[image ALT: Much of my site will be useless to you if you've got the images turned off!]
Bill Thayer

[image ALT: Cliccare qui per una pagina di aiuto in Italiano.]

[Link to a series of help pages]
[Link to the next level up]
[Link to my homepage]

[image ALT: link to previous section]
Ch. 17
This webpage reproduces a chapter of
The Romans
on the Riviera and the Rhone

by W. H. Hall

originally published by Macmillan & Co., Ltd.,

The text is in the public domain.

This page has been carefully proofread
and I believe it to be free of errors.
If you find a mistake though,
please let me know!


[image ALT: link to next section]
Ch. 19

 p170  Chapter XVIII
Via Aurelia

Section 3: Julia Augusta.

The construction by Augustus B.C. 12 of the Via Julia Augusta between Vada Sabata and the Var (about 93 miles in length) served at the same time to prolong the inland Trans-Apennine section of the Via Aemilia Scauri and its coast section (afterwards Via Aurelia) to the Var.

The section of the Julia Augusta between Vada and the Var was therefore the only portion of that road which Augustus had to make entirely new. For the rest of its course between Placentia (the Trebia) and Vada Sabata he had only to improve parts of two roads previously existing. The entire course of the Via Julia Augusta was thus made up:


Part of the Via Postumia repaired from Placentia to Dertona


Part of the Via Aemilia Scauri repaired from Dertona to Vada


Section of the Via Julia Augusta made new from Vada to the Var


The course of the Via Julia Augusta is fortunately determined for us by the inscription on one of its milestones  p171 found in the Valley of Laghet, one mile to the westward of the monument of Turbia. It is identical with the Via Aurelia in its course along the coast, between Vada and the Var, and only dropped its original name — Julia Augusta — when it became absorbed by the Aurelia of the Antonine Itinerary, as was the case with the Via Aemilia to the east of it, and the Via Domitia to the west.

In making two distinct roads of the Aurelia and the Julia Augusta, M. Lentheric is undoubtedly wrong. Had he taken the trouble to study the Corpus Inscriptionum Latinarum or the inscriptions on the numerous milestones to which he vaguely alludes as found in the territory of Nice,1 he would have avoided the error into which he has fallen of prefixing to his Riviera, Ancient and Modern a map showing the Via Julia Augusta branching from the Aurelia at Turbia and leading from Vence to Frejus by Auribeau. When Augustus carried his Via Julia Augusta over the Alps at Turbia, there was no Via Aurelia to branch off from, nor at any later period, as there never was but one Roman road there at any time. For the later Roman road to Vence (Vintium) branched off the Aurelia at Antibes.

The milestone which decisively determines the direction of the Via Julia Augusta is one of the three with which the visitor is confronted immediately on entering the Municipal Library at Nice. It is numbered 8,102, and is entered, Vol. V Pt. 2, p955 of the Corpus Inscriptionum Latinarum2 amongst the 26 milestones therein numbered from 8083 to 8109 inclusive as belonging to the Via Julia Augusta. That road is described by Mommsen as leading

"Placentiâ Vada, Vadis ad Varum," i.e. from Placentia (synonymous with the river Trebia, there forming its  p172 junction with the Po) to Vada (Sabata) and from Vada to the Var. The inscription runs:


Imp. Caesar. divi

Trajani Parthici Filiusº

divi Nervae N. Traja-

-nus Hadrianus Aug

pont. max. trib. potest. IX

cos. III viam Juliam

Aug. a flumine Treb

bia,º quae vetustate

interciderat, sua

pecunia restituit


The number CCXVI, at the top, gives the distance of that milestone from the 'Trebbia,' while that at the bottom, DCV, shows the total mileage from Rome by the circuitous route via Placentia and Rimini, the Julia Augusta having, so to speak, running powers onwards to Rome, over the Via Aemilia Lepidi and the Via Flaminia. The smaller letters in the number CCXVI are not decipherable in the original inscription, but Mommsen has supplied them on the convincing ground that another milestone (No. 8095), which stood four miles nearer the Trebbia, is found inscribed CCXII at the top and DCI at the bottom.

In Vol. over Part 2, p828, Mommsen gives the following table of the composition of the inland circuitous route from Rome to the Var, which Augustus adopted in preference to the coast road, in spite of its being about 200 miles longer:

Via Flaminia. Roma Ariminum m. p. c. CCXXI
Via Aemilia Lepidi. Arimino Placentiam " CLXVIII
Julia Augusta Via Postumia. Placentia Dertonam " LII
Via Aemilia Scauri. Dertona Vada LXXIX
Strata ab Augusto. Vadis ad Varum A. V. C. 742 (B.C. 12) XCIII

fiunt m. p. circiter


 p173  Augustus apparently considered that the alternately swampy and precipitous nature of the coast between Rome and Genoa rendered the effective maintenance of a reliable thoroughfare along the shore an impossibility. The Via Aurelia of the Antonine Itinerary, which attempted it for a time, became practically impassable, as early as the beginning of the 5th century of our era. We gather this from the following passage in the charming poem of Rutilius Namatianus, describing his return journey from Rome to his native Gaul in the reign of Emperor Honorius. In it, he explains that he preferred the risks of the sea to those of the land route:—

"Electum pelagus: quoniam terrena viarum

Plana madent fluviis, cautibus alta rigent:

Postquam Tuscus ager, postquamque Aurelius agger

Perpessit Geticas ense vel igne manus,

Non Silvas domibus, non flumina ponte cohercet:

Incerto satius credere vela mari."a

In the age of Dante, the coast road between Lerici (on the bay of Spezia) and Turbia had become proverbial for its impracticability:

Tra Lerici et Turbia, la più diserta,

La più romita via, è una scala

Verso di quella, agevole ed aperta.

Il Purgatorio III.49.3

It was necessary, in order to unravel the complications of the Via Aurelia, to explain the course of the Via Julia Augusta, the coast section of which enters into its composition. If, however, one follows the course of the via Aurelia as laid down by the Itinerary of Antonine, in its entirety, one would also have to include as "Coast-road," or Aurelia, the part of the inland course of the Julia Augusta as far as Dertona and thence back to Genoa by the Via Postumia. But this would be  p174 obviously absurd, as the essential feature of the Via Aurelia (viz. "coast-road") would be lost and the distance from Vada to Genoa, 28 miles by the coast, would be increased to 130.

In my opinion we are justified in declining to follow this aberration of the Itinerary, which is to be ascribed to an error of the copyist, for whom the intricate nature of the whole subject provides some excuse.

It seems much more reasonable to adopt the short cut or compendium indicated by the Table of Peutinger. Such an error goes a long way to explain the exaggeration in the impossible total 797 miles, given in the heading of the Via Aurelia of the Itinerary as the distance from Rome to Arles, which is really about 600 miles by rail via Vintimiglia and Marseilles.

Between Vada Sabata and the Var, the thread of the Via Julia Augusta may still be picked up at not unfrequent intervals. It sometimes followed the shore and was sometimes carried a mile or more inland. It was carried over the marshy ground to the east of Albenga, partly on a raised causeway, and partly on arches, which for a length of nearly 200 yards are still extant, forming a striking monument of Roman solidity now known by the name of 'Ponte Lungo.'

[image ALT: missingALT.]

Ponte Lungo near Albenga

Between Albenga and Alassio the Roman road forms one of the most charming walks on the whole Riviera, being carried under olive groves round the shoulder of the mountains for about four miles at a considerable height above the sea, of which it commands glorious views.

On both sides of the picturesque bridge, partly Roman and partly mediaeval, at Andora, the Strada Romana may again be traced, as it crosses the wide valley. Between Diano Marina and Oneglia, the Roman road is (I believe) identical with the steep winding track overhanging the sea, which is much to be recommended to pedestrians, as forming a picturesque short cut between the above-named places.

 p175  From Bordighera to Vintimiglia, the Strada Romana runs parallel with the sea, about a thousand yards inland, past the chapel of St Roche, which displays a Roman entablature let into an angle of the wall.

The passage of the road through Vintimiglia is attested by three milestones preserved in the disused church of San Michele — a most picturesque edifice, on the extreme northern edge of the precipitous promontory, on which the now dismantled fortress is built.4

Between Vintimiglia and Mentone, as all the world knows, the Via Julia Augusta is to be seen in all its naked simplicity, as it is carried over the rocks as you approach along the coast those modern gardens of the Hesperides where the Commendatore Hanbury delights to exercise much appreciated hospitality.

As the Roman road rounds the corner of the Red Rocks (Balze Rosse) famous for its cave dwellings, where prehistoric human skeletons may still be seen in situ, its narrow proportions are proved by the 7 ft.‑8 ft. wide ledge chiselled out of the rock for its passage.

In a narrow roadway winding through the western end of Mentone, the Strada Romana may again be recognized and followed across low-lying orchards before it scales the high ground, overlooking the Cap Martin, where the Roman Station of Lumone was situated.

Crossing the col, which separates the bay of Mentone from that of Monaco, the Via Julia Augusta sweeps round the foot of the village of Roquebrune, whence it mounts by an ascent, nowhere difficult, to Turbia distant six miles from Lumone. It is a walk strongly to be recommended and quite within the compass of an  p176 ordinary pedestrian, starting as the case may be either from Mentone or Monte Carlo, and using the funicular railway at whichever end is most convenient.

In its descent from La Turbie (the Alpe Summa of the Antonine Itinerary) towards the Var, the Via Julia Augusta entirely avoided the difficulties and dangers encountered in the carrying of the Cornice road along the face of the precipices behind Eze. It turned off the main road, about half a mile from the monument of Augustus, to the right down the valley of Laghet, which offers an easy and inviting short cut to Cimiez — the Roman Cemenelum — by La Trinité in the valley of the Paillon. For the Romans were making for Cimiez, avoiding the Greek town of Nice, which remained dependent on Marseilles after the establishment of the Roman Empire, as Strabo informs us.5

It is surprising how little attention has been paid to the valley of Laghet, which must have served from the earliest times as the main thoroughfare between Italy and Gaul. With their practical eye, the Romans seized on it at once, as a natural opening ready made for their road.

While the distance from Turbia to Cemenelo is only nine miles by the valley of Laghet, it must be nearly 20 by the circuitous route of the Cornice, where a road never existed before the time of Napoleon.6 In carrying his Via Aurelia by the Cornice, as a separate road from the Julia Augusta, M. Lentheric was clearly in error, as I have already pointed out.

Neither the Itinerary of Antonine, nor the Table of Peutinger shows more than one road, which avoiding Nice, as a Greek settlement, made straight for Cimiez by the valley of Laghet. Nine of the 26 milestones  p177 belonging to the Via Julia Augusta noted by Mommsen were discovered in the valley of Laghet, whereas not a single one has ever been found on the Cornice road.

The modern road, leading to the famous monastery of Laghet, branches off from the Cornice road within half a mile of the village of La Turbie, and the Via Julia Augusta in its turn again branches off from the modern Laghet road a few hundred yards further on, being cut along the face of Monte Sembola. It is only traceable now as a road as far as a quarry, about half a mile from its bifurcation from the modern road, which descends the hill to the monastery.

[image ALT: missingALT.]

Roman milestone
in valley of Laghet,
near Nice

Beyond the quarry, the Via Julia Augusta becomes a mere footpath, and so continues, till it is lost amongst terraced patches of cultivation on the slopes of Monte Sembola. The milestone represented on the opposite page is still in situ, although the road has collapsed entirely. This illustration is from a drawing made by Sir James Harris, the British Consul at Nice, to whom I am much indebted for conducting me to the spot.

The inscription on the milestone proves that it is one of those originally placed there in the reign of Augustus, as his name is still quite legible on it as well as the distance from Rome, DCVII miles. It is numbered 8105 in the Corpus.7

On a later occasion, when we revisited the valley of Laghet together, I was fortunate enough to find and draw the attention of Sir James Harris to another Roman milestone in situ, which had escaped his observation. It is distant exactly one Roman mile from Turbia and marked DCV from Rome.

This milestone appears to be a duplicate, as far as the mileage from Rome is concerned, of that presented by Sir John Boileau to the Municipal Library of Nice,  p178 the inscription on which I have quoted above in full. But the milestone DCV still in situ is one of those originally set up by Augustus, whereas that in the Municipal Library marks the restoration of the Via Julia Augusta by Hadrian. It is numbered 8101 in the Corpus, where the inscription is given.

It was a common practice of Roman Emperors to place milestones referring to their own work of reparation by the side of those of their predecessors. Three of the nine milestones discovered between Turbia and Cimiez are duplicates, in the sense of marking the same total of miles from Rome.

Besides the nine having inscriptions of some kind, there are several other half-milestones which may or may not be in situ. Other fragments have been built into terrace walls, — one of which I examined in the company of the British Consul and Vice-Consul.

To excuse himself for breaking up a milestone which he found on his ground, a peasant-owner remarked to us that there was no object now in maintaining records whole, since the territory around Nice had been transferred from Italy to France in 1860.

No spot in Europe has proved so rich in yield of Roman milestones as the valley of Laghet, and yet it is still uncertain at what point the Via Julia Augusta crossed the Paillon to mount up to Cimiez.

The Author's Notes:

1 Riviera, Ancient and Modern, p14.

2 Vol. V in two parts can be inspected at the Municipal Library.


'Twixt Lerici and Turbia, the most desert,

The most secluded pathway is a stair

Easy and open, if compared with that.

Longfellow's Translation.

4 In his Storia de Vintimiglia, sold at the "Drogheria" in the main street of the Upper Town in the absence of a bookseller's shop, Signor G. Rossi cites a letter from Caelius to Cicero bemoaning his bad luck in being sent to Vintimiglia amidst the snows of the Alps at Christmas time to quell a riot occasioned by the assassination by adherents of Pompey of one of their citizens for having entertained Julius Caesar on his way to Spain at the outset of the Civil War. This as far as I know is the solitary classical text referring to the presence of Julius Caesar on the Riviera.

5 Strabo, IV. Cas. 184, ἡ μὲν Νικαία ὑπὸ τοῖς Μασσαλιώταις μένει καὶ τῆς ἐπαρχίας ἐστίν.

6 The rough track forming a short cut from Riquier to Villefranche is probably all that remains of the old Greek road, which was bound to keep near the shore.

7 The words "Tribunicia Potestate," of which the initial letters are wanting on the milestone opposite, would supply a clue to its date, if followed, as they usually are, by a number, indicating the number of times the emperor in question had been invested with the tribuneship.

Thayer's Note:

a The entire poem is onsite; the passage cited is I.37 ff.

[image ALT: Valid HTML 4.01.]

Page updated: 26 Aug 16