About UsContact

Research on Pompeii
Research Projects on Pompeii
Topography Online Searchable Bibliography,  Book Reviews, eBooks Literature, Inscriptions, Graffiti Vesuvius,  Museums, Images,  Representations,  VisitPompeii in the NewsLinks to other Online ResourcesSearch the SiteClick Here to Join the Pompeii Group


Directionality of Pompeii’s Urban Streets



For over two centuries, ruts in the streets of Pompeii (Figure 1) have been looked on as merely curious and quaint reminders of the survival of the mundane. Then, in 1991 Sumiyo Tsujimura published, "Ruts in Pompeii - the traffic system in the Roman city" [1] , and for the first time ruts were systematically considered for their informational potential. This form of evidence has no longer gone unnoticed. [2] In the following study of traffic patterns in Regio VI (Figure 2), I will supersede Tsujimura by combining ruts with diagnostic wearing on curbstones and the width of each street to reconstruct the directional flow of ancient traffic. Additionally, observations on the nature of a possible system of traffic regulation are noted and combined with the physical evidence to suggest how such a system operated in the northwest of Pompeii [3] . The streets to be discussed are Via Consolare, Vico di Mercurio, Via di Mercurio, Via della Fortuna / Terme, Vico dei Vettii, and Via del Vesuvio (Figure 3).


Ruts are an important remnant of past behavior for a number of reasons. First, they are positive proof of the presence of wheeled traffic. Second, by being a continuous form of evidence, they connect distant and diverse areas in an unbroken physical line. Ruts chronologically unify a physically disperse but continuous behavior; [4] in essence, ruts stretch time across space. [5] Finally, ruts instruct the system of traffic regulation by being positional. The existence of rutting on one street proves the possibility of travel while the absence of rutting can suggest a restriction on traffic.


Rutting has two forms: straight (Figure 4) and curved (Figure 5). The absence of rutting also informs the traffic system by suggesting the absence of traffic. In Figure 6, a stone blocks the street, dramatically emphasizing the restriction on travel implied by the ruts. Traffic volume was also reduced by the width of the street. Some streets could carry two-way traffic (Figure. 7) while others were only wide enough for one-way at a time (Figure 8). [6]


Rutting in the street, whatever its width, is most important at intersections. Here ruts mark not only activity but also decision. Going straight on a straight street merely shows conformity to the curbstones, while turning or continuing on at an intersection demonstrates intentionality. Intersections prove where traffic went and imply where it might not have been allowed. Of course, the absence of evidence, is never proof positive. Still, the absence of ruts in one area or direction, combined with the preponderance of ruts in the another direction can suggest that cart drivers [7] shared a common mental template of allowed action. 


Let us examine the intersection of Vico di Mercurio and Via Consolare (Figure 9) to see the interaction of these factors. This intersection is the example par excellence of this methodology, and the benchmark from which much of the rest of Regio VI is measured. Figure 10 is a view east down Vico di Mercurio at Via  Consolare. One can clearly see the southern rut curving northward and the parallel northern rut. [8] Figure 11 is a view from the south.


More importantly, the wearing on the curbstone requires that the cart was moving west bound and therefore turning to the north. Figure 11a offers a closer look at the curbstone, illustrating this diagnostic wearing pattern, while Figure 12 is a diagram of how this process occurs. A west bound cart turning to the north will cause both curved ruts and a diagonal cut across the stepping stone by constant grinding from iron wheel rims trying to make an ever tighter turn. The reciprocal wearing by the opposite wheel against the curbstone allows one to trace the course of contact. As the cart moves west and begins the turn to the north, the wheel turns counter-clockwise and comes into contact with the curbstone. The wheel remains in contact as it rotates downward, marking a diagnostic curve of wearing. Moreover, the depth of the wear increases as it moves west, further demonstrating a tight turn to the north.


Therefore, one may summarize the physical evidence from this intersection as follows (Figure 13) Ruts curve into the intersection, positively identifying either a south to east or west to north turn. Wearing on the stepping stone adds emphasis to the story of the ruts. Wearing on the north curbstone is diagnostic of a west to north turn, establishing direction. The tightness of the turn indicates that carts were trying to join the northbound lane of traffic along the east curb and avoid the southbound carts along the west curb. Furthermore, there is an acute absence of evidence for a north to east turn [9] as well as for a wider, north bound turn to the west side of Via Consolare. [10] Translating the physical evidence into a structure of traffic (Figure 14), one can now say that Vico di Mercurio was strictly one-way to the west with the majority of its flow exiting to the north. Additionally, Via Consolare was a two-way street with southbound traffic in the west and north bound traffic in the east. Finally, Via Consolare answers the pivotal question of which side of the road the Pompeians drove. They drove on the right. [11]


Determining which side of the street the Pompeians drove is key to decoding two-way streets such as Via Consolare and Via della Fortuna. Therefore, evidence from a single intersection is not sufficient. Figure 15 shows the visible wearing on a stone standing just off the north curb of Via della Fortuna (Figure 16). [12] The downward wear, though over a much shorter distance is consistent with westbound cart traffic. Repeated action of wheels rolling along the right (north) curb [13] caused this diagnostic wear, yielding evidence from outside an intersection. This westerly direction is supported a block further by wear on a fountain guard stone (Figure 17) just west of Vico dei Vettii (Figure 18).


A topographic survey of Regio VI can determine several facts which add to the conception of traffic flow (Figure 19). First, by width and by being dead-ends [14] , most of the streets north of Vico di Mercurio had to carry carts in both directions, one at a time, as alternating one-way streets. [15] Furthermore, the southern sections of Vico del Fauno and Vico del Labirinto were limited in carrying traffic.  The former, colored in brown, was being repaved at the time of the eruption. [16]   No access was possible from Vico di Mercurio. Conversely, the latter –in green- is newly paved, thus not rutted, and blocked to traffic in the south by the large stone already (Figure 6). In the north, Vico del Labirinto was narrowed by a fountain on the east curb at Vico di Mercurio. Access was still possible for a small cart. Therefore, Vico del Fauno allowed no traffic while Vico del Labirinto could have supported only alternating one-way traffic [17] .


Where Vico di Mercurio meets its wider namesake (Figure 20 and Figure 21), evidence is in greater abundance and bolsters the conclusions already derived. The westbound ruts dramatically unite the sections of Vico di Mercurio in a lazy ‘S’ pattern. Parallel curved ruts demonstrate a right turn from the east side of the intersection, while a single southbound rut shows carts heading for the west curb. On the opposite, or West Side, there is also weak evidence for a left turn from the east side of Via di Mercurio. This left turn is supported by complimentary wearing on the western stepping stone, as is a right turn from southbound traffic on the West Side of Via di Mercurio. Therefore, from the East Side of the intersection, traffic on Vico di Mercurio continued as a westerly one-way, and turned both north and south. On Via di Mercurio, traffic moved in both directions and diverted its flow to the west from both southbound and northbound directions (Figure 22).


At the intersection of Vico di Mercurio and Vico dei Vettii (Figure 23 and Figure 24) the ruts connect westbound across the junction indicating it as a one-way across the entire region. Additionally, this intersection demonstrates that evidence can be ordered diachronically. Figure 25 shows an older rut running beneath a later stepping stone which blocked its path indicating an intentional change in the flow of traffic and northbound carts may have been bared. On this evidence [18] I offer the tentative designation of Vico dei Vettii as a one-way southbound street.  


Armed with the evidence from a cursory explanation of these crucial loci in Regio VI, it is now possible to build a street map of traffic direction. The evidence at Via Consolare and Vico di Mercurio proved north and southbound traffic on the former and strictly westward flow on the latter (Figure 14). Topographic evaluation defined many streets as alternating one-way streets (Figure 19). At the two Mercurios the Vico stayed strictly westbound while the Via provided both north and south directions (Figure 21). Moreover, Vico di Mercurio was proven to be westbound all the way across the Regio (Figure 25). Vico dei Vettii was tentatively attributed as a strictly southbound street (Figure 23). Via del Vesuvio, the eastern boundary of Regio VI is two-way north and south and (Figure 26) intersects with Via della Fortuna / Terme. This final street is two-way east - west and connects to Via Consolare (Figure 27). Wearing against north side stones (Figure 15 and Figure 17) locates the westbound traffic on the right. The direction map is now complete.







[1] Tsujimura, Sumiyo “Ruts in Pompeii – the traffic system in the Roman city” in Opuscula Pompeiana (Kyoto, 1991) p.

[2] Innumerable thanks must be given to the Directors of the Anglo-American Project in Pompeii (Ms. Jarrett Lobell, Dr. Rick Jones, and Dr. Damian Robinson) and to Prof. Pier Giovanni Guzzo and the superintendency of Pompeii who allowed such notice to taken at all.

[3] Such a system of traffic regulation would, of course, have been citywide. However, the scope of the project has only allowed for Regio VI to be studied.

[4] Unbroken ruts are evidence for the action of traffic. Simultaneously, ruts represent the volume of traffic over time by depth and make connections over space from a single action. In other words, in a diachronic sense one can infer the amount and duration of traffic from the amount rutting. Likewise, by dividing that aggregated traffic into a single, synchronic event, one can see that different places connected by the ruts are cemented in chronological unity.

[5] The behavior of driving a cart from one place to another leaves its mark as a rut on the paving stones, making the entire street a single chronological unit. When ruts adjust for datable obstacles in the road, such as fountains, they provide a post quem date for the event of paving. Thus the paving of the entire street can be dated and provide a benchmark to those quoining and abutting that street.

[6] One may deduce four categories of possible travel based upon the width. Wide streets could be two-way, or two-lane one-way; narrow streets could be strictly one-way or they could alternating one-way (that is who ever was first on the street set the direction).  Tsujimura, p. 62 sets the minimum width of a two way street at 3.21 meters.

[7] In the absence of signage, cart drivers (who seem to have been a specialized group) must have operated at least from a shared concept of acceptable behavior.

[8] The remnant of the rutting is less apparent on the north side due to the relatively tighter turn on the inside compared to the outside of the turn.

[9] That is not say there is no wearing on the east face of the last curbstone on the east side of Via Consolare before the intersection with Vico di Mercurio. Indeed there is a slight diagonal wearing on the end of the same curbstone which suggests that traffic did cross the intersection and turn to the south along the west curb of Via Consolare. The wide arc of the turn explains the faintness of the wearing and the general absence of ruts.

[10] Therefore, Via Consolare was not a two-lane one-way northbound street.

[11] In fairness to the British,  there is evidence that in ancient Britannia the Romans drove on the left. See Walters, Byrn “Swindon: Blunsdon Ridge”  Current Archaeology #163. P.258.

[12] This stone is located at VI.XIV.14

[13] Like the wear on the larger curbstone, the contact of the wheel rim grinding downward – but over a shorter distance – causes a diagnostic pattern from the upper right to lower left. Furthermore, as the depth of wear increases across the face of the stone, the associated rut gets visibly closer to the curb.

[14] There may have been an open area in which to turn around at the end of these streets. It is possible that an unpaved road which ran between Vico di Narciso and Vico del Labirinto existed. However, the complete absence of doorways suggests it did not exist, or was of no consequence in the system of traffic.

[15] These streets are: Vico di Narciso, Vico di Modesto, Vico della Fullonica, Vico del Fauno, and Vico del Labirinto.

[16] There is a precipitous drop at the intersection with Vico di Mercurio from the absence of paving stones. In the south traffic was likely preempted by the presence of road construction.

[17] There is no evidence at all for the use of this street by carts, except for the expectation that  the 5 front doors (all into Insula 14) along its length would have needed or desired vehicular service.

[18] Additionally, the entrances to the street at the distal intersections are narrow. Moreover, the narrowing of the street by the positioning of stepping stones and fountain supports this conclusion.




Copyright 2006


Copyright 2006