a y e d b y arches in the t o w n s c a p e . Just as t h e y w e r e erected to m a r k the points at w h i c h m a j o r roads reached the city ( A p p i a , Flaminia, cf. the arches at R i m i n i , Susa, and in G a u l , at O r a n g e and G l a v u m ) , so, inside the t o w n they f o r m e d a m o n u m e n t a l b a c k g r o u n d to a street (the arch o f T i t u s at the top o f the via Sacra) and in the t o w n setting h a d the same role as the trophies that lay at intervals a l o n g the m a i n roads in the o p e n c o u n t r y . A l o n g s i d e the R o m a n streets, in addition to houses, there w e r e temples, fountains, altars (for e x a m p l e , the compitum Acili w h e r e the vicus Cuprius crossed a r o a d r u n n i n g t o w a r d s the Carinae) and statues. W e m a y n o t e here that the ' g o l d e n milestone', erected b y A u g u s t u s in 20 BC in capitefori romani11, sub aede Saturni, consisted o f a m a r b l e c o l u m n w i t h g i l t - b r o n z e plates affixed to it, displaying the distance o f the great t o w n s in the E m p i r e ( w e m a y

72

The archaeology of Roman roads

compare French road measurements taken f r o m N o t r e - D a m e in Paris). This should not be confused w i t h the umbilicus Romae, a neighbouring m o n u m e n t dating f r o m the Later Empire. Imperial streets w e r e wider, especially those laid out after the fire in R o m e at the time o f N e r o . It is w o r t h w h i l e recalling w h a t Tacitus said in the Armales: R o m a n building-land on w h i c h N e r o ' s palace had not encroached, was not, as happened after the fire caused b y the Gauls, used haphazardly for housing: n e w construction had to observe a building line, streets w e r e widened, the height o f houses was reduced, open yards w e r e provided and porticoes erected to protect the fronts o f blocks . . . These measures, w h i c h w e r e popular because they were useful, also enhanced the beauty o f the n e w t o w n . S o m e people thought, h o w e v e r , that the former lay-out o f R o m e had been m u c h better for the health, since the narrowness o f the lanes and the height o f the houses prevented exposure to the burning heat o f the sun, whereas today, those w i d e spaces, offering no shade, are scorched in far m o r e unbearable heat 1 8 . Balland has considered these w o r d s and thinks that they refer to a p r o g r a m m e i n v o l v i n g reconstruction and control rather than n e w d e v e l o p m e n t 1 9 . Nevertheless in one part o f the Nova Urbs (translated f r o m the Greek 'neapolis', meaning ' n e w city') the principles o f classical t o w n planning w e r e introduced: the eastern part o f the Sacra Via, lined b y arcades, as w e l l as the clivus Palatinus. T h e Domus Aurea was entered through enormous porticoes o f monumental proportions. In particular, anxious that his n e w creation should f o l l o w in the Hellenistic tradition, N e r o hoped to expand R o m e towards the sea b y constructing n e w harbour installations. T h e last w o r d s o f the quotation f r o m Tacitus m a y w e l l reflect detailed talks that w o u l d have been based o n the theories put f o r w a r d b y H i p p o d a m o s o f Miletus. In practice, the term Nova Urbs - originally to be the title o f a plan o f m a j o r development before the fire o f 64 - became, through the force o f circumstances, the scheme for reconstructing the districts in the old t o w n . N e r o did not h a v e either the technical or the financial means to compete w i t h the large-scale w o r k s in Alexandria b y carrying out plans that had been b e y o n d Caesar's p o w e r : that is, to transform R o m e into a Neropolis, w h i c h b y f o l l o w i n g the example o f the great classical cities o f the Hellenistic period 2 0 , w o u l d be a reminder o f the glories o f Alexandria. B u t the Emperor was quite w e l l aware that this project created practical and aesthetic problems in the e c o n o m i c field and in t o w n planning. T h e n e w type o f rationally designed domestic architecture was to b e c o m e the rule in the second century. Under the Republic, street construction in R o m e w a s the responsibility o f the censors; resurfacing, paving, cleansing, as the lex Julia municipalis tells us, w e r e the duties o f the aediles (one for each o f the f o u r districts), the duumviri viis in urbe purgandis (under Augustus IV viri viarum curandarum) and the duumviri viis extra propiusve urbem Roman passus mille purgandis, w h o disappeared w h e n A u g u s t u s introduced the curatores in charge o f streets outside the city limits. S o m e v e r y unusual magistratures w e r e also created: a curator viis sternundis (a consul in 92 BC) ,

The archaeology of Roman roads

73

though we do not know if his sphere of action was R o m e or'Italy; a curator viarum e lege Visellia (a tribune in 71 BC) ; and a curator viarum (Fabricius, ten years later). Under the Antonines, inscriptions mention a procurator silicum21, and a procurator regionum urbis adjuncto sibi officio viarum sternendarum urbis partibus duabus2,2. These magistrates had the task of seeing that public property was respected and that there was no movement of heavy waggons from sunrise until two hours before sunset, except in cases of material intended for public works such as building and demolition, a concession that did not apply on holidays (according to the lex Julia municipalis2,3). The ban was renewed later . The way in which private owners encroached on public property created concern that was again manifest in the Late Empire 2 5 . Thus, in Constantinople, on 1 1 October 398, Arcadius and Honorius ordered the city prefect to pull down lean-to buildings that were reducing the width of approach roads 26 ; again, on 1 November 439, Theodosius 11 and Valentinian 111 instructed the prefect to lay charges against those who were including lanes wholly or in part within their property^ 7 . B y contrast with the old towns, cities that were newly built ex nihilo, and in particular the colonia, had the advantage of a genuinely classical lay-out, following the orthogonal model of Hippodamos. In the De Lege Agraria28, Cicero contrasts Rome and its irregular plan, resulting from its uneven site, with a wealthy lowland city and its well-planned grid of spacious avenues. It is known that the foundation ceremony, while well recorded in Roman times, went back to an Etruscan origin 29 . Making use of offsets to astronomical bearings, the augur traced out the axes of the city, the cardo and the decumanus maximus, with the aid of the groma. There is information on this technique from surveyors who were describing how land was centuriated. The next stage was a ritual furrow: the sulcus primigenius marked the line of the rampart. This event is pictured on a relief sculpture from Aquileia and on coins commemorating the foundation of new colonies, in addition to being summarised by several writers. Servius 30 tells us that an Etruscan town could not be considered properly founded (Justa) unless it had three temples (dedicated to Jupiter, Juno and Minerva), three gates and three streets. H o w can this definition be reconciled with an orthogonal plan, which should apparently include four gates, one at each end of both axes? In practice, the circuits of Etruscan towns were adapted to the terrain and none o f these cities seemed to have had a regular plan. A passage from a work by Festus (AD 339), which points out that the seat of the major gods, on the Acropolis, lay to the north, from where they could watch and protect the city, puts us on the track of a solution: since the way to the north was excluded, only three gates and three streets in fact remained so the gods enjoyed a panoramic view of the town. It does appear that the first Etruscan cities were irregular, with the criss-cross plan coming into use only when colonies were founded, closely linked with Greek ones. Etruria was in constant touch with Ionia, the homeland of Hippodamos of Miletus, who codified these theories of city-planning in the first half of the fifth century BC. The dating on both sides seems to coincide. A t Marzabotto, founded at the end of the sixth century BC on a plateau overlooking the Reno, the

74

The archaeology of Roman roads

insulae are b o u n d e d b y streets o f a m p l e dimensions, w i t h w i d t h s o f 1 5 m f o r the m a j o r axes and 5m for the m i n o r . W e m e e t this t o w n pattern again in the n o r m a l l a y - o u t o f R o m a n c a m p s , w h o s e f o u r g a t e w a y s , oriented in R e p u b l i c a n times o n the cardinal points, w e r e l i n k e d b y t w o great streets intersecting at r i g h t angles t w o - t h i r d s o f the w a y a l o n g the m a j o r axis: the 5 o f t - w i d e decumanus maximus ran f r o m the porta praetoria t o the porta decumana, w h i l e the via principalis, 100ft w i d e , e x t e n d e d f r o m the porta principalis dextra to the porta principalis sinistra. W e k n o w this a r r a n g e m e n t f r o m P o l y b i u s . U n d e r the E m p i r e , H y g i n u s , in his Liber de munitionibus castrorum, c o n f i r m e d b y the results o f e x c a v a t i o n s , g i v e s the details o f a r e c t a n g u l a r c a m p , in w h i c h the m a i n axis, the decumanus, 60ft w i d e , is cut b y the via principalis (60ft) and the via quintana (40 or 50).

PROVINCIAL T O W N S

A s e x a m p l e s o f p r o v i n c i a l t o w n s w e shall consider, first, an Italian t o w n o f ancient o r i g i n but regular plan, P o m p e i i , and, second, a t o w n o f c o l o n i a l t y p e in G a u l - A u t u n . P o m p e i i lay on the site o f a t o w n s h i p that w a s native at first, then G r e e k , Etruscan, G r e e k again, then S a m n i t e and finally a R o m a n c o l o n y in 80 BC. O n the w h o l e , its plan is regular and the anomalies can b e e x p l a i n e d b y the existence o f earlier nuclei. T h e m a i n axes are t w o decumani r u n n i n g e a s t - w e s t , the via di Nola and the strada dell' Abbondanza w i t h t w o cardines r u n n i n g n o r t h - s o u t h , the via di Stabia, l i n k i n g the porta Vesuvio w i t h the porta di Stabia, and the street j o i n i n g the porta di Nocera and the porta di Nola. T h e decumanus maximus (strada dell'Abbondanza) is 8-$m w i d e , w i t h p a v e m e n t s o f a b o u t 4 m ; f o r the via di Nola the figures are 8m and 3 - 5 m ; the cardo maximus (via di Stabia) is nearly 7 - 5 m w i d e , the p a v e m e n t s a b o u t 4 m . T h e other streets v a r y f r o m 3 to 5 m . T h e y are all p a v e d . D o o r w a y s o f b o t h houses and shops o p e n e d o n to the p a v e m e n t s (margines, crepidines), o n w h i c h there must h a v e b e e n placed benches and stalls to take a d v a n t a g e o f the d a y l i g h t . A t intervals, l a r g e slabs o f stone a l l o w e d the streets to be crossed after h e a v y M e d i t e r r a n e a n s h o w e r s . O w n e r s w e r e e n j o i n e d to k e e p the r o a d clean in front o f their d w e l l i n g s , u n d e r the w a t c h f u l e y e o f the aediles. W a l l s w e r e c o v e r e d w i t h graffiti, inscriptions and caricatures. T o these w e m u s t add a v a r i e t y o f pictures o f deities and signs o f all kinds. It is possible to c o n j u r e u p , in these q u i e t streets w h e r e t o d a y in s u m m e r the tourist hurries a l o n g u n d e r a b a k i n g sun, the c o l o u r f u l , n o i s y , s w a r m i n g c r o w d s : children w i t h their schoolmasters, acrobats, p r o m i n e n t m e n in sedan chairs or litters w i t h their clients a r o u n d t h e m , soldiers, sailors, actors, passing travellers, slaves f r o m distant p r o v i n c e s , w e l l - d r e s s e d w o m e n o n their w a y to the baths or a s h o w , water-sellers, pedlars, porters, h o r s e m e n , all a d d i n g to the c o n f u s i o n - n o t to m e n t i o n b e g g a r s and thieves. W e s h o u l d n o t f o r g e t the idlers g a z i n g at the t h r o n g f r o m the c o o l interiors o f the barbers' shops and the c u s t o m e r s seated at table in inn (thermopolia) and g a m i n g - h o u s e . It is an accepted fact that in M e d i t e r r a n e a n countries, the m a j o r i t y o f business transactions take place in the

The archaeology of Roman roads

io

75

Paved street in Pompeii

street. W h a t wealth was accumulated in this w a y ! T h e r e was the w h o l e range o f activities f r o m the l u x u r y shops in the strada dell'Abbondanza to the h u m b l e stalls in the side-streets. T h e ruts left b y w a g g o n s in the paving stones g i v e some idea o f h o w m u c h traffic there was. H o w e v e r , s o m e streets w e r e reserved for pedestrians. Elsewhere, the watertanks, at cross-roads, and the street fountains must h a v e been an obstacle to m o v e m e n t . W e should also notice that, as in all R o m a n t o w n s , under the p a v i n g there ran a carefully laid system o f sewers. If w e n o w turn to Gaul, A u g u s t o d u n u m - present-day A u t u n - w a s built b y A u g u s t u s b e t w e e n 15 and 10 BC at a j u n c t i o n o f natural routes b e t w e e n the basins o f the Loire, Seine and R h ô n e , in order to replace the f o r m e r capital o f the A e d u i , Bibracte, in accordance w i t h a consistently held policy. This was to m o v e centres o f population d o w n on to l o w l a n d w h e r e they could b e w a t c h e d , w h i l e receiving at the same time all the benefits o f Romanisation. A u t u n w a s an important meeting-place o f routes: roads led towards D e c i z e , Orléans and B o u r g e s ; towards A v a l l o n , S e m u r , Saulieu and Sens; t o w a r d s Besançon, C h a l o n , M a c o n and L y o n s . W e should r e m e m b e r that the A u t u n m i l e -

The archaeology of Roman roads

76

S'. IVliltfl^M»

i OwMMwrf

Polvanrfi* t«

11

Map of Autun, ancient and modern

The archaeology of Roman roads

77

stone 3 1 , a road sign dating f r o m the beginning of the third century, opened up to travellers vistas of far horizons, with a road to R o m e and a road to the Rhine. T h e Antonine Itinerary placed Autun on the road to the Ocean, leading through A u x e r r e , Sens, Lutetia, Beauvais and Amiens. Study o f this road has shown h o w well its course was adapted to ground features. It was very carefully constructed ; the agger was given flagstone kerbs, while sections have indicated that successive layers of metalling brought its height up to three feet 3 2 . O f the gates in the city walls, t w o have disappeared : that on the south (the R o m e gate, with its foundations recognised in the sixteenth century and which was called in medieval times the gate ' o f marbles', probably because o f its wealth o f ornamental stone) and the one on the west (the St-Andoche gate, excavated in 1850, so that part of its elevation is known). O n the east side, the St-André gate is still visible, through which passed the road f r o m the Loire to Besançon and Langres ; it was defended b y projecting towers. Finally, athwart the road f r o m Sens to the north, can be seen the A r r o u x gate. Both surviving structures comprise double central carriageways with semi-circular arches supporting an upper storey and flanked b y side-passages. T h e surface of the R o m a n streets lies at a general depth o f i-6m. There is a filling of small pebbles in the gaps between the large flagstones. A section was recorded across the street to R o m e w h e n a cutting was dug for the railway in 1866-67. Apparently, after 270, the paving was removed f r o m the summum dorsum, which was then rebanked with building debris and given fresh paving in the reign o f Constantine. T h e streets f o r m a geometrical checker-board pattern such as w e associate with the colonia, but somewhat adapted so that it is oriented to fit into the axes of the city's diamond shape (the main axis runs north-west by south-east). This m a y be recognised as the plan of a camp adjusted to ground conditions. It will be seen that the streets tie in with the towers in the city walls, whose circuit has been established. T h e cardo maximus comes f r o m the direction of the Saône and goes as far as the A r r o u x gate. It is 1,250m long, but it continues f o r another 1 , 5 7 0 m by a series of short straight sections that m a y indicate the edge o f the pomoerium. T h e h i g h w a y was 8m wide and had t w o pavements, each of 4m. There were two decumani at right angles to the cardo, one reaching the west gate after a zig-zag (at the approach to the forum?), while the other, which was to become the road to Besançon, lay parallel, 300m a w a y , aiming f o r the east gate. T h e insulae were determined by the streets and their short sides ràn parallel to the cardo. The areas of the insulae vary and this m a y be explained b y the desire to have different sizes o f land parcel or merely by a wish f o r variety, such as m a y be suggested by another important detail in the lay-out o f the city : the visual effect of stepped building levels in the higher parts of the town. Successive city plans (of which w e have records since the sixteenth century) show that this geometric pattern, more or less abandoned in the Middle Ages, when there w e r e t w o nuclei centred on the castle and the Marchaux, was unconsciously reused, although at a higher level, when n e w development took place in the town.

78

The archaeology oj Roman roads

Similar regular street patterns survive in the majority o f city centres in old French towns : Rouen, Orléans, Tours and Aries.

3 THE R O A D IN THE

LANDSCAPE

A l o n g the roads and outside the towns (beyond the pomoerium) lay cemeteries where monuments, in all their variety, put on display the success o f the dead and their families and allowed the deceased, thought to be pursuing a more leisurely existence in the next w o r l d , a moment's awakening as the living went by. A m o n g many inscriptions expressing this vague belief, at times bitterly and sceptically, w e m a y quote: ' D o not overlook this epigram, O traveller, halt awhile to hear it and k n o w it ere y o u d e p a r t . . . A l l w e . . . dead are n o w nothing more than bones and dust. I have told y o u some h o m e truths, be on y o u r w a y , traveller, that I m a y not seem to y o u over-talkative for a dead m a n 3 3 . ' Propertius, on the other hand, waxes indignant against this social mixing : 'I am in no wise pleased to have m y name on v i e w in the middle of a r o a d 3 4 . ' Examples o f these cemeteries must include, o f course, the Alyscamps, along the road to Italy where it leaves Aries, one o f the best-known sacred sites in the R o m a n West. In spite of damage in modern times and in spite o f the artificial manner in which empty sarcophagi have been placed in line (the finest are in the museum), the site remains as thought-provoking as the via Appia in R o m e . This Christian burial ground, with its series o f 1 7 chapels and its thousands o f graves, g r e w up around the tomb o f St Geneseus. Such cemeteries o f G a l l o - R o m a n and Merovingian times were later to inspire m a n y an epic legend, which saw the numerous tombs as evidence o f mighty battles in the past. Another example m a y be found outside Glanum, where the ancient Celtic road is lined b y Gallo-Greek burials. Here the inscription on the mausolée des Antiques was turned to face travellers coming along the road f r o m Spain and into the t o w n through the municipal arch. T o m b s m a y also be found in the open country, as in the case o f the mausolea or tumuli (in Belgium), belonging to the great estateowners, which made landmarks f o r travellers, even at night (there is an elegiac couplet, taken f r o m Virgil, on a t o m b south o f Lambasa 3 ). Relics o f these holy places, whether they w e r e cemeteries or centres o f worship, have survived in unusual circumstances, often near cross-roads: thus, close to old roads in N o r t h Africa, still lie numerous tombs o f ' h o l y - m e n ' (marabouts). It is this sort o f association that makes these isolated burial places so interesting archaeologically. In addition to tombs, milestones and engineering w o r k s 3 6 f o r m e d part o f the road landscape. W h e n agricultural land was developed, water-course and road sometimes came in for combined treatment. There is often an aqueduct beside a road, as for example f r o m N o v é a n t to Gorze where the underground section o f the aqueduct (the

The archaeology of Roman roads

79

Sarrazin stream) and the old road lie parallel. Such is frequently the case in the R o m a n countryside. T o quote Pliny the Elder: 'Agrippa b r o u g h t the aqua V i r g o a distance of t w o miles along the Praeneste road f r o m the point w h e r e the side-road joins it at the eighth milestone 3 7 .' T o the south of Constantine, the aqueducts, with their characteristic courses, even seem to have been, in certain places, the main feature around which the entire h u m a n development of the landscape was set in motion. Usually aqueducts entered t o w n s at the gateways. Short of an aqueduct, there w e r e watertanks, troughs or wells for travellers to quench their thirst. As Pliny the Elder tells us once more 3 8 : 'Fertile land should be left as m e a d o w and irrigated w i t h rainwater f r o m the public road.' Ancient tracks followed b y livestock (transhumance roads) always have ponds strung out alongside them. W h a t about the vegetation cover around R o m a n roads? W e k n o w that w o o d l a n d was avoided because of ambushes (cf. B o o k i of the Gallic War in w h i c h Caesar describes a long detour m a d e in order to miss the forests and n a r r o w glens of the D o u b s valley). Roads often skirted w o o d l a n d w i t h o u t going into it. B u t alongside m a n y routes there must have been copses - w h e r e a halt could be m a d e away f r o m the sun 3 9 - and also sacred groves. Four relief sculptures at least f r o m Gaul illustrate a horse-drawn vehicle passing a tree 4 0 . Does this merely symbolise the countryside or are w e to visualise plantations beside the road? Pliny the Y o u n g e r states that R o m a n s were n o t u n m o v e d b y the magic of the landscape and his uncle sings the praises of the vines hanging f r o m the y o u n g elms in n o r t h e r n Italy, one of the features of the P o countryside that w e r e to be the delight of all foreign visitors. A n d it is well k n o w n that the Romans also liked to plant w o o d l a n d walks on their country villas. W h i l e tree-planting in antiquity must remain a mystery to us 4 1 , it plays an important part in the appearance of roads: Palladio considers that they should be lined b y trees, while Sully and Gautier g o on to r e c o m m e n d l o w , quick-set hedges, 'so that they w o u l d offer n o advantage to brigands'; they w o u l d give 'the roadside beauty and firmness of line'. Even away f r o m the centres of population, there w e r e sanctuaries along the roads, w h e r e people paid their respects to Hercules 4 2 , M e r c u r y , Mars, the Dioscuri 4 3 , the Lares Viales (invoked by Plautus), and the divinities attributed to the meeting-point of t w o , three, or four roads (Biviae, Triviae, Quadruviae). These little places of worship were often sited at p r o m i n e n t features on the j o u r n e y , such as cross-roads and fords, and m i g h t conceal earlier cults (in Gaul, for example, there are columns to the horseman-god close to fords, while Caesar in the Gallic War quotes the protection given to travellers b y Mercury). A passage f r o m Apuleius 4 4 tells us h o w varied these holy spots w e r e : ' W h e n pious travellers meet a sacred g r o v e o n their w a y or some holy place, they usually offer a prayer and an ex-voto, halting awhile T h e n a list follows: ' A n altar garlanded w i t h flowers, a g r o t t o adorned w i t h shady foliage, an oak tree bearing horns, a beech bedecked w i t h skins, a holy m o u n d enclosed b y a wall, a tree trunk sculpted b y an adze into a h u m a n shape, a patch of turf charged w i t h the aroma f r o m the b u r n i n g of libations, a rock sweet-smelling from perfume.'

8o

The archaeology of Roman

roads

4 STREET PATTERNS A N D C O U N T R Y ROADS. CENTURIATION BOUNDARIES The previous notes on city streets necessarily form part of our subject, since there is a close connection, at any rate in theory, and sometimes verified on colonial land, between the patterns of city streets and of those serving farmland. This is made clear by a text in Hyginus concerning Ammaedara (Haidra, in Tunisia) 45 . At this point we should redefine centuriation, which as w e know comprised a grid traced out on the ground. After he had marked on the two axes at right angles to each other, the decumanus maximus and the cardo maximus, the Roman surveyor (agrimensor, gromaticus) laid out in relation to these a series of parallels, which appear on the ground as lanes flanked by side-ditches or lines of stones cleared from the fields. The land divisions thus obtained were called centuriae (originally one centuria for 100 men) and were further subdivided in a geometric fashion by footpaths leading to the various lots, by drainage ditches or irrigation trenches and by lines of stones. The centuriae occurred in several dimensions, the commonest being composed of 20 actus each of 120ft, about 710m. The ideal arrangement according to the surveying manuals, was that the pertica or centuriated land as a whole should be divided into four equal areas starting from the very centre of the city; but this was theoretical only and scarcely applied except on the colonia. W e know of one example at least: Ammaedara (Haidra) in Tunisia. It is worth noting here that the orientation of

12

Part of the cadastral inscription from Orange (after A. Piganiol, 16th Supplement to Gallia, 1962, p. 109)

The archaeology of Roman Roads

81

street grids, land-plots and main roads in relation to each other m a y date the successive stages b y w h i c h colonisation t o o k place. T h u s , in northern Italy, centuriation w h i c h is tied in w i t h the via Aemilia o f 187 BC must be c o n t e m p o r a r y w i t h it or later. O n e o f the most extraordinary inscriptions to h a v e survived is the cadastral registry o f O r a n g e , w h i c h outlined o n flagstones the boundaries o f the centuriae; a f e w important landscape details; exact i n f o r m a t i o n about the siting o f the lots in relation to the axes; the areas o f tributary o r colonial land; the land t a x ; and finally, the n a m e o f the official adjudicator. T h i s material evidence, placed alongside the illustrations in the textbooks o f the gromatici, proves that there must h a v e been land registries in general use, i n v o l v i n g some fairly exact m a p - m a k i n g (and o f the latter, the Peutinger Table gives o n l y a sketchy idea). Let us take as an e x a m p l e the centuria DDXVI-CKI f r o m land registry B : T h i s f r a g m e n t w i l l fit exactly o n to the m o d e r n map. T h e R o m a n engraver has m a r k e d the river Berre and the h i g h w a y (the A g r i p p a road), w h i c h bridges it at the L o g i s de Berre. T h e n suddenly he has b r o u g h t the road to a halt at the precise point w h e r e the edge o f the plateau des Echirouzes rises f r o m 73 m to 124. T h e old road w e n t up o n t o the h i g h g r o u n d in the direction o f M o n t e l i m a r , for, 8km south o f that t o w n , the discovery o f an inscribed milestone 4 6 i n f o r m s us that Tiberius repaired that road in 31-32. S o o u r engraver left o f f m a r k i n g the course o f the road w h e r e the straight stretch ended. A t the L o g i s de B e r r e a milestone bears the n a m e o f Constantine . . . T h e A g r i p p a road is n o w the D - 1 5 8 and at the L o g i s de B e r r e the m o d e r n road avoids g o i n g u p o n to the plateau des Echirouzes, but turns to the left towards D o n z e r e , j u s t w h e r e the ancient draughtsmen finished his line 4 7 . O n e o f the most surprising things arising f r o m the examination o f maps and aerial photographs f o r traces o f centuriation is the w a y in w h i c h limites h a v e survived. O n e is struck b y short sections o f road, w h i c h do not fit in w i t h the present-day n e t w o r k , and turn sharply or end suddenly for n o clear reason. B u t these stretches o f road, once projected, are f o u n d to be in alignment. T h e y m a y be continued b y a footpath, ditch or stream or b y some local b o u n d a r y , sometimes simply b y the edge o f cultivated land: these sudden changes s h o w h o w far the old courses still persist. Such alterations in alignment to right o r left, h o w e v e r slight, are evidence o f the g r o w t h and d e v e l o p m e n t o f f a r m i n g b u t there is a l w a y s a return to the original axes, striking p r o o f h o w straight these w e r e . Place names like 'dismano' (from decumano), 'cardito' ( f r o m cardo) and 'limite' (from limes) - in Italy - p r o v i d e extra evidence. W e must emphasise that these land registries c o v e r e d large areas o f the R o m a n w o r l d : almost the w h o l e o f Italy and Spain, N o r t h A f r i c a (as far as agricultural land w a s concerned), Y u g o s l a v i a and the D a n u b e valley, and P r o v e n c e . T h e r e are traces in Greece, the N e a r East, B e l g i u m , G e r m a n y and even in Britain. T h e r e can be n o d o u b t that on the g r o u n d the parcels w e r e v e r y clearly demarcated. T h e g e o m e t r i c c h e c k e r - w o r k pattern f o u n d in the provinces resulted f r o m the lines

82

The archaeology oj Roman roads

set out when the land was developed, with the centuriae inscribed as it w e r e on the surface; the limites are best revealed b y the network o f communications. O n e main road m a y serve as a decumanus maximus for one or several land surveys. M a n y other roads w e r e built to fit in with those constructed f o r centuriation; as far as possible their courses w e r e made to coincide with the limites of the centurial parcels. W h e n it was impossible for the road to maintain a straight line because o f unfavourable terrain, a detour was made to reach the next limes so as to avoid expropriating land or interrupting the system o f centuriae. T h e roads were indeed the key to this economic set-up, with their posting inns and granaries. It was not only on the main highways that road and countryside w e r e in harmony with each other. T h e geometric lay-out o f the secondary roads, which w e r e also based on the cadastral survey, was the best method o f ensuring that all land parcels had the same ease of access; exactly the same system was used when the American West was colonised. T h e road n e t w o r k was the means by which land was distributed and divided into lots and so f o r m e d the main feature whereby the estates w e r e developed. Thanks to their stony metalling, often running alongside waterways, the old roads have survived, helped too by the fact that the tenancies o f the properties through which they ran (the regulation Iterpopulo debetur, the subject o f Saumagne's excellent study) did not allow trespass. It is f o r this reason alone that, accidentally, centuriation has persisted in the V e g a de Carmona. It can be easily seen that tracking the details o f the land registries which were probably in their day the best w a y o f carrying out colonisation allows us to obtain a realistic estimate o f the progress o f both conquest and settlement, showing h o w thoroughly and permanently R o m a n isation took hold. T h e discovery o f the land divisions will provide the only proper f r a m e w o r k in which to plot random archaeological finds and, together with the distribution o f the villae and vici, lay the basis o f a statistical survey o f both the d e m o graphy and the economy, so enabling us to understand what R o m a n civilisation was at bottom: a f r a m e w o r k for local initiative, which filled the gaps. T h e w h o l e structure combining adherence to principle and flexibility in application gave stability to the land, which was the foundation o f the economy o f the ancient world. It is striking to observe h o w the guiding lines o f the past, which were usually dictated b y the very nature o f things, still make their presence felt in the modern landscape. Modern developments take place at points geometrically determined b y age-old coordinates. Roads and railways f o l l o w directions that avoid cutting across farmland, w h e r e the field boundaries still correspond to the ancient land divisions. T o w n s are still g r o w i n g where the old axes crossed. Large farms lie close to R o m a n ruins, in the angles of centuriae. In this w a y , the land surveys, not very visibly, f o r m the f r a m e w o r k to living in many provinces. T h e unchanging landscape is one of the most certain facts in historical geography.

5 ROAD

CONSTRUCTION48

It happens that one o f the f e w ancient texts giving information on the different phases of construction o f a R o m a n road is a poem by Statius in praise o f the via

The archaeology of Roman roads

83

Domitiana, a short cut along the Appia49. D o m i t i a n had it built at great cost between Sinuessa and Pozzuoli, skirting the sandy, marshy coastline so as to avoid the r o u n d a b o u t route t h r o u g h Capua. T h e text reads: T h e first task here is to trace furrows, ripping u p the maze of paths, and then excavate a deep trench in the g r o u n d . T h e second comprises refilling the trench w i t h other material to m a k e a foundation for the road build-up. T h e g r o u n d must n o t give w a y n o r must bedrock or base be at all unreliable w h e n the paving stones are trodden. N e x t the road metalling is held in place on b o t h sides by kerbing and n u m e r o u s wedges. H o w n u m e r o u s the squads w o r k i n g together! S o m e are cutting d o w n w o o d l a n d and clearing the higher g r o u n d , others are using tools to smooth outcrops of rock and plane great beams. There are those binding stones and consolidating the material w i t h b u r n t lime and volcanic tufa. O t h e r s again are w o r k i n g hard to dry up hollows that keep filling w i t h water or are diverting the smaller streams. These lines are very m u c h to the point, in spite of the abrupt, rhetorical style affected b y Statius, and Duval's useful c o m m e n t a r y 5 0 is followed here. First it is clear that the strip of land involved in the road was m a r k e d o u t by furrows, so obliterating earlier tracks: there has been a view that the w o r d limites could, technically speaking, apply to the byroads giving access to centuriation in C a m pagna. T h e road trench had to reach, if n o t bedrock, at any rate a sufficiently firm foundation, which was reinforced b y r a m m i n g , piles or b r u s h w o o d (crates). Sections w e r e built at a time, as w i t h m o d e r n m o t o r w a y construction; this m e t h o d of building m a y well explain the slight changes of alignment observable o n the g r o u n d in the case of R o m a n roads. T h e trench was packed w i t h materials b r o u g h t f r o m elsewhere, the b o t t o m layers being m a d e of stones, gravel and sand successively and the spine of the road cambered to aid drainage. T h e surface was given a covering of paving stones held in place by a kerb (umbo) f o r m e d of similar stones set o n edge supported on the outer side b y a stony revetment rather than a sidewalk. O t h e r wedge-shaped stones (gomphi)51 acted as clamps: they w e r e j a m m e d between the stones of the kerb as well as in the paving 5 2 . M a n y different trades w e r e engaged in the w o r k : woodcutters, quarrymen, carpenters, stonemasons and navvies. T h e surrounding hills provided flagstones, paving stones or rubble and in particular timber for reinforcing the foundations, for use in constructions of all kinds (new fords, bridges, suspended roadways) or again for heating lime kilns. Employing m o r t a r (here made of l i m e 5 3 and pozzolana instead of sand) was unusual on roads, except on very difficult g r o u n d . T h e approaches to the road must have been kept well drained, b u t Statius does n o t m e n t i o n side ditches. H e next deals w i t h the bridge t h r o w n across the Volturno. T w o other technical sources must be noted dealing w i t h road construction. T h e first was written b y Julian and occurs in a letter to his friend and tutor Libanius just as the e m p e r o r was setting off on a campaign against the Persians 5 4 : In the territory of Chalcis, I came across a road running t h r o u g h the remains of

84

The archaeology of Roman roads

an Antiochan winter camp. This highway was, so to speak, half-marsh, halfmountain, and hard going withal. The marshy ground was crossed by a layer o f stones, seemingly cast there by intent, but in an artless fashion, unlike other territories, where, in the case o f roads, as for masonry, stones are rammed together into the built-up soil o f a causeway, as if it were mortar in a wall. Julian, keen strategist that he was, used to march at the head o f his troops, keeping a watchful eye on the roads. A solidly built construction o f stones set in soil as described here is not often found and was kept for difficult sections, swampy areas, deep glens and high ground exposed to snowfall; nevertheless it is a feature providing proof of a Roman origin. In modern times, the system o f construction has depended on the use o f graded stones, well packed and tamped. The last quotation comes from the Gothic War by Procopius and describes the Appia between Rome and Capua, which was paved by Appius with stone from various sources: 'After much laborious smoothing, the slabs were cut into polygonal shapes and he then laid them together without using lime [some manuscripts say 'without metal'] or anything else. And they were fitted together with such care and the gaps so well filled that, to the onlooker, they appear to be the work not o f man, but o f nature 55 .'

6 ROAD BUILDERS W e should remember at this point the text quoted earlier from Siculus Flaccus about the classification o f roads, their construction and their upkeep. In the case o f the main roads, the importance o f the army's role has already been emphasised, with its cadres of civil or military engineers, who were often veteran volunteers (evocati). These men did not rely on merely empirical methods, a fact amply demonstrated by the straightness o f the boundaries o f centuriation over dozens o f miles, despite obstacles and interruptions imposed by relief. It is not easy to visualise how the surveying was carried out, although manuals do contain ideas on triangulation. Sighting instruments had already been devised {groma, dioptra) which allowed signals to be picked up from one point to the next and overcame problems created by differences in level. This would be worth further investigation, starting with what is known from Greek science. It would be wise, too, to spot what has survived beyond the Middle Ages. Fustier's useful book 5 6 , La Route, shows clearly the part played by surveying instruments in the development o f road engineering 57 . Roman engineers certainly aimed at high points in the landscape, for from one viewpoint to the next it is possible to check the bearings o f straight sections o f road 58 . T o carry out the work there must have been a work-force o f some size, provided by the army and helped, of course, by conscripted civilians, as Cicero's Pro Fonteio tells us 59 : like Bugeaud's army in Algeria, the legion built the roads, established the land survey and registry, thus helping in the settlement o f the natives and their

The archaeology of Roman roads

85

Romanisation. W e have very f e w ideas about this large-scale w o r k . Detachments were put at the disposal of the specialists responsible for undertaking the land survey and at the same time laying out roads: thus the milestones in southern Tunisia bear the name of the proconsul and the number of the legion that took part on the road survey: Legio tertia Aug[usta] leimitavit, C. Vibio Marso proco[n]s[ule|60. W e also k n o w of a land conveyance effected under Septimus Severus south of the Hodna Chott: the transaction, prepared by the freedman Epagathus (probably a Greek mensor) and the prefect's adjutant, Manilius Caecilianus, was negotiated by an evocatus in the Third Legion, M . Gennius F e l i x 6 1 . Another inscription worth quoting dates f r o m a d 145 and was found on a rockface in the gorges of Tighanimine (Aures). It commemorates the w o r k of a unit brought f r o m Syria to face a native rebellion: 'Under the reign of the emperor Caesar T . Aelius Hadrianus Antoninus, Augustus, pious, father of his country, consul for the fourth time, and of Marcus Aurelius, Caesar, consul for the second time, by order o f Prastina Messalinus, imperial legate and propraetor, the vexillatio of the V I Legion Ferrata built this road.' It is noticeable that the majority of main roads were pioneered by military operations. For example, on its return from the first Samnite w a r (343-40), the R o m a n army did not come back along the via Latina, but followed the coast through the territory of the Aurunci, thus blazing the trail of the Appia on a line that had already been k n o w n to traders, at least since the hegemony of Etruria. In the early third cencentury, operations against the Umbrians o f Mevania and Narnia and against the Senones took into account the route that became the Flaminia. Great strategic roads were built by the military in Gaul under Agrippa f r o m bc 16-13 i n Dalmatia and Pannonia under Tiberius f r o m AD 6 - 9 6 2 , in the Rhineland and the Danube valley under Claudius, and in Asia Minor under the Flavians. These highways were policed f r o m fortified posts, which soon developed into army colonies, both in Italy and in the provinces. Examples include the R o m a n colony of Terracina (329-26), guarding the most dangerous pass on the Appia, the Lautulae; the Latin colonies of Cales (337-34) and Fregellae (328-25) which protected the via Latina against the raids made by Samnites and other enemies f r o m the south (Pyrrhus took this road as far as Anagni and perhaps Praeneste); the Latin colonies of Alba and Carsioli on the Apennine w a y ; and the colonies of Dyrrachium, Pella and Philippi on the via Egnatia. These posts had, when first set up, undeniable strategic importance, but soon became in addition centres of population and development. In the countryside between the large towns, there came into being a pattern of townships and markets based on economic factors, recalling once again Bugeaud's policies in Algeria hence the existence of those fora created by roads, with names taken f r o m the great colonisers: Forum Appi on the Appia, where Appius Claudius Caecus had been bold enough to erect a statue of himself wearing a royal crown 6 3 . In peacetime, the decision to open up a road lay with the censors 64 , eventually with the praetors. Under the Empire the heads o f state took a personal interest in the road system and commemorated new road construction by striking special

86

The archaeology of Roman roads

coin issues : thus w e have the Augustan denarii o f 16 BC variously displaying a milestone in celebration o f repair w o r k or t w o triumphal arches with equestrian statues and trophies on a bridge or a double arch with rostra and a quadriga on a bridge, or again a double arch or a fivefold one bearing a chariot drawn by elephants. This shows the importance attached to roads b y the government. A s early as the eulogy o f Appius Claudius the building o f the Appia was put on the same plane as military victories or political deeds.

7 S T R U C T U R E A N D F E A T U R E S OF R O M A N

ROADS

T h e possibility o f digging test trenches m a y not often occur and results are not clearcut: it is seldom that datable material comes to light f r o m the build-up of the road or f r o m the side-ditches. In any case, n o construction can be regarded as typical. B u t there are many ground exposures in quarries and public w o r k s which allow sections o f old roads to be studied, once they have been scraped clean, while it is simple and cheap to carry out geophysical profiles o f roads by electrical or magnetic methods, such as at Izernore (p. 107). In the early seventeenth century, in his Histoire des grands chemins de l'Empire romain, written at the royal c o m m a n d 6 5 , N . Bergier elaborated a theory that has been popular f o r a long time, arising f r o m the laying o f floor surfaces and pavements (expolitiones) and the building of tiled flat roofs. Bergier believed that he had been able to establish that the f o l l o w i n g layers w e r e laid in succession on beaten earth up to a thickness o f 1 or i $m. statumen, several rows of flat stones, bound with mortar or clay; rudus or ruderatio, a watertight foundation raft, made of a concrete containing small pebbles, broken stone and brick, well rammed d o w n ; nucleus, a finer concrete, with a lime and sand cement reinforced with broken tile, rolled out in layers ; finally, summum dorsum, with a central camber, sometimes sloping straight d o w n on each side, for drainage, made o f gravel (as for viaeglarea stratae) or paving stones (viae silice stratae). Bergier's theory, which was to have w i d e acceptance b y archaeologists, seems to have originated as follows : whereas the w o r d pavimentum had only one meaning for Vitruvius and Pliny the Elder, namely the actual floor inside houses, Bergier connected the w o r d with the French 'pavé', which signified the w h o l e o f a road, and so followed a false argument. Since pavimentum = 'pave , then all the information given by Vitruvius in this respect must apply to R o m a n roads ; the author's research led him to believe he had identified the various layers making up the pavimentum. There is a certain similarity between the build-up of roads and that o f paved floors, but a R o m a n road is not a 'buried wall'. Its structure is both more variable and more complicated. As a general rule, what is seen is : a foundation course (sometimes involving stakes), consisting o f a rubble layer

The archaeology of Roman roads

87

with stones set on edge to aid drainage, always the concern o f R o m a n engineers; a build-up of resilient material 66 (sand); a surfacing, not necessarily paved, especially in the earlier period (the first paving appeared in 1 7 4 / B C 6 7 , but on the Appia not until the time of Hadrian) and in the open country 6 8 , usually along the easier stretches.

13 R o m a n road f r o m R h e i m s to T r i e r : plan and section at Florenville. Details: (a) section: I recent fill 2 undisturbed natural 3 clay 4 sand (b) plan s h o w i n g surface metalling (after J . Mertens, 'Les routes romaines de la Belgique', Industrie, no. 10, O c t o b e r 1955, p. 3 1 )

Nevertheless the manner in which roads w e r e built varied a great deal, even along the same route, according to the firmness o f the subsoil and the available materials. T h e interpretation o f most sections is complicated b y the repairs and remetalling that occurred on many occasions. Sometimes iron slag was used (cf. place names like ' v y Ferrat'). W e should remember, too, Ulpian's distinction 6 9 between, on the one hand, viae terrenae, mere surface roads, and, on the other hand, glarea stratae and silice stratae, properly metalled roads. Broise has recently published some detailed observations made in the HauteSavoie: T h e road is made o f successive layers o f gravel, often reaching one metre in thickness, the result in m a n y cases o f a number o f resurfacings. People have spoken in the past o f consolidated gravel and w e have often noticed ourselves that the gravel o f roads in Boutae (Roman Annecy) had a binding o f compacted clay o f thin mortar so as to make up what w e w o u l d call today a rough concrete. Surfacings o f large stones, whether or not roughly shaped, have also been noticed over stone foundations; but the fact that there is no metalling today on the surface does not mean that there was none originally, f o r paving, especially in the vici, m a y have been torn up for later building-work.

88

The archaeology of Roman roads

This point will be noted later with reference to African roads. W h e n they passed through population centres, some roads included sidewalks (margines, crepidines) with a raised edge (umbo). There are many raised causeways bordered b y lines o f shaped slabs set upright and possessing side-ditches. A m o n g s t the ditches, several types can be picked out with the aid o f aerial photographs 7 0 : quarry-ditches immediately alongside, which provided ballast and material for remetalling and drained the surface-water; the verges are then either side of the causeway or else 20m out, with a width o f i - i to 1 7 m , sometimes more. There are such ditches even within towns. T h e y are not unique to R o m a n roads. W e m a y observe that, thanks to the retention o f ground moisture, even when these ditches are filled, they allow vanished roads to be seen f r o m the air when the causeway has been destroyed (a practice which had official backing when major roadwork took place, as in the eighteenth century'. These ditches sometimes have legends attached to them, as at C a g n y (Somme), where the bands of dew running along them are said to mark the w a y trodden by St Domitius and St Ulpha. boundary-ditches lying 20m more or less on each side; their presence m a y explain the shape o f elongated fields along the roads which correspond to strips o f land that long remained uncultivated (they m a y well have a different colour). There m a y have been a ban on ploughing or on building (we k n o w this f r o m inscriptions on the edges o f aqueducts) 7 1 or a verge m a y have left f o r road-users (a bridleway for horsemen, or a grassy belt f o r herds, as in Spain) or again the natives m a y have been forbidden to use a strip o f land, at any rate in the early days and in woodland favourable to ambush. There is mention in Strabo o f such measures o f security along the Ligurian coast 72 and in m o r e recent times w e m a y note the practice o f cutting d o w n vegetation on' either side o f a road within musket-range 7 3 ; finally, h o l l o w - w a y s exist 7 4 , lined by side-ditches lying a javelin's throw a w a y which appear to be connected with former military camps and m a y have served to allow the safe movement o f troops or the entry o f water supplies, fodder and livestock. W e have the example o f the great depression linking the t w o R o m a n camps stationed in front o f Gergovia. Raised banks, bundles o f b r u s h w o o d or palisades may have completed these bracchia-type ditches, which must thereby have acted as an obstacle to the movement o f surface water. Roads varied considerably in width. Augustus had issued decrees (known through surveyors' writings) fixing the width of a decumanus maximus as 40ft ( 1 3 m ) , that o f a cardo maximus as 20 and o f other decumani and ordines as 12. B y r o a d s w e r e allowed 8ft. If the widths that w e see today corresponded to the original widths, then w e might have a chronological guide. B u t the standards used varied with the importance o f the road, and this changed with the times, as is shown b y the differences in the width and thickness o f the deposits visible in one and the same section.

The archaeology of Roman roads

89

In towns there is a connection with the openings through gateways (thus at Autun, 50ft, that is 8m, with t w o sidewalks of 4ft; an order of 1720 speaks of 60ft). Usually in the countryside roads are i 5 - 2 o f t wide to allow vehicles to pass each other but w e have to reckon with cases of narrowing where the terrain was difficult! It is unwise, apart f r o m the main strategic roads, to attempt any classification of the old highways. Ruts, both in town (as at Pompeii) and in country, may have been the result of wear and tear, but where they are deep (6-30cm) and with sharp edges, clearly shaped by pick and hammer, they were obviously intentional. O n a w k w a r d ground, close to a sheer drop, they served as rails, guiding the waggons, for these had front wheels higher than the actual frames, making independent steering impossible. Bends had to be negotiated therefore on a wide arc. T h e use of these rails, sometimes in conjunction with transverse grooves to prevent horses f r o m slipping 7 5 , may have started earlier than the Romans. In some places, parking lanes are said to have been recognised, which raises the problem that this would involve one-way traffic 76 . Gauges have been claimed varying f r o m 1-05 to 1-85m; the figures differ according to the w a y in which the measurements were made (from the centre line of each rut or between inside edges). Comparisons have been sought with the wheel-bases o f ancient vehicles, prehistoric (waggon burials) or Roman, and with modern railway gauges. It is quite possible that careful study of the standards used could yield evidence about dating (i"3m in antiquity, 1 -45m in the Middle Ages ?). Alongside the paved w a y in use by heavy waggons, there often ran parallel tracks providing easier going for both walkers and riders, especially when faced with oncoming traffic. Palladio has pointed out these sand or gravel strips, which were separated f r o m the higher traffic lane in the centre b y lines o f stones set on edge. W e need only to travel in Spain, even today, to notice in the verges o f many important roads the tracks followed by sheep in seasonal movements. O n each side of the Appia, close to Itri, Fustier has observed a line of prismatic boundary stones, like those edging the road. T h e y are spaced out irregularly and mark out a beaten track 2-^ft wide for beasts of burden, so extending the band in public use to 32ft 7 7 . A strip of clay soil running alongside a paved road may also be evidence of an older road of mere beaten earth. W e should take special note of the profiles of R o m a n roads, both longitudinal and transverse. T h e builders avoided obvious problems and there was a preference for a course on dry level ground or along the so-called military crest, below the top of the slope, where one could see without being seen. Nevertheless difficult situations were tackled purposefully 7 8 . One such difficulty occurred on marshy ground, where the road was given a proper causeway, and not just an earthen ridge 7 9 , upon dykes which later on in the Middle Ages may have been used to hold back ponds (as in the Sologne). Professor Mertens has recovered some interesting information about the via Mansuerisca in the Hautes-Fagnes, B e l g i u m : he found a wooden framework,

The archaeology oj Roman roads

90

probably in the native tradition 80 , involving piles and culverts, overlain by a builtup causeway: T w o rows of cross-beams were placed on the marshy soil for the whole length of the road. The distance between the beams was 2m. There was a gap between the internal ends of each pair of beams, while the external ends projected some 40cm beyond the basic timber seating. These outer ends were slotted to take vertical stakes so pinning the framework to the ground. On the cross-beams were fixed lengthwise with the road two lines of joists to carry the sides of the highway. The joists bore a transverse 'corduroy' of tree trunks and on this in turn there lay limestone flags cemented with clay, covered again by the road-metalling of gravel and pebbles. Tree trunks were discovered under other roads at rivercrossings 81 . Other examples can be pointed out. In the marshlands of the Ems and the Hunse, wooden roads have been found with morticed timbers and at Kembs, near the old course of the Rhine, traces of duckboarded tracks. The 'Ferronne highway' in the valley of the Therain, lying in the marshy country west of Clermont, had a foundation of stones laid chevronwise. Modderman has noticed a layer of timbers probably belonging to a road running along the bank of the Rhine 82 . Drouyn observed at a point at which a Roman road crossed swampy ground layers of tree trunks laid criss-cross fashion 83 . Caesar noted such construction close to Sens 84 and reference can also be made to the passage in Tacitus' Annates85 describing the Pontes Longi at the mouths of the Rhine: 'Caecina received the order to cross at all

14

Section of a road built o v e r on unstable g r o u n d : the via Mansuerisca in the Hautes-Fagnes (after J . Mertens, op. cit., p. 39)

The archaeology of Roman roads

91

speed b y w a y o f the L o n g Bridges. T h i s is a n a r r o w h i g h w a y in vast marshes, built in the past b y L. D o m i t i u s ; b e y o n d it there is o n l y slime and thick, tenacious m u d , in w h i c h the w a t e r w a y s are difficult to m a k e o u t ; all around rise l o w w o o d e d hills . . . Caecina w a s w o n d e r i n g h o w he c o u l d at one and the same time repair the t i m e - w o r n bridges and drive a w a y the e n e m y . ' Place names like ' P o n t - l o n g ' or ' L o n g - p o n t ' m a y w e l l take us back to similar structures. W e should also note the timber roads ( b r u s h w o o d bridges) preserved in the peat o f the La B r è c h e marshes west o f Breuil-le-Sac (Oise) 8 6 ; they w e r e m a d e o f timber frames resting on a brushw o o d foundation held in place b y stakes, fixture being m a d e b y w e d g e s not m o r tices, the w h o l e covered b y a double floor o f beams supporting w a t t l e w o r k and then b y sand. T w o stretches are k n o w n each o f 600m, w i t h a w i d t h o f 3 9 to 18m and a thickness o f 60cm. T h e course w a s f o r k e d , the eastern approach being paved. T h e supporting layer o f b r u s h w o o d was strengthened in s o m e places b y stout tree trunks and, close to the river, b y quarry stones cemented w i t h marl. T h e a m o u n t o f material used per hectare has been estimated at 2,000 cubic metres o f timber and 4,000 cubic metres o f b r u s h w o o d . C o m p a r i s o n s h a v e also been suggested w i t h a timber track in the peat b o g s o f Somerset (Meare Heath), dated to the local Late B r o n z e A g e (750-500 BC) and w i t h the 'road o f life' in the G i r o n d e marshes, but the age o f the causeways at Breuil-le-Sec is m u c h in dispute and the proffered dating ranges f r o m a prehistoric origin to the M i d d l e A g e s . T h e interest o f such structures is that they can be carbon-dated. Engineering w o r k s proper w i l l be discussed later. It is useful at this point to compare descriptions o f various roads, as they appear b o t h o n the surface and in section, f r o m different provinces o f the Empire. T h e Syrian road f r o m A p a m e a to T h e l e d a provides one e x a m p l e 8 7 . It is 6 m w i d e w i t h a central spine and kerbs at the side or raised edges. In l o w e r parts liable to flooding, it becomes a causeway, interrupted at intervals b y a setting o f large flat stones. T h e edges o f the causeway slope g e n t l y d o w n to the g r o u n d o n either side. T h e stone setting, laid b e t w e e n the middle rib and the kerbs, f o r m s a hard road surface o f fitted paving. Crossing the marshland, the w a y is m a r k e d b y a layer o f flagstones, level w i t h the surface o f the water. A n o t h e r road w o r t h describing is that f r o m Bosra (Syria) t h r o u g h Damascus to P a l m y r a 8 8 . It is 5 - 5 m w i d e and is c o m p o s e d o f t w o strips, each sloping a w a y f r o m the centre to f o r m a slight h o g ' s back w i t h a line o f stones m a r k i n g the m i d d l e rib. T w o further such lines m a k e up kerbs on each side. T h e s e three parallel features are constructed w i t h r o u g h l y h e w n slabs, rectangular rather than square in shape and set e d g e to edge. B e t w e e n them there is a road surface o f basalt slabs laid in an irregular pattern. T h e three lines o f stones project about 15 or 20cm a b o v e this surface. W h e n a w a d i had to b e crossed, an e m b a n k m e n t w a s built to ease the gradient. T h e e m b a n k e d sections are o f the same w i d t h as the road and their sides are a l w a y s sheer. T h e y are constructed in dry-stone masonry. T h e external faces consist o f r o u g h l y dressed blocks w i t h side measurements b e t w e e n 30 and 50cm. C a r a v a n roads in the desert, b e y o n d the limes, h a v e left little g r o u n d evidence, f o r their courses w e r e defined o n l y w h e n crossing stony c o u n t r y . In order to spare

92

The archaeology of Roman roads

the feet of the animals in a camel train, small rocks and sharp fragments were pushed to each side to form parallel verges, which have survived as gently curving ridges 89 . But the style of building at the posting-stations; the method of establishing watchtowers which could exchange visible signals; the provision of water supplies (by well or tank); and finally the actual distances of the stages - all these factors show that the Romans were responsible. On the whole, the roads follow the obvious course over the desert. When the going was not firm under foot, a higher route was sought, half-way up mountain spurs. Mostly, w e simply find viae terrenae, humble beaten tracks with edges each side, and with eventual later improvements. Where floods were a threat, there occur at regular distances cross-ribs of flat stones laid athwart the normal longitudinal features, enclosing in effect a series of macadamised strips. The paving slabs used on the low ground are flat above, but on their lower sides bear a bevelled point. There are, according to the author, two stone edges 1 2 m apart which serve either to contain the surfacing material or simply to mark out the course of the road, at any rate in the interior of the province. Sometimes there are four lines of stones dividing the highway into three lanes, the middle one being 6 5 m wide. In general the impression is given of an organisation whose methods were firm, but not harsh, based on a sound knowledge of local geography (there were winter and summer roads) and carried out with flexibility. The map of Roman roads is the map of natural routes. They run straight with bends and detours when the ground requires. One is struck by the regularity of the stages and the wideness of the highway. There is no hesitation in tackling difficult terrain (swamps, lava-flows) at the cost of major works, but as soon as possible the road seeks out the easiest course, the most picturesque places and the watering-points. It must be added, that as regards the materials used in the actual road construction, there is still a shortage of careful, on-the-spot observations backed by sampletaking for petrological analysis 90 . The study of sands by grading in particular should allow quarries to be identified, since special qualities of material may have to be obtained at some distance. In a f e w exceptional cases, details of interest have come to light, as, for example, the use of Euganean trachyte, which was transported by water to pave some roads in the lowland around Padua, in the same w a y that basalt from the Vivarais was employed in Gaul. It may well be that the systematic study of grades of material will enable some advances to be made in dating. Model sections have been published by Mertens in Belgium 9 1 or in Italy (Alba, via Valeria) as well as by Fustier 92 . They bring out the variety of solutions adopted by Roman road-builders when having to adapt themselves to individual problems (in the Crau, a natural conglomerate was used, while elsewhere w e find a mass of clay well rammed down, 'a technique as yet unrecorded for counteracting the absence of hard material by working up local substitutes'). In Alsace there is a road made of crushed Gallo-Roman titles. Even in Italy, Fustier comes to the conclusion that there is no general theory. The course of a road is as much determined by geographical factors as are the

The archaeology of Roman roads

93

materials to build it. There is also evidence for much reuse in the building and repair of roads 93 . The following is an excellent description of a section: The old road is built in a wide depression dug into the ground and filled with fine sand. The width of the roadway between the kerb-stones set on edge is 5 '2m and the sandy sides of the depression each form margines (parallel tracks) of i-5m, which may have served as bridle-paths. The foundation course is made of small slabs of local limestone, bedded on a scatter of other stones resting on the sand, so as to form a perfectly flat surface which shows no signs of wear and tear. The next layer, which rises to the level of the kerb-stones or just above, is slightly cambered and less even and has traces of remetalling. A rather loose paving lies at this level between the western margo and the edge of the side-ditch running parallel to it and some 1 5 m away. A n as of Constantine 11 was found just above the second layer. A t the higher levels, the side-path is no longer in service and the fifteenth-century highway takes on that cobbly appearance so often associated with farmyards. The later resurfacings have been dispersed by ploughing. The kerb-stones measure 28 by 12 by 4cm 94 . In conclusion, what must be emphasised is the great variety of approach to road construction in the past and the adaptability of road builders in planning a route and in using local methods, which is understandable, since it was the natives living alongside the roads who often built them and anyhow kept them under repair. W e should remember, too, that on the ground the same road may pursue several courses in service at the same time or successively, as a result of changes in land use. Some branch roads lost importance, other roads were upgraded or downgraded for economic or political reasons and as a result or urban development.

8 ENGINEERING

WORKS

BRIDGES

In the first place, w e shall deal with the type of bridge that came to be built in many places, thanks to Roman skill in the construction of arches. Opus caementicium was used for the foundation and the framework of arches and piers, together with opus quadratum or latericium for the facing. The method of building (similar to that for aqueducts), and the sites are known in a variety of ways: textual references, especially technical works like those by Vitruvius; commemorative coin issues; relief sculpture (notably Trajan's Column); mosaic floors (the Aries bridge); illustrations in old prints; archaeological finds (including the results of dredging, which may bring up oak piles, protected by shoes, flanged iron tips); and finally place names provided by maps and land surveys. Examination of a plan and analysis of the site on the ground may suggest the

94

The archaeology of Roman

15

roads

A mosaic from Ostia: the Aries bridge

possibility of a bridge (for example, a vanished bridge over the Tiber along the line of a street in Rome and again the destroyed bridge over the Rhone at the point where the projection of the main axis of the Aries amphitheatre meets the river). Building operations were carried out in two ways: in Mediterranean lands, the piers could be erected directly on the river bottom, when the bed was dry in summer. In North-West Europe, where there are no seasonal differences, piers were constructed with the aid of piles and timber scaffolding. In a chapter devoted to the problem of masonry construction in water 95 , Vitruvius gives some details and we learn of the employment of artisan divers 96 . Roman engineers could not build flattened arches of great width, so, in order to avoid extra spans and piers, they raised the apex of the central arch well above flood level. This arrangement required steeply sloping approach ramps, which carried traffic across the river on a series of arches allowing free passage to boats and flood waters. The river banks sloped the opposite way to the ramps and the side arches were successively lower and narrower. The road over the bridge had therefore a pronounced humpback. Building details are of interest: pier-abutments serving as buttresses and cut-waters, very high vaulting (the rise corresponding to at least one-third of the opening), constructed from large blocks with offset joints or using parallel arch beams normally without ties, tympana linked by curved jointing, and, in the case of the small arches, tall slender voussoirs. Sometimes there can be seen holes for bronze clamps that have vanished, corbels that once supported timber centering and gaps for scaffolding. Attention may be given, too, to the sectional construction of the roadway, the type of parapet, and the decoration (cornices, lintel-courses, pilasters and capitals, inscriptions).

The archaeology of Roman roads

95

Such surviving evidence forms a useful comparison with classical texts dealing with bridge-building, in particular Caesar's famous description o f the bridging o f the Rhine 9 7 : H e built a bridge in the f o l l o w i n g w a y : he coupled together pairs o f timbers, one and a half feet thick, sharpened a little w a y up f r o m the end and measured according to the depth o f the river, so that they stood t w o feet apart. W i t h the aid o f rafts he lowered them onto the river bed and set them firm b y ramming, not like ordinary piles in an upright position, but leaning over with the flow of the current. Opposite to these and forty feet downstream, he set t w o further timbers, coupled in the same w a y , leaning against the current. These t w o pairs o f timbers had t w o - f o o t wide transoms set into them and across the g a p ; between them, one each side, braces w e r e fixed to keep the couples apart. W i t h the piles thus both splayed and clamped together in opposite w a y s , the w h o l e construction had such strength, thanks to natural laws, that the greater was the thrust o f the current, the m o r e the structure held together. U p o n the transoms w e r e placed cross-pieces and upon these again battens and wattlework. Furthermore, downstream, sloping piles were fixed to reinforce the whole structure like a buttress and take the force o f the river. Others were arranged a little w a y upstream f r o m the bridge, so that if the barbarians sent d o w n tree trunks or boats to destroy the construction, the shock w o u l d be lessened.

T h e bridges o f R o m e W e shall n o w sum up what is k n o w n about the bridges o f R o m e , following Le Gall's masterly treatise on the T i b e r : T h e Sublicius B r i d g e (from sublica, pile or timber) was probably erected b y the Etruscans, although it has been attributed to Ancus Martius 9 8 . It is famous f o r having been defended b y Horatius C o d e s against P o r s e n n a " . O v e r it the R o m a n s fled to C a e r e 1 0 0 . It was certainly an ancient structure, as evidenced b y the superstition attached to it and the archaic style o f building, which was respected to the point of tabu: no metal, beams with f i x e d ends (using w o o d e n braces?), perhaps to allow easy dismantling. T o V a r r o , of course, w e o w e the suggested e t y m o l o g y o f pontifex: pontufices . . . a ponte arbitror: nam ab his Sublicius est factus primum ut restitutus saepe101, and some have gone so far as to see this function as a survival f r o m the lake-village culture, with the bridge having originated as a track borne on piles. B u t these explanations remain uncertain, despite medieval parallels (Order o f the Pontiff Brothers, confirmed b y Clement hi in 1 1 8 9 , the Brotherhood o f the H o l y Ghost). T h e Sulpicius Bridge, often carried a w a y b y floods and restored, at any rate until the fourth century, is k n o w n through a coinage issue o f Antoninus Pius (140-43), showing Horatius C o d e s swimming in the Tiber, and through an issue o f Marcus Aurelius (180). Its position, downstream f r o m the Aemilius B r i d g e , has been established f r o m an analysis o f traffic conditions (city walls, ease o f access).

96

The archaeology oj Roman roads

The Milvius Bridge ('Ponte Molle') may date from the fourth century: after the capture of Veii, a w a y across was needed, in the first place made of wood. Then, having served the via Veientana, the bridge was used by the Clodia (late fourth century), the Cassia (second half of the third century) and the Flaminia (220). It is clear that local geography determined the exact site. There must have been an approach causeway over the swampy ground 1 0 2 . In 207 BC the people of Rome made their w a y along it to meet the messengers from the Metaurus. The bridge seems to have been rebuilt in stone by the censor for 109, M . Aemilius Scaurus, with each end of wood, thus being easier to destroy. In the Res Gestae, Augustus takes pride in having repaired all the bridges along the Flaminia, with the two exceptions of the Milvius and the Minucius. On that occasion, statues were erected to him upon triumphal arches at each end of the road, at Rimini and at the Milvius Bridge 1 0 3 . The bridge, after many repairs, still carries modern traffic. The Aemilius bridge or pons Lepidi, of which one arch survives ('Ponte Rotto' was its name in 1598) downstream from the Isola Tiberina, was the first stone bridge in Rome. It owes its name to the censor for 179, M . Aemilius Lepidus, who built the Aemilia Basilica and who was responsible for the bridge-piers 104 , carrying a wooden roadway, and destroyed by a storm. The arches were the work of P. Scipio Africanus and L. Mummius, censors in 142. W e know of restoration at the time of Augustus (an entrance arch) and medieval repairs by the popes, who left a commemorative inscription. The remaining fragment is composed of travertine and peperino stone with vaulting of tufa from the Grotta Oscura. There are indications of corbels. A sombre note on Roman life: it was a popular spot from which to commit suicide 105 , while Heliogabalus' corpse was thrown from this bridge into the Tiber 1 0 6 . The Fabricius bridge, built in 62 BC, links the left bank with the Isola Tiberina. An inscription informs us more than once: L. Fabricius G. f . cur[ator] viar[um] faciundum coeravit, and, on the central arch: idemqueprobavit. Another inscription: M. Lollius M. f . Q(uintus) Lepidus M(anii) f . cos. ex. s(enatus) c(onsulto) probaverunt, corresponds to a repair after floods in 21 BC. This was also a suicide spot 1 0 7 . The bridge, $-8m wide, was decorated with herms. It is built of travertine in those parts exposed to view, but the rest of the vaulting is made of tufa; the facing of travertine and brick is of later date. The Cestius Bridge must date from the end of the Republic (first century BC) , since building initiated by emperors bears their name. A fourth century restoration turned it into the pons Gratiani. The bridge of Agrippa or of Aurelius Antoninus (restored) corresponds to the present 'Ponte Sisto' (see below, pons Valentiniani). Nero's bridge (its medieval name, the piles are known) led to the circus from the Vatican. Hadrian's bridge (Aelius) is well preserved. It is dated to 13 3-34 by the inscription: Imp. Caesar Divi Traiani Parthici Jilius, Divi Nervae nepos Traianus Hadrianus

The archaeology oj Roman

16

Roads

97

T h e Fabricius b r i d g e in R o m e

Augustus pontifex maximus tribunic(ia) pot(estate) XVIII, cos.Ill fecit. T h e structure comprises: three central flattened arches, t w o small ones on the right bank and three on the left serving to carry the approach roads; piers w i t h triangular cutwaters; and in the bridge structure, iron ties covered w i t h lead. T h e r o a d w a y , 4"75m w i d e , is paved w i t h p o l y g o n a l slabs o f lava-stone and has t w o sidewalks

98

The archaeology oj Roman roads

3'ira wide made of rectangular travertine slabs cramped together. Pilasters today carry Bernini's angels, but were preceded by columns bearing monumental statues: these turned the bridge into a ceremonial way leading to the mausoleum of Hadrian. There is a bronze medallion showing the bridge with three large central arches and two steep approach roads, each carried on two arches, balustrades made of stone slabs ending in little pilasters, and, on the right bank, a river wall and step. The access roads have been altered, as the town has changed. In medieval times the bridge was in heavy use, for it linked the city with the Vatican. The pons Valentiniani underlies the 'Ponte Sisto', it was the ancient pons Agrippae (Agrippa built it and adorned it with statues), which was restored by Antoninus Pius, then by Valentinian and Valens. A whole arch has been found lying in the river and fragments are known of the architectural and epigraphic detail, coming from a triumphal arch erected by Antoninus at the entrance to the bridge. The inscription announces: T o the emperor Caesar, our lord Flavius Valens, very great, pious, fortunate, victorious and triumphant, always Augustus, the Senate and the Roman people, for the foresight which has always been common to him and his illustrious brother. The honour of dedicating Valentinian's bridge, set up according to the needs of the Eternal City, after its completion has been handed over by decision of the very great princes to L. Aurelius Avianus Symmachus, a most worthy man, former prefect of the town. The latter must have directed the work during his prefecture, which was completed on 10 March 365 1 0 8 . Pieces of gilt-bronze statues have been found (first to third century). The pons Gratiani, dating from 370, was demolished only in 1885-89. The orator Symmachus spoke about it in his eulogy of Gratian 1 0 9 and, addressing the Rhine, he declared: 'Well, O horn-bearer 1 1 0 , take care not to think thyself the equal of Tiberinus, since you both sustain monuments of Princes; he has been crowned and thou art under the yoke.' This bridge, made of reused materials, involved one of the largest known arches (length 48m, opening 23-65m), with a system of iron cramps set in grooves and sealed with lead, earning it the name of'ponte Ferrato'. Concerning the pons Theodosi, known through medieval sources and near the A ventine Hill, according to the order followed by the list in which it is mentioned, we simply know that artisan d i v e r s 1 1 1 were employed and the construction had given rise to jobbery. As for the pons Probi, there was a restoration of Nero's bridge. There must have existed bridges of which we have no knowledge (in particular, wooden bridges over the Tiber). Nero had thought of the scheme which involved a bridge between Palatium and the Capitol 1 1 2 .

The archaeology oj Roman roads

99

Roman bridges in Italy A m o n g the most unusual were : Augustus' bridge at Rimini (AD 1 4 - 2 1 ) which carried the Flaminia on five l o w arches and five high ones (the central span measuring 3 2m) and was embellished with triumphal arches ; the Savignano bridge over the Rubicon, which survived until the recent w a r ; the Narni bridge over the Nera (AD 10) ; the Ascoli Piceno b r i d g e 1 1 3 over the via Salaria; Domitian's bridge, close to the mouth of the Volturno, described by Statius; the ponte Leproso, at Benevento, on the Appia ; the Fossombrone bridge (third century AD) ; the very fine bridges in northern Italy (Ivrée, Pré-St-Martin, Val de Cogne, Aosta, St-Vincent, Châtillon) ; the bridge of Porto Torres in Sardinia, a humpback structure with seven arches. Roman bridges in Gaul There is no complete inventory for G a u l 1 1 4 , although the evidence exists for native Gallic bridges - through place names 1 1 5 and through Caesar's text - at Lutetia 1 1 6 , atMelun, on the Aisne, at G e n a b u m 1 1 7 , at Geneva, on the A l l i e r 1 1 8 , andatPonts-deC é 1 1 9 . W e may note by w a y of example, the following bridges : at Ascain over the Nivelle (Pyrénées-Atlantiques); Flavian, close to St-Chamas; Julian, near A p t ; at Vaison over the Ouvèze ; at Viviers over the Escoutay ; at Champagne (CharenteMaritime) ; at Saintes (decorated with the Germanicus arch, n o w removed), destroyed under Louis Philippe; the bridge of boats at Aries with arches and milestones, k n o w n through a mosaic in the Place des Corporations d'Ostie; at Narbonne, used as a foundation for houses; and, of course, the bridge-aqueduct at Gard. Other remains are found at A m b r o i x on the Vidourle, at Domqueur, Ganagobie, Lavoute, Mane, Mondouilloux, Outremecourt, Pont-Serme, Rolampont, StThibéry on the Hérault, Ste-Colombe, Sommières, Tarascon, Vendelais, and Le Vignau. According to Blanc, there appear to be the ruins of 15 bridges in the ravines of the Ardèche. The R o m a n bridge at Geneva, which replaced its Gallic predecessor (cut by Caesar in $8 BC) 120 a little upstream on the axis of the decumanus, has been described by Blondel. There is knowledge of a masonry pier and the piles o f another pier in the layer of clay, serving as a foundation for a timber framework which must have supported the masonry (square oak posts and wrought-iron shoes with square nails). The structure, which dates f r o m the end of the second century and was twice restored in antiquity, rested on both banks, formerly set back, and on the island in the Rhône. The wooden roadway, 5-5-6111 wide, was carried between the piers on such stout trestles that they lasted until the sixteenth century. T h e 19 spans o f i2-8m gave a length of 220m, with a bend in the course at the island. T h e crossing was guarded by a fortlet on the left bank and a tower on the island.

IOO

The archaeology of Roman roads Other bridges in the R o m a n world

In Belgium, there are possible examples at Arlon and Montignies-St-Christophe. In Germany, the army established a number of bridges for military operations, in Caesar's train 1 2 1 . The crossing at Pfalzem-Stadtbredimus, in use since prehistoric times, has yielded the piles of a pre-Roman bridge, dated by dendrochronology to 1 6 8 - 1 4 9 BC. This small bridge was followed by a Roman one in 30 AD similar to that at Coblenz. A lead medallion illustrates the Rhine bridge at Mainz-Kastel. But the bridges that have been most carefully studied are those at Trier (by Cuppers, 1969). Their siting has been explained with reference to local geology and surface drainage, the road network and the development of R o m a n technique, not to mention other crossings over the Rhine and Moselle and their tributaries and bridges in the Danubian lands, Switzerland, France and Britain. At the point at which the valley is widest and both banks are favourable to human settlement, a massive bridge was built about 44 AD, with ashlar piers resting on a system of piles. As the river level is changeable, construction had to be on a large scale where there was a junction of vital roads. This first bridge was mentioned by Tacitus in his account of the Batavian w a r 1 2 2 (70 AD : the Moselle bridge linking the colony with the other bank). Remains of the bridge were found during w o r k on the bed of the Moselle and they conformed with what w e k n o w of the roads and constructional details. T h e piles were made of oak trunks 50cm in diameter, their tips protected by iron shoes. Seven piers were found resting on these piles, set in rows to f o r m a pentagonal base measuring overall io-2m by 19-5m. Flat planking had been laid over a clay floor well rammed down with broken stones into the gaps between the piles. T h e whole foundation supported the ashlar blocks which stood to a height of 8m and bore a wooden superstructure.

The archaeology of Roman roads rj If

u u rr ir \r a -a m a 0u n a a a $ i n c c: il 0 O 0ar> t* 0n a a 0 0 o J•> •O 3 a if *u • at n

o c> 0 • 9 ,• ®n t? m \iO rj ( , r* •J

IOI

p o s a i , cruci o

d* IP g> t. CO t^ -on o a n J r> a j o a n o o o t, n u ci

ty .» o a n t ' A a n « .J .

«

" '

»'ft T

17

3"

/j

as *J ;j a a 'Jo 0 a a ft : ^ n !-; a a • u n r, c ? o C • a 3 n o n 'i r 11

£

VoJuos

¿if

-J

» a a a o n n D ' ,t

.¿¿V' c'a o ^

-

ù 0

a

a o 3 Oa .

.

R é m y ) , E r n a g i n u m (St-Gabriel), T a r u s c o (Tarascon), U g e r n u m (Beaucaire), N e m a u s u s (Nimes). 65

Historia Naturalis, m, 20, 136.

66

In the Late E m p i r e , the D r ô m e valley was preferred. C f . the distribution pattern o f coinage f r o m Marseilles in the C e l t o - L i g u r i a n regions, H. R O L L A N D in R. Et. Lig., x v , 1949, pp. 139-48.

67

68

S A L L U S T , Histories, 11, 96, Civil War, i, 13, 109: P o m p e y trail blazed b y Hannibal, but one, t o w a r d s the sources o f the (as the P o was k n o w n ) .

69

L I V Y , v , 34, 8.

70

Gallic War, 1, 10, 4.

71 72 73 74 75

76 77

origin, cf. H O R A C E , Epodes, 4, 14: a m a n w e a r i n g o u t the A p p i a n W a y w i t h his Gaulish ponies. 3

Pack animals, able to use the poorest o f tracks, f o u n d e m p l o y m e n t t h r o u g h o u t protohistory (tin route). T h e supply train o f the R o m a n a r m y included e v e n m o r e beasts o f burden than waggons.

4

T h e hipposandal is a veterinary fitting intended to protect a sore f o o t that is unshod. In L'Attelage et le cheval de selle (Paris, 1931), C o m m a n d a n t LEFEBVRE DES NOETTES claimed that R o m a n horse-trappings comprised simply a saddlecloth and a bridle. M o r e precise detail can be obtained f r o m the panels o f w e a p o n s on the arch at O r a n g e , cf. R. A M Y , P . - M . D U V A L . . ., L ' A r c d ' O r a n g e , Paris, 1962, p. 87 and pp. 138 sq. It is the use o f t w o spurs that appears to be later than the C a r o l i n g i a n period.

5

A n d the same r o o t as veredus, a Gallic steed. Paraveredus, o f course, a supply-horse, g a v e palefroi. T h e w o r d 'cheval' is also p r o b a b l y Gallic. A l l these terms are discussed in the Dictionnaire des Antiquités.

4 ; cf. A P P I A N , did not f o l l o w the opened up a n e w R h ô n e and Eridan

iv, i , 2 - 4 . For this point see also R . D I O N , op. cit. S T R A B O , iv, 6, 7. iv, 6, 11. C f . Actes du Colloque international sur les cols des Alpes, Antiquité et Moyen Age, B o u r g , 1969, Orléans, 1 9 7 1 . Paris, 1969. T A C I T U S refers to this passage in Histories,

iv,

6

Georgics, in, 536; Aeneid, x i , 138.

7

C f . J U V E N A L , Satires, in, 254. Annuaire du Collège de France, i960, p. 379.

8 9

77C f . H. Jürgen E G G E R S , Der römische Import im freien Germanien, 1951. 79 Gallic War, iv, 17. 80 Annais, x m , 53. 81 Cf. S U E T O N I U S , x x x i . 8 2 C f . T A C I T U S , Annals, x i , 20. 83 Annals, x m , 53 and Histories, v, 19; cf. S U E T O N I U S , Claudius, I. T h i s study has been resumed b y R . D I O N in Annuaire du College de France, 1964, p. 425. 8 4 For the W e s t e r n Alps, cf. pp. 165 sq. 8 5 C f . T A C I T U S , Annals, 1, 20, 1. 8 6 N o t e that their nationality changed w i t h t i m e : Syrians tended to replace Italians. 8 7 C f . J. N . V O N S A D O W S K I , Die Handelsstrassen der Griechen und Römer durch das Flussgebiet der Oder, Weichsel, der Dniepr und Niemen an die Gestade des Baltischen Meeres, A m s t e r d a m , 1963. 88 Historia Naturalis, vi, 208. 8 9 Ibid., vi, 173. 9 0 Ibid., x x x v i i , 45. 9 1 Ibid., vi, 181 and 184. 92 Ibid., vi, 40. 93 Epitoma rei militaris, Iii, 6. 94 11, 5, 34-

78

Chapter 4

10

In Gaul, skill w i t h skill w i t h w o o d .

11

P. S A L A M A , -Considérations sur les transports routiers romains, en particulier dans la p r o v i n c e d ' A f r i q u e ' , Actes du 79e Congrès nat. Soc. sav., A l g i e r s , 1957, p. 291.

C f . T A C I T U S , Annals, m, 9.

2

Their

reputation

varying

according

their

accompanied

11, 2.

13

x , 38. vin, 5, 8, 17, 28, 30, 47. T h e heaviest loads must h a v e been d r a g g e d o n sledges or carried b y w a t e r : V i t r u v i u s m e n t i o n s w a g g o n s f o r m o v i n g large c o l u m n s or architraves. B u t in the quarries, special g r o o v e d tracks w e r e needed. In other sources, ships appear to h a v e been used f o r obelisks (AMMIANUS M A R C E L L I N U S , x v n , 4, 14). C f . Gallia, 1963, p. 332: relief f r o m J a n d u m , s h o w i n g horse and carriage. For examples f r o m L u x e m b u r g , cf. C h . - M . T E R N E S , Index des travaux concernant l'archéologie . . ., L u x e m b u r g , !9 PP- 264-267. MIRKOVIC (M.), 'The Roman Road Naissus - Scupi and the stages Ad Fines', Z.Ant., X , i960, pp. 249-257MUELLER (R.), 'Die Geographie der Peutingerschen Tafel in der Rheinprovinz, in Holland und Belgien', Geogr. Anzeiger, 1926, Heft 9 - 1 0 (and Gotha, 1966). TOMOWSKI (T.), 'Beitrag zu einer Rekonstruktion der Strecke Scupi-Stobi auf der T.P.', Z.Ant., X I , 1961, pp. 1 1 3 - 1 2 5 .

Examples of studies involving the 'Antonine Itinerary' 1. In Gaul BARRIÈRE (P.), 'A propos des voies antiques des Cadurques, Organisation et circulation', R.E.A., LIV, 1952, pp. 102-108. BARRUOL (G.), MARTEL (P.), 'La voie romaine de Cavaillon à Sisteron, sous le Haut-Empire', R.S.L., X X V I I I , 1962, pp. 125-202. BERCHEM (D. van), 'Les origines de l'Itinéraire d'Antonin et ses rapports avec les édits relatifs à la perception de l'annone', B.S.A.F., 1934, pp. 2 1 2 213, cf. Mém. S.N.A.F., 1937, pp. 172 sq. BOUDREAU (M.), 'Voies romaines en pays boien', Bull. Soc. archéol. Bordeaux. LXII, 1957-62 (64), pp. I I I - , I I 8 .

MATHERAT (G.), 'Le problème topographique de Litanobriga', M.A.I., X I V , 1951, 2, pp. 1-60. MERLAT (P.), 'Les noms d'Ouessant', list of place names in Annales de Bretagne, LXII, 1955, pp. 379391, cf. 'Considérations générales sur l'établissement d'une carte du réseau routier en Armorique antique et observations particulières sur une carte des voies romaines de la cité des Vénètes', N.A.A., 1955, pp. 300-332. MUELLER (R.), 'Oppidum Atuatucorum', Gymnasium, L X I , 1954, pp. 326-339. OLDHAM (R. D.), 'The Antonine Itinerary of the Aurelian Road between Aix and Aries', G.J., L X X I I , 1928, p. 58. REBUFFAT (R.), 'Les stations corses de l'Itinéraire d'Antonin', A.F.L.A., XLIII, 1967 (Études classiques, II), pp. 2 1 7 - 2 2 7 . ROBLIN (M.), 'L'emplacement de Litanobriga', B.S.A.F., 1943-1944, pp. 129-130. ROBLIN (M.), 'L'habitat ancien dans la région de Pont-Sainte-Maxence', Mélanges A. Piganiol, Paris (1966), pp. 1 0 8 7 - 1 1 1 0 . 2. For other provinces, the following are useful : BLOCH (R.), 'Une campagne de fouilles dans la vallée du ChélifT, M.E.F.R., 1941-1946, pp. 9-42. CHATELAIN (L.), 'Carte du Maroc romain', B.S.A.F., 1945-1947, p. 195. DESJARDINS (V.), 'Note sur les martyrs de Regia', B.S.G.A.O., L X V , 1944, pp. 7 7 - 8 1 . DERINGER (H.), 'Die römische Reichsstraße AquileiaLauriacum', Carinthia, C X X X I X , 1949, pp. 1 9 3 221. EUZENNAT (M.), 'Les voies romaines du Maroc dans l'Itinéraire d'Antonin', Hommages à A. Grenier, pp. 595-610; cf. Tocolosida (last station in the Antonine Itinerary), B.A.C., N.S. 1 - 2 , 1965-1966 (68), pp. 160-161. GEYER (P.), 'Orum der Rand (It. A. 20)', Archiv, f . Latein. Lexikogr., IX, 1896, p. 300.

Bibliography GOODCHILD (R. G.), 'The Coast Road of Phoenicia and its Roman Milestones', Berytus, IX, 1949, pp. 91-127 (accuracy of the Antonine Itinerary) (cf. B.S.R., X I X , 1951 (52), pp. 43-77: identification of a station in Tripolitania). HURTADO (V.), 'Las provincias hispano-romanas y las vias de Antonino', El Miliario extravagante, V, 1 9 6 4 , pp. 1 0 2 - 1 0 8 .

HUSSEY (Chr.), 'Netherby Hall, Cumberland', Country Life, C V , 1949, pp. 1 4 2 - 1 4 5 . KUCHTNER (L.), ' D i e Entfernungsangaben fur die

Römerstraß Pons Aeni ad Castra im Itinerarium Antonini', Verhandl. d. Hist. Vereins für Niederbayern, LXXVIII, 1952, pp. 94-98. LA LOMIA (M. R.), 'Ricerche archeologiche a Vito Soldano', KSkalos, VII, 1961, pp. 157-165. LIGOTTI (A.), 'Identificazione di Calloniana', A S . S . O . , X I - X I I , 1 9 5 8 - 1 9 5 9 , pp. 1 2 3 - 1 3 0 . LOPES PEGNA (A.), 'Itinera Etruriae', S.E., XXI, 1 9 5 0 - 1 9 5 1 , pp. 4 0 7 - 4 4 2 -

MAYERSON (Ph.), 'The Desert of Southern Palestine according to Byzantine Sources', P.A.Ph.S., CVII, 1 9 6 3 , pp. 1 6 0 - 1 7 2 .

MELONI (P.), 'I miliari sardi e le strade romane in Sardegna', Epigraphica, XV, 1953 (55), pp. 20-50. MERCATI (G.), 'S. Isicius (It. A. 27)', R.Bi., N.S. IV, 1 9 0 7 , pp. 7 9 - 8 0 . MORAN (P. C . ) , GALLENT (G. G.), Vias y

Poblaciones

romanas en el norte de Marruecos, Madrid, 1948. MOUTERDE (R.), 'Une dédicace d'Apamée de Syrie et l'Itinéraire d'Antonin', C.R.A.I., 1952, pp. 355-363. MUELLER (R.), 'Römer und röm. Poststraßen im Jülichen Land', Heimat. Kalender des Kreises Jülich, 1952. PP- 5 6 - 6 4 .

OPELT (I.), 'Das Grab des Riesen Goliath (It. A. 31)', Jb. A.C.,

III, I960, pp. 1 7 - 2 3 .

PASSMORE (A. D.), The Roman Road from Caerleon to Silchester. Itinerarium Antonini Augusti. Iter Britanniarum XIII, Swindon, 1948. RIVET (A. L. F.), 'The British Section of the Antonine Itinerary', Britannia, 1, 1970, pp. 34-82. ROLDAN HERVAS (J. M . ) , ' S o b r e los acusativos con

"ad" en el Itinerario de Antonino', Zephyrus,

X V I I , 1966, pp. 1 0 9 - 1 1 9 .

SAA (M.), AS Grandes Vias da Lusitania. O Itinerario de Antonino Pio, I, Lisbonne, 1956 (57). STEFAN (Gh.), 'Scythica. A propos d'un passage de l'Itinerarium Antonini, 225, 3', Balcania, VII, 1944, 2, pp. 344-348. UGGERI (G.), 'La terminologia portuale romana e la documentazione dell'Itinerarium Antonini', S.I.F.C., XL, 1-2, 1968, p. 225. VELKOV (V.), Klio,

X X X I X , 1 9 6 1 (62), pp. 2 1 5 - 2 2 1

(a castellum in the Antonine Itinerary). WHEELER (G. H.), 'Textual Errors in the Itinerary of Antoninus', E.H.R., 35, 1920, pp. 377-382; 67, 1 9 3 2 , pp. 6 2 2 - 6 2 6 .

233

3. For the Maritime Itinerary : DUPRAT (E.), 'A propos de l'Itinéraire maritime: Citharista=La Ciotat', Institut historique de Provence, Congrès . . . , 1939, 46. LUGAND (R.), 'Notes sur l'Itinéraire maritime de Rome à Aries', Ai.E.F.J?., XLIII, 1926, pp. 124-139. MOUQUET (J.), 'Sur quelques ports de l'Itinéraire maritime', ß . A , X X X I V , 1931, 2, pp. 123-135. UGGERI (G.), 'Sull'Itinerarium per maritima loca da Agrigento', A. et R., N.S., XV, 1-3, April-Sept. 1 9 7 0 , p. 1 0 7 . There is a collection of papers on the Antonine Itinerary in the Bulletin 3-4,1968, of the Société Gaule. Topographical studies making use of several itineraries In general, for the French sections of the itineraries, reference may be made to E. DESJARDINS, Géographie de la Gaule romaine ; Public, des Bénédictins, Historiens de France-, A. MOLINIER, Les Sources de l'histoire de France, Paris, 1901. ALBERTINI (E.), 'Notes critiques sur l'Itinéraire d'Antonin et la Table de Peutinger', M.E.F.R., XXVII, p. 463. ALVES PEREIRA (F.), 'Consideraçôes sôbre a interprataçâo do Itinerarium romano na parte relativa às ligagôes viârias de Olisipo a Emerita', Mem. Acad, das Ciências de Lisboa, Cl. de Letras, II, 1937, pp. 33-111BARADEZ (J.), 'Réseau routier de la zone arrière du limes de Numidie', Limes-Studien, III, Bale, 1957 (59). PP- 19-30. BEJARANO (V.), 'Fuentes antiguas para la historia de Salamanca', Zephyrus, VI, 1955, pp. 89-119. DERINGER (H.), 'Die röm. Reichsstraße AquileiaLauriacum', Carinthia, C X X X I X , 1949, pp. 193— 221.

DOERNER (F. K.), 'Dusae ad Olympum', Studies Robinson, I, pp. 374-379FAVRET (P.-M.), 'Hypothèses sur l'identification des stations des voies romaines de Reims à Verdun et de Reims à Toul', B.C.T.H., 1936-1937, pp. 465474 (cf. id., Riobe, ibid., pp. 461-464). FISCHER (H.), 'Geschichte der Kartographie von Palästina', Z.Pal.V., 1939, pp. 169-189. GOODCHILD (R. G.), 'The Coast Road of Phoenicia and its Roman Milestones', Berytus, IX, 1949, pp. 91-127.

HENSCHELL (H.), 'Die Geographie Südwestgalliens nach den röm. Itineraria', B.Ph.W., 1917, p. 47. MARTIN (J.), 'Geminiacus vicus', Namurcum, 1948, pp. 47-52. MUELLER (R.), 'Die Angaben der röm. Itinerare über die Heerstraße Köln-Eifel-Reims', Festschrift P. Meyer, Münstereifel, 1933; ed. Tabula Germaniae secundae. Orte und Zeugen der Peutingerschen

Bibliography

234

Tafel und des Antoninischen Itinerars, KölnLindental, 1946. MUELLER (R.), 'Oppidum Atuatucorum', Gymnasium, L X I , 1954, pp. 326-339. POLASCHEK (E.), 'Die Tabula Peutingeriana und das Itinerarium Antonini als topographische Quellen für Niederösterreich', Jb. für Landeskunde von Niederösterreich, X X V I , 1936, pp. 38-50. STIGLITZ (H.), 'Rom. Lager . . . am norischen Limes', J.O.E.A.I., X L V I , 1961-1963, Beibl. pp. 1 4 3 - 1 7 2 . STOLTE (B. H.), 'Driemaal Noviomagus', Numaga, XIII, 1966, pp. 233-235. Ptolemy's 'Geography' WILBERG (F. G.), Translation into Latin (6 parts), Essen, 1838-1845. NOBBE (C. F. A.), 3 vol., Leipzig, 1843-1845 (index). LANGLOIS (V.), Géographie de Ptolémée, a photolithographic reproduction of the Greek manuscript in the Vatopedi monastery on Mount Athos, Paris, 1867. MUELLER (C.), Paris, Didot, I, 1, 1883; I, 2, 1901 (Latin translation). CUNTZ (O.), Die Geographie des Ptolemaeus, Berlin, Teubner, 1923, Monumenta cartographica Africae et Aegypti. DANGIOLO (J.), Latin translation, Paris, 1931. STEVENSON (E. L.), English translation, N e w York, 1932. YOUSSOUF K A M A L , C a i r o ,

1926-1951.

Other geographical texts Expositio totius mundi et gentium, editor J. Rougé, Paris, 1966. Dimensuratio provinciarum. cf. A. RIESE, Geographi latini minores, Heilbronn, 1878 and R.E., item 5, 1 (1903), col. 647. Studies of the Ravenna Cosmography (Ravennati Anonymi Cosmographia) E d i t i o n s b y PARTHEY ( G . ) , PINDER ( M . ) a n d SCHNETZ

(JO. PP- 178, 186, 190, 1 9 1 , 223-224, 2 3 4 - 2 3 5 , 239. 259, 267; X V I , 2 , 1 9 5 8 , pp. 335-336, 366,432; X V I I , 2, 1959, pp. 420, 422-423, 428-429; etc. GOODCHILD (R. G.), ' T h e Coast Road of Phoenicia and its R o m a n Milestones', Berytus, IX, 1949, pp. 91-127. HATT (J.-J.), ' D é c o u v e r t e d ' u n e borne r o m a i n e à Seltz', Etudes haguenoviennes, IV, 1962-1964, pp. 363-364HELENA (Ph.), 'La borne milliaire de D o m i t i u s Ahenobarbus et le trajet audois de la voie D o m i tienne', Bull. Comm. archéol. de Narbonne, X X I I , 1949-1950 (51), pp. 88-105. HIRSCHFELD (O.), 'Die römischen Meilensteine', Kleine Schriften, 1907, pp. 7 0 3 - 7 4 4 = Sitzber. Berl. Ale., 1907, p. 165. HULTSCH (F.), Griechische und römische Metrologie, Leipzig, 1864-1866. INNEREBNER (G.), 'Alstraßen der Deutschgegend. Römischer Meilenstein in Laurein?', Schiern, X X X I , 1961, pp. 2 1 0 - 2 1 1 . JALMAIN (D.), 'Lieues romaines', Bull, du groupe archéol. du Nogentais, III, no. 4, pp. 10-11. KÖNIG (I.), ' Z u r Dedikation r ö m . Meilensteine, Digesta 43, 7, 2; 50, 10, 3 - 4 , Chiron III, 1973, pp. 419-427. KÖNIG (I.), Die Meilensteine der Gallia Narbonensis, Berne, 1970. LANFRANCHI (G.), 'Les itinéraires à travers les anciennes unités de longueur', Fiches A.A.A.M., 1 9 7 1 , 1, p p . 223-228. LABROUSSE (M.), ' U n milliaire inédit de Constantin à Castelnau-Magnoac', Pallas, IV, 1956. LABROUSSE (M.), ' D e u x milliaires de la route romaine de Toulouse à N a r b o n n e ' , Pallas, VI-VII, 1958, facs. 3, pp. 55-58. LEBEL (P.), Mém. Comm. Antiq. dépt. Côte d'Or, X X I V , 1954-1958, pp. 22-23 (correction of C.I.L., XIII, 9045).

LELONG (Ch.), 'Sarcophage taillé dans une b o r n e milliaire du III e siècle t r o u v é à Saint-Martin de Tours', R.A.C., J u l y - S e p t . 1969, p. 221. L o u i s (M.), 'Les bornes milliaires de la voie D o m i tienne entre le R h ô n e et le Vidourle', B.C.T.H., 1 9 4 3 - 1 9 4 4 - 1 9 4 5 (51), pp. 547-578. cf P- 228. NESSELHAUF (H.), 'ZU den Funden neuer Leugensteine in O b e r g e r m a n i e n ' , Germania, 1937, pp. 1 7 3 - 1 7 5 . OxÉ (A.), 'Die römische Meile, eine griech. Schöpf u n g ' , B.J., 1 3 1 , 1926, p. 219. Cap. de Frégate PETESCH, 'La "Belle B o r n e " de Cusey', Bull. Soc. hist, et archéol. de Langres, XIII, 174, 2nd quar. trim. 1959, p. m (cf. B.C.T.H., 1959-1960 (62), p. 68). PFLAUM (H. G.), ' D e u x milliaires de Gaule n a r b o n naise', B.S.A.F., 1962 (64), p. 45. PFLAUM (H. G.), 'P. Licinius Gallienus nobilissimus Caesar et i m p . M . Aurelius N u m e r i a n u s nobilissimus Caesar Augustus. A la lumière de d e u x n o u v e a u x milliaires d ' O u m el Bouaghi', B.A.A., II, 1966-1967, pp. 1 5 7 - 1 8 2 . T h e Revue des études anciennes (R.E.A.) gives brief details about n e w discoveries of milestones. ROLLAND (H.), ' U n n o u v e a u milliaire de l'itinéraire de Peutinger', C.R.A.I., 1962, pp. 76-80. SALAMA (P.), ' U n e b o r n e milliaire archaïque de l'Afrique romaine', ibid., 1963, pp. 142-148. Cf. POINSSOT-MERLIN, B.C.T.H., 1 9 3 4 - 1 9 3 5 , p. 3 2 1 . SALAMA (P.), ' D é c h i f f r e m e n t d ' u n milliaire de Lepcis M a g n a ' , Libya Antiqua, II, 1965, pp. 39-45. SALAMA (P.), 'Le milliaire archaïque de Lorbeus', Mél. de Carthage, 1964-1965 (66), pp. 9 7 - 1 1 5 . SANCIER (R.), 'Étude n u m é r i q u e d ' u n tronçon de voie d'Aleth à V o r g i u m ' , Mém. Soc. d'émulation des Côtes-du-Nord, 1963, p. 1. SPRATER (F.), 'Römerstraßen und Meilensteine', Festschr. Reineke, 1950, pp. 131-132. SUSINI (G.), 'La preparazione del corpus dei milliari' (C.I.L., XVII), Epigraphica, X X X , 1968, pp. 1 8 0 181 (cf. H . H . , 'Das corpus miliariorum. Stand der Arbeiten', ibid., X X X I I , 1970, p. 188). TOUTAIN (J.), 'Les n o u v e a u x milliaires de la route de Capsa à Tacape', Mém. S.N.A.F., 1903, p. 153. WALSER (G.), 'Meilen u n d Leugen', Epigraphica, J a n . - D e c . 1969, p. 84 (cf. Actes du Colloque sur les cols des Alpes, B o u r g , 1969 (Orleans, 1971), pp. 55-57)WALSER (G.), ' Z u zwei germanischen Meilensteinen', M.H., X X V I I - 4 , 1970, pp. 255-264. WALSER (G.), ' A n e p i g r a p h e Meilensteine in der Schweiz', Chiron IV, 1974, pp. 457-466. WALSER (G.), 'Die R e p r o d u k t i o n von MeilensteinInschriften', Actes V Congr. intern. Épigr., C a m bridge, 1967 ( O x f o r d , 1971), p. 437. WUILLEUMIER (P.), Inscriptions latines des Trois Gaules, Paris, published b y the C . N . R . S . N e w milestones are recorded f r o m no. 461 o n w a r d . N e w discoveries of milestones are also noted in :

Bibliography Proceedings of the International Congresses E p i g r a p h y (indexes), Année epigraphique, Fasti Archaeologici.

on

Studies of Italian milestones BRACCO (V.), ' L ' E l o g i u m di Polla', R.A.A.N., X X I X , 1954, pp. 5 - 3 7 (cf. i960, p. 149). BRACCO (V.), ' A n c o r a sull'Elogium di Polla', ibid., X X X V , i960, p. 149 (cf. N.Sc., 1953, p. 343). BRUSIN (G.), ' U n n u o v o miliario da Chienes e la strada della Val Pusteria', F.A., III, 4249. CARETTA (A.), ' D u e f r a m m e n t i di colonne miliari dell'agro Laudense', Epigraphica, XI, p. 44. DEGRASSI (A.), ' U n n u o v o miliario Calabro della via Popillia', Philologus, 1955, p. 259. DEGRASSI (A.), ' N o u v e a u x milliaires archaïques', Mélanges A. Grenier, Brussels, 1962, p. 499. D i VITA (A.), ' U n milliarium del 252 a.C. e l'antica via A g r i g e n t o - P a n o r m o ' , Kôkalos, I, 1955, p. 10. D i VITA (A.), ' U n a recente nota e la datazione del miliario siciliano del console C . Aurelio C o t t a ' , Latomus, 1963, p. 478. FORNI (G.), 'La strada r o m a n a da Hadria (Atri) a Pinna (Penne) (a proposito di u n miliario dai pressi di Castilenti)', Abruzzo, I, 1963, 1-2, pp. 3 - 1 1 . LUSSANA (A.), ' A l c u n e osservazioni sulle pietre miliari della Transpadana, della Venezia e della Liguria', Epigraphica, IX, 1947 (49), pp. 6 8 - 8 0 . MELONI (P.), 'I miliari sardi e le strade r o m a n e in Sardegna', Epigraphica, X V , 1 - 4 , p. 20. MIRABELLA ROBERTI (M.), ' N u o v i miliari dalla T r a n s padana e dalla Venetia', Atti III Congr. Intern. epigrafia, R o m e , Bretschneider, 1959, p. 353. PITIMADA (L.), 'S. O n o f r i o Catanzaro, cippo miliare', N.S.A., 1953, 7 - 1 2 (1954), P- 343SCIARRA (B.), ' U n miliario della via Herculea? nel M u s e o provinciale di Brindisi', Epigraphica, X X X I I , 1970, p. 162. SOTGIU (G.), Studi sull'epigrafia di Aureliano, U n i v . di Cagliari, 1961. SOTGIU (G.), ' U n miliario di Gallo e Volusiano " r i s c o p e r t o " a Villanova T r u s c h e d u ' , Studi Sardi, X X I I , 1971-72 (73), pp. 290-292. STYLOW (A. U . ) , 'Ein neuer Meilenstein des M a x i m i n u s T h r a x in Sardinien u n d die Strasse KaralesO l b i a ' , Chiron, IV, 1974, pp. 515-532.

R o m a n roads : construction and course AGACHE (R.), 'Présence de fossés parallèles à certaines voies romaines', Bull. Soc. des Antiquaires de Picardie, 3rd quarter, pp. 258-264 (cf. the special n u m b e r s 5, 6 a n d 7 of the Bulletin de la Société de Préhistoire du Nord). BULLE (M.), 'Geleisestrafien des Altertums', S.B.A. W„ Phil. hist. Kl., Jahrg., 1947, H e f t 2, p. 134.

237

CAILLEMER, 'Les voies à rainures chez les Anciens', Congrès archéologique de France, X L V I , 1880, pp. 277-289. CLAES (P.), 'Les fossés-limites de la chaussée BavaiC o l o g n e dans la région de Liberchies', Helinium, IX, 1969, p. 138. Del PACE (Cl.), 'Sopra l'uso della pozzolana nei cementi degli antichi e specialmente nella costruzione delle strade consolari dei R o m a n i ' , Saggi di dissertaz. acad., Accad. Etrusca di C o r t o n a , I X - V I I . DUVAL (P.-M.), 'La construction d ' u n e voie r o m a i n e d'après les textes antiques', B.S.A.F., 1959, pp. 176-186. FORBES (R. J.), Notes on the history of ancient roads and their construction, A m s t e r d a m , 1934, 2nd ed. 1964. Cf. Studies in Ancient Technology (2. Land transport and road building), II, 2, Leiden, 1955. FOUET (G.), ' Q u e l q u e s coupes de voies romaines régionales', Actes du XXVe Congrès d'Études régionales. ' L u c h o n et les Pyrénées centrales', Tarbes, 1970, pp. 16-24. FUSTIER (P.), ' N o t e s sur la construction des voies romaines en Italie', R.E.A., i960, pp. 9 5 - 9 9 ; 1961, p p . 276-290 (cf. 1958, p. 82). FUSTIER (P.), ' N o u v e l l e coupe de la voie r o m a i n e de L y o n à R o a n n e ' , R.E.A., J a n . - J u n e 1962, p. 124. FUSTIER (P.), ' E t u d e technique sur u n texte de l ' e m p e r e u r Julien relatif à la constitution des voies romaines', R.E.A., L X V , 1963, 1 - 2 , pp. 114-121. GODET (J.), ' D é c o u v e r t e s le l o n g de la chaussée (romaine) B r u n e h a u t à Petit-Enghien', Annales du Cercle archéologique d'Enghien, XI, 1958 (59). HALBERTSMA (H.), ' V e e n b r u g g e n en h u n G e b r u i k . . . ' , Ber. van de rijksdienst . . . oudheidkundig Bodemonderzoek, X I I - X I I I , 1962-1963, pp. 193-209 (road o n a t i m b e r base). JOUDOUX (P.), 'L'art de l'ingénieur chez les R o m a i n s ' , Le mouzi, Sept. 1961, p p . 4 2 - 4 8 ; Jan. 1962, p p . 3 3 - 3 7 ; J u l y 1962, pp. 11-16. LECORNEC 0.), ' S t r u c t u r e de la voie r o m a i n e VannesAngers', Annales de Bretagne, NAA, L X X I I , 1965, pp. 175-178MATHERAT (G.), 'Les ponts de fascines de Jules César à Breuil-le-Sec (Oise)', R.A., 1936, I, pp. 5 3 - 9 4 ; 1937,1. PP- 3 8 - 6 2 . MATTY DE LA TOUR, Les Voies romaines, système de construction et d'entretien, 1865, manuscript, Bibliot h è q u e de l'Institut, 1866, Part 2, Vol. 3: the a u t h o r carried o u t m o r e than 300 sections. MERTENS (].), coupes in F.A., V, 4889; VII, 4432, etc., cf. R.E.A., 1956, p. 290. D E RUETTE (A.), ' H i v e s : coupes de la chaussée r o m a i n e A r l o n - T o n g r e s ' , Ardenne et Famenne, II, 1959. PP- 120-122, 2 fig. SIAT ( A . ) , STIEBER ( A . ) , W E I L ( R . ) , ' N a t u r e

pétro-

g r a p h i q u e des m a t é r i a u x d ' e m p i e r r e m e n t d ' u n e voie r o m a i n e à Dambach-la-Ville', C.A.A.H., X, 1966, p p . 6 9 - 7 1 (cf. ibid., XIII, 1969, p. 39).

238

Bibliography

SUSINI (G. C.), 'Considerazioni sulle carregiate nelle vie urbane romane', A. e M. d. Dep. Storia patria prov. Romagna, X I I - X I V , 1960-1963, pp. 297-300. ULRIX (F.), 'Recherches sur la méthode de traçage des routes romaines', Latomus, X X I I , 1963, 2, pp. 157-181. VASSELLE (F.), 'Coupe de la chaussée romaine à la sortie de Roye, de celles d'Amiens-Vermand et de Saint-Just-en-Chaussée', B.S.A.P., LU, 1966, pp. 216-220. VASSELLE (F.), 'Découverte d'une chaussée romaine à Amiens', ibid. 1967, pp. 1 2 6 - 1 2 7 .

Roman bridges and dykes ALAGNA (E. P.), 'I ponti romani di Santa Marinella', Boll. Informazioni associaz. archeol. Centumcellae, IV, 1962-1963, no. 4, pp. 42-46. ALLAIN (Dr.), 'Pont gallo-romain et gué pavé à Argenton-sur-Creuse, Indre', R.A.C., II, 1963, p. 1 3 1 . AUBIN (H.), 'Die Rheinbrücken im Altertum und Mittelalter. Eine Kriegs - und wirtschaftsgeschichtliche Studie', Rhein. Vierteljahrsbl., VII, 1937, pp. 111-126. BAKALAKIS (G.), 'The "Classical" bridge at Amphipolis', A.J.A., 74, 3, July 1970, p. 289. BALLANCE (M. H.), 'The Roman Bridges of the Via Flaminia', P.B.S.R., X I X , 1 9 5 1 , pp. 7 8 - 1 1 7 . BARCACILA (A.), in Stud. Cere. Istor. Veche, X V I I , 4, 1966, p. 645 (on scene C I of Trajan's Column). BARRUOL (G.), 'Le pont romain de Ganagobie', Gallia, X X I , 1963, 2, pp. 3 1 4 - 3 2 3 . BEDESCHI (V.), 'Una strada e un ponte in Brescia antica', Commentari dell'Ateneo di Brescia, 1 9 4 6 1947, pp. 69-76. BLANC (A.), 'Ponts gallo-romains et très anciens de l'Ardèche et de la Drôme', Gallia, X X V , 1, 1966, pp. 77-99. BLANCHET (A.), 'Découvertes faites récemment dans le lit de la Loire, à Ancenis', B.S.A.F., 1950-1951, p. 28. BLONDEL (L.), 'Le pont romain de Genève', Bull. Soc. hist, et archéol. Genève, V, 1 9 2 5 - 1 9 3 4 , pp. 1 2 8 - 1 4 0 . BLONDEL (L.), 'L'emplacement du pont de César sur le Rhône', Genava, 1938, pp. 1 0 5 - 1 1 5 . BLONDEL (L.), 'Pont romain de Genève', ibid., 1954, p. 205. BUNDGÂRD (J. A.), 'Caesar's Bridges over the Rhine', A.Arch., X X X V I , 1965 (66), pp. 87-103. COLOMB (J.), Ponts du Moyen Age, Cannes, 1967. CÜPPERS (H.), 'Vorröm. und röm. Brücken über die Mosel', Germania, 45, 1967, 1 - 3 , pp. 60-69. CÜPPERS (H.), Die Trierer Römerbrücken, Mainz, Philipp von Zabern, coli, du Rheinisches Landesmuseum, Trierer Grabungen u. Forsch. V, 1969. DEGRASSI (N.), 'Il ponte romano di Olginate e la Strada da Bergamo a C o m o ' , Riv. archeol. d. antica

prov. e diocesi di Como, f. 127, 1946, pp. 5 - 2 3 . DONINI (A.), Ponti su monete e medaglie, Rome, 1959, Pontif. Univ. Gregoriana. DUBOIS (J.), 'La chaussée des Césars et le pont de Saint-Avertin', B.S.A.T., X X X V I , 1970, pp. 1 0 9 - 1 4 1 , 13 fig. DYMOND (D. P.), 'Roman Bridges on Dere Street, County Durham : with a General Appendix on the Evidence for Bridges in Roman Britain', Arch.J. C X V I I I , 1961, pp. 1 3 6 - 1 6 4 . ERSOVIC (N. I.), ' A u sujet du passage du Rhin par Jules César', V.D.I., 1968, 140, pp. 1 2 0 - 1 3 0 . FUGIER (H.), 'Le latin pontifex et les faiseurs de chemins latins et indiens', R.E.L., 1961 (62), p. 68. Gallia, bridges or piers at : Baudven (XII, 2,1954,439) ; Besançon, X I , 1 , 1953, 1 3 4 ; Champagne (timber), X I X , 2, 1961, 424, fig. 4 6 - 5 1 ; Kembs, VIII, 1950, 166; Metz (piles), X V I , 2, 1958, 328; Pont d'Areuilhes, X V I I , 2, 1959, 4 1 6 ; Pont des Arches, X X , 2, 1962, 529; Substantion, X I X , 1, 1961, 1 2 ; Vaison, XII, 2, 1954, 460; Villeperdrix, X V I I I , 2, i960, 372, fig. 16. GALLIAZO (V.), I Ponti di Padova romana, Padua, 1971. GAZZOLA (P.), Ponti romani, Contributo ad un indice sistematico con studio critico bibliografico, Florence, Olschki, 1963. GOSE (E.), 'Neue Beobachtungen an der Römerbrücke', T.Z., X X V I I , 1964, p. 153. HARMAND (J.), 'Sur les "ponts de fascines" de Breuille-Sec', B.S.A.F., 1959, p. 2 7 1 . HATT (J.-J-), 'Découverte des vestiges d'un pont romain à Kembs', C.A.A.H., 1952, p. 83. KRAUS (O.), 'Die römische Rheinbrücke bei Köln', B.J., C X X X , 1935, p. 232. LABROUSSE (M.), Toulouse antique, Paris, 1968 (on bridges, pp. 303-309)LAMBOGLIA (N.), 'Restauri e ricerche sui ponti romani (Vado Ligure)', Riv. Ingauna, 1952, 1 , p. 24. LAMBOGLIA (N.), 'Restauri e ricerce sui ponti romani di Val Ponci (Finale Ligure)', Riv. Ingauna, J a n . March 1954, pp. 9 - 1 4 . LAUR-BELART (R.), 'Die Römerbrücken von Äugst im hochrheinischen Strassennetz', Festschrift E. Vogt, Zürich, 1966, pp. 241-246. LEANING (J. B.), 'The Date of the Repair of the Bridge over the River Chabina', Latomus, X X X , 2, A p r . June 1 9 7 1 , p. 386. MANIÈRE (G.), 'Voies et ponts antiques dans la commune de Saint-Martory (Haute-Garonne)', Gallia, X X V I I , 1969, 2, p. 163. MATHERAT (G.), 'Les ponts de fascines de J . César à Breuil-le-Sec (Oise)', R.A., VII, 1936, pp. 53-94. (Cf. B.S.A.F., 1935. P- 89-) SCHETTER

(W.),

VON U S L A R

(R.),

'Zu

den

pontes

longi' (Tac., A., I, 63, 3), Gymnasium, L X X X V I I , 1 9 7 1 , pp. 201-224. SCHWAB (H.), 'Keltische Brücke zw. Cornaux (Neuenburg) und Gals (Bern)', S.P., X X X , 1966,

Bibliography pp. 9-10. TUDOR (D.), ' U n p o n t r o m a i n i g n o r é dans la région d u B a s - D a n u b e ' , Latomus, 1 9 6 1 , p. 501.

R o m a n roads and the study of place-names (toponymy) T h e m a j o r i t y o f w o r k s o n t h e ancient roads t o u c h o n t h e question. T h e list g i v e n b e l o w concentrates m a i n l y o n p u b l i s h e d material dealing w i t h r o a d - n a m e s ( h o d o n y m y ) . W h i l e t h e e x a m p l e s q u o t e d are p r i n cipally French, studies are also q u o t e d f o r o t h e r countries to help in t h e c o m p i l a t i o n of series for c o m parative purposes, as r e c o m m e n d e d b y t h e a u t h o r in 'Pour une toponymie archéologique comparée', Hommages Herescu, 1964, p p . 93-96. AEBISCHER (P.), ' L ' é t y m o l o g i e d u français "chaussée" et u n passage d u " R o m a n d u B r u t " ' , Onomastica, V , 1953. PP- 1-7ANDRÉ (J.), 'Les n o m s latins d u c h e m i n et de la r u e ' , R.E.L., 1950, pp. 1 0 4 - 1 3 4 . ARMAND-CALLIAT (L.), 'Lieux consacrés à M e r c u r e dans la m o n t a g n e chalonnaise', B.C.T.H., 19431945, p. 167. BLANCHET (A.), ' H y p o t h è s e à p r o p o s de la voie R é g o r d a n e ' , C.R.A.I., 1945, p p . 4 5 8 - 4 6 3 . BONFANTE (G.), ' T r a c c e di t e r m i n o l o g i a palafitticola nel v o c a b o l a r i o latino?', A.I. ^ . . X C V I , 1 9 3 7 - 1 9 3 8 , p. 62 (this hypothesis, accepted b y G . DUMÉZIL, Jupiter, Mars, Quirittus, p. 126, is rejected b y G. PATRONI, R.I.L., L X X I I , 1 9 3 8 - 1 9 3 9 , p p . 16-24). BRUNEL (Cl.), 'Le n o m de la voie R é g o r d a n e ' , Romania, 1958, p p . 2 8 9 - 3 1 3 . BRUNOT (F.), Histoire de la langue française. Vol. VII, Paris, A . C o l i n , 1926, p p . 2 0 2 - 2 2 7 ( n e w edit. 1967). BUCQUET (H.), ' Q u e l q u e s précisions sur les " c h a u s sées" anciennes et m o d e r n e s ' , R.I.O., V, 1953, pp. 129-130. CARNOY (A.), ' T o p o n y m i e des chaussées r o m a i n e s en B e l g i q u e et dans les régions avoisinantes. Essai d ' h o d o n y m i e . ' , A.C., X X I I I , 1954, P a r t 1, p p . 5-28. CHENON (E.), 'Les termes " g r a n d c h e m i n " , " c h e m i n r o y a l " et les anciennes voies r o m a i n e s au M o y e n A g e ' , B.S.A.F., 1925, p p . 3 0 0 - 3 0 1 . CHEVALLIER (R.), La Toponymie des voies antiques, N o t i c e t e c h n i q u e n o . 5, G r o u p e d ' A r c h é o l o g i e a n t i q u e d u T . C . F . , Paris. CHRISTMANN (E.), ' " K e r n , K i m , K ü m m e l " , als Benennungen für Römerstraßen von LuxemburgM e t z bis S p e y e r - L a u t e r b u r g ' , Germania, Anzeiger, 27, 1943, PP- 72-79CHRISTMANN (E.), KAISER (K. W.), 'Beiträge der F l u r n a m e n f o r s c h u n g zur R ö m e r s t r a ß e n f o r s c h u n g in d e r Pfalz', B.N., N . S . , I, 1966, pp. 1 7 9 - 2 1 3 . CONNEAU 0 . ) , 'La P o u y a d e , La P o u j a d e , La P o u g e , n o m s de voies antiques', Bull, archéol. du Vexin français, 4, 1968 (69), p. 1 1 9 .

239

CONNEAU (J.), Les Appellatifs de voies antiques et leurs dérivés en toponymie française, duplicated typescript, 1969. D u CANGE, sections: 'calceia', ' c a m i n u s ' , 'leuga', 'ru(t)a'. ESCALLIER (E.), ' U n itinéraire alpin d e G a r g a n t u a ' , Bull. Soc. Et. H.A., 1954, p. 4 2 ; 1956, p. 94. FOUCAULT (J. A. de), 'La r o u t e dans la t o p o n y m i e d u Pas-de-Calais', R.I.O., J u n e 1968, p p . 1 0 1 - 1 0 5 . FOURNIER (P.-F.), ' L ' o r i g i n e d u m o t chaussée', Bull, philol. et hist, du C.T.H., I960 (61), p p . 4 1 - 5 3 . GILLES-GUIBERT (M.), 'Le c h e m i n dans la t o p o n y m i e d u M i d i de la France', École nat. des Chartes. Positions des thèses, 1952, p p . 6 3 - 6 7 . GILLES-GUIBERT (M.), ' N o m s des r o u t e s et des c h e m i n s dans le M i d i de la France au M o y e n A g e ' , Bull, philol. et hist, du Comité des T.H., I960 (61), PP- 1-39GRENIER (A.), 'La t o p o n y m i e des voies gallor o m a i n e s ' , R.C.C., 1 5 - 3 0 M a r c h 1930. GRIERA (A.), ' A l g u n a s designaciones del c o n c e p t o via en catalan, via Domitia, I. Études de linguistique consacrées au S u d de la France et au N o r d d e l'Espagne', Annales Fac. Lettres Toulouse, 1954. GYSSELING, Mededelingen van d. Veren. van Naamkunde te Leuwen, 1957, 24 (names related t o roads in N o r t h e r n Gaul). HARSH (P. W . ) , ' A n g i p o r t u m , platea a n d vicus', C.Ph., X X X I I , 1 9 3 7 - 1 9 5 0 . HERBILLON 0 . ) , ' R o u t e s et c h e m i n s en t o p o n y m i e g a l l o - r o m a i n e ' , R.B.Ph., X L V I , 1968,1, p p . 81-86. HEURGON (J.), 'La f i x a t i o n des n o m s de lieux en latin d ' a p r è s les itinéraires routiers', ibid., X X V I , 1952, pp. 169-178. HOCHULI (E.), Einige Bezeichnungen für den Begriff Straße, Weg und Kreuzweg in Romanischen, Z u r i c h A a r a u , 1926. HOYOUX (J.), ' F i g u r e et destinée d e la chaussée B r u n e h a u t au M o y e n A g e et à l ' é p o q u e m o d e r n e ' , Bull. Inst, archéol. liégois, L X V , 1945, p p . 7 1 - 9 4 . IRIGOIN (P.), ' M o n t j o i e s et oratoires', B.M., X C I V , 1935, PP- 145-170. LANG (L.), 'Les lieuxdits " S p i e g e l b e r g " et leur i n t é r ê t a r c h é o l o g i q u e ' , Soc. d'hist. et d'archéol. de Saverne, L I I I - L I V , 1966, p p . 4 4 - 4 5 . LAUR-BELART (R.), ' Z w e i alte Straßen ü b e r den B ö z b e r g ' . S.P., O c t . 1968, p . 30; cf. X X X I I , 4, 1968, p. 72. LEBEL (P.), 'Voies r o m a i n e s d ' A f r i q u e et de G a u l e ' , R.A.E., II, 1 9 5 1 , p. 192 ( c o m p a r i s o n of n a m e s o f road-stations a n d inns) (cf. ibid., 1950, p. 31 ; 1953, P- 353)LOUIS (R.), ' A p r o p o s des M o n t j o i e s a u t o u r d e Vézelay. Sens successif et é t y m o l o g i e d u n o m " M o n t j o i e " ' , Pub. annuelles Soc. des fouilles . . . Yonne, p l a c e - n a m e series, I, A u x e r r e , 1939. MATHERAT (G.), 'Les chaussées B r u n e h a u t et les anciennes voies r o m a i n e s en B o u r g o g n e et en

240

Bibliography

Picardie', Annales de Bourgogne, XI, 1939, pp. 121-126. M U R E T (E.), ' N o m s de lieux, vestiges archéologiques et vieux chemins', Revue d'hist. suisse, II, 1931, pp. 409-420. NIEDEREHE (H. J.), Strasse und Weg in der galloromanischen Toponomastik, Geneva-Paris, 1967. OLIVIERI ( O . ) , 'Di alcune tracce di vie romane nella toponomastica italiana', A.G.I., 1934, pp. 200-201. PALLU DE LESSERT (C.), 'La syntaxe des routiers romains et les déformations des noms de lieux dans l'Afrique romaine', Mém. Ant. Fr., LXV, 1906, pp. 115-138. QUENTEL (P.), ' U n n o m des anciennes routes, C a r haix', R.I.O., XVIII, 1966, pp. 255-270. ROBLIN (M.), 'Enquêtes pour l'hagiotoponymie française', Annales, 1934, p. 342. ROBLIN (M.), 'Les Montjoies', B.S.A.F., 1945, pp. 45-47 (