A Diachronic Perspective on Directionality: Detour and the Pattern of Pompeian Traffic
At the time of the eruption in AD 79, wheeled vehicles entering and leaving Pompeii (Figure 1 and Figure 2)by the Herculaneum Gate faced an increased difficulty in accessing the rest of the ancient city. These cart drivers were already governed by social institutions that limited and informed their decisions for navigating the urban streets. At very least, these institutions operated as a general consensus among cart drivers for driving on the right-hand side of the street. At most, an assertion yet unproven but becoming more likely, a unified system of one and two-way streets directed the activity of driving in Pompeii. Now, these same cart drivers faced a new, physical impediment: a detour.
This paper addresses the patterns of traffic on specific streets in the northwest of Pompeii from a diachronic perspective. By combining the evidence for the repaving of the streets with the wearing patterns recorded on those streets it is possible to generate a relative sequence of construction events. This sequence reveals which streets were closed to vehicles and at what times, which in turn determines the possible routes of traffic. Moreover, with the ‘fairly’ secure ante quem date of AD 79, it is also possible to make a more educated guess about how long it took to cause those famous wheel ruts of Pompeii.
Ruts are, however, only one form of evidence for traffic. The wearing patterns on curbstones, narrowing stones, and stepping-stones locate not only the presence of traffic but also may define its direction. These directionally diagnostic wear patterns– such as this parabolic curve (Figure 3), or these over-ridding marks (Figure 4) – are found in complimentary positions across the ancient city. It is tempting for this reason to argue for a citywide institution governing the behavior that produced these wearing patterns and caused them to be in their specific locations. However, these evidentiary forms have no necessary, logical connection to one another; the cause of their existence could be as different as the reason why two coins of Nero can found at the same level in two different trenches. Yet if the evidence is silent on why the patterns came into existence, these forms of evidence still enlighten us as to when the events happened. This is because, like the coins, wearing patterns have a necessary relationship to the context with which they are associated. 1
For coins, this context is likely a soil deposit, for traffic wearing patterns, this context is the paving stone, curbstone, or stepping-stone onto which they are inscribed.
Thus, the street is itself is a stratagraphic unit, its ruts are as cuts (Figure 5) and its paving junctures (Figure 6)and bound both the spatial and temporal extents of that unit. The similarity with traditional stratigraphy continues as the phases of construction are linked by the synchronic features that exist on or extend across them. 2For instance, the wearing on Vico del Farmacista (Figure 7) stops abruptly at its paving juncture with Via Consolare. This is positive evidence for the interruption of traffic. 3Conversely, while the paving juncture of Via Consolare and Via delle Terme proves that these streets were laid at different times, the ruts (Figure 8) which cross both paving events link the activity on those streets chronologically. Carts could only have made those ruts after both streets were repaved. This is positive evidence for the resumption of traffic. 4
Preceding the current, cambered street surfaces there was an earlier pavement made up of somewhat smaller pieces of lava stone. This older surface still remains on Via delle Terme between the intersections with Via Consolare and Vico di Farmacista (Figure 9). Additionally, the original curbstones associated with these oldest paving stones, made of Sarno stone, are still in place along the south curb. 5Conversely, in the resurfaced eastern part of the same street all curbstones are of lava stone. Although this area provides a glimpse of the older street conditions, its existence, when all the streets around it have been renovated, is puzzling.
That is until one considers the materials found in and around this area. Strewn on the southern section of Vico del Farmacista are loose paving stones (Figure 10 and Figure 11). 6
Moreover, two large paving stone ‘blanks’ are built into the 1978 construction in front of Casa Fabius Rufus (Figure 12 and Figure 13). These ‘blanks’ are large spheroid lava blocks, which slice nicely into the pyramidal or dome shaped paving stones (Figure 14 and Figure 15). Such materials indicate that this western most section of Via delle Terme was not repaved because it was being used as a staging area and work zone right up to the day of the eruption.
The first street to be resurfaced was the eastern section of Via delle Terme . Once the construction was underway the traffic pattern was altered considerably (Figure 16). Vehicles entering the city could reach the eastern environs only by making a eastbound left turn on to Vico di Mercurio – a street demonstrably westbound in later phases 7– or by traveling south down Vico del Farmacista to its intersection with Via Soprastanti. The directional wearing on Vico del Farmacista, which is from this phase, 8indicates northbound traffic, again opposite of the proposed direction. Thus, with no access to the city for southbound traffic, one must be consider that Vico di Mercurio and Vico del Farmacista were required to permit travel in two directions during this time or that only exiting traffic used the Herculaneum gate in this period.
The second phase of repaving was dramatically more disruptive to the movement of vehicles through the area(Figure 17). Once the eastern end of Via delle Terme had been finished, the integral Via Consolare was opened to construction and closed to traffic. Every street in area was affected by this construction event, effectively closing off the northwest of the city. The section of repaving extended from the sharp line juncture at Via delle Terme (Figure 18) in the south to a less distinct convergence just north of the fountain at Vico di Narciso and Via Consolare. This latterjuncture (Figure 19) can only be detected by the absence of rutting in the south section and its prevalence in the northern section up to the Herculaneum gate. It is unsurprising, therefore, that there is no wearing, directionally diagnostic or otherwise, during this phase since there was no traffic to create it.
After the Via Consolare resurfacing was finished, traffic once again rushed through the area (Figure 20). And although the southern section of Vico del Farmacista is the smallest and least important street in terms of volume of traffic, its repaving (once Via Consolare was reopened to traffic) caused distinctive and at first glance contradictory patterns of wearing. It is these patterns that illuminate the phases of construction and, moreover, it is this repaving of Vico del Farmacista that deflected the flow of traffic off its normal path.
This is because when Vico del Farmacista was open to vehicles it reduced the volume of traffic that had to negotiate the constricted southern area of Via Consolare (Figure 21). Remove the possibility for southbound traffic by the construction on Vico del Farmacista and all vehicles that entered Pompeii’s most important landward gate 9had to pass through a section so narrow that it could permit only alternating one-way traffic. Likewise, the Herculaneum gate was also the most important gate for exiting traffic, a fact which forced northbound carts to compete for this space. Rutting on both sides of this bottleneck bear this out: two lanes of ruts converge into one and then diverge after exiting the Consolare bottleneck (Figure 22 and Figure 23). Moreover, there is a diagnostically worn narrowing stone against the west curb in this restricted area that is associated with this latest phase 10of construction and demonstrates northbound traffic (Figure 24).
Still, it is the places where the evidence for traffic does not exist that puts a necessary chronological relationship on the individual paving events. In fact, such areas of negative evidence are also found in a diagnostic form and locations. These are triangle-shaped sections of paving stones without rutting within a paving event where other areas do preserve ruts. Instead of being smooth, the paving stones in these areas are rough-hewn and pocked (Figure 25). Moreover, this rough surface is similar to that seen on the unlaid paving stones (Figure 26) found on the south end of Vico del Farmacista and elsewhere.Such rutless triangles exist in two 11very instructional places (Figure 27): on Via Consolare at the intersection with Vico del Farmacista (Figure 28) and on Via delle Terme where it meets the construction zone(Figure 25). Because no ruts exist in these places, no 12vehicles passed through here, proving what the absence of rutting and position of loose paving stones on Vico del Farmacista suggests; Farmacista was the final thing to be resurfaced, and, that event took place after traffic on Via Consolare had resumed. As mentioned above, a rut on the northern section of Vico del Farmacista abruptly terminates at the juncture with Via Consolare. Necessarily, this northern portion of Farmacista must be older than Consolare and, as the rutless triangle makes evident, must not have been available to carry ‘through-traffic’.
The effect on the ground of this final phase, therefore, was to deflect a significant proportion of northbound traffic onto a detour. Instead of fighting ‘up-stream’ against southbound traffic in the Consolare bottleneck, many cart drivers trying to exit the city choose to turn up Vico di Modesto. From here they then turned left onto Vico di Mercurio, and finally turned right back onto Consolare. Wearing made by these redirected vehicles is both prominent and diagnostic. At the point of deflection, the corner of Via delle Terme and Via Consolare, the lava curbstones are eroded (Figure 29), including one which displays a diagnostic directional curve (Figure 30). Additionally, the ruts associated with this wearing pass over a gap in the paving that once held a narrowing stone. 13Rut depth lessens as many carts split off to try their luck on the more direct Via Consolare (Figure 31). Then, because the older paving on Vico di Modesto had pre-existing ruts, rut depth resumes north of the fountain.
The drivers that choose to turn up Vico di Modesto next had to make a left hand turn onto Vico di Mercurio(Figure 32). Unfortunately, the curbstones on both sides of Vico di Modesto at this intersection are missing, leaving only a trace curved rut in the center of a disturbed group of paving stones (Figure 33). Still, the volume of vehicles turning here can be inferred from the depth and consistency of wear at the end of Vico di Mercurio. This intersection with Via Consolare is the archetypical example of combination of evidential forms (Figure 34). Two sets of curved ruts are preserved in the intersection and the strong wearing that exists on the southwest edge of the stepping-stone (Figure 35) is complimented by the diagnostic parabola on the northwest curbstone (Figure 36). From this point the cart driver would have had to negotiate only other northbound traffic (Figure 37).
Such was the state of affairs when Vesuvius erupted in AD 79. It can only be assumed that the eastern section of Via delle Terme would eventually be resurfaced, but it might have still have been in use for the construction on Vico del Fauno (Figure 38 and Figure 39). It is now possible to see the urban streets of Pompeii from a diachronic perspective, as infrastructural elements as important and necessary as the fountains that date them. However, we are at a loss to say with specificity who decided resurfacing was necessary 14, who did the work, how it was done 15, or who paid for it. Still, the archeological evidence does cast light on one fact: how long it took to wear out a street.
Two historical events pin down the chronology: the Augustan age construction of Pompeii’s aqueduct, with its associated fountains, and the eruption of Vesuvius in AD 79. Since the fountains are not cut into the paving stones, they are at least contemporary, if not earlier, than the street surfaces. 16In fact, on Via dell’Abbondanza (Figure 40) ruts can be seen adjusting to avoid a fountain, necessitating that the paving stones came after the fountain. Therefore, these street surfaces were no more than a century old and probably a decade younger still.
However, several other factors must be equally considered. Most importantly, there is wearing on the fountain(Figure 41) and its guard stones at Consolare and Modesto, but there are no associated ruts on the Consolare surface. This indicates another paving event occurred between the arrival of the fountains and the eruption. If we split the difference, the current surface could only be about 40 to 50 years old. Now, it still must be explained why Vico di Modesto and the northern section of Vico del Farmacista (Figure 42), both demonstrably earlier than the Consolare pavement, are in so much better shape than the western end of Via delle Terme (Figure 43). While the fountain and this section of delle Terme have no stratagraphic relationship, the presence of three heavily eroded stepping stones set into this oldest pavement is strong evidence.
The existence of stepping-stones presupposes a nearly fluvial environment in the street, which can only be provided by the constant overflow of aqueduct fed fountains. Such an expectation of water is also echoed in the tall curbs 17, which canalize the cambered street, themselves punctuated by the drains of private dwellings (Figure 44)that spill out into the street. 18Thus, the fountains are the raison d’etre of the stepping-stones, making the western surface of Via delle Terme post aqueduct. At the same time, the unusual style of construction and its poor state of preservation suggests that this pavement is older than that found on other streets. For these reasons, it appears we must now divide our century by three.
It is my opinion that it not only possible to date the resurfacing events by their relationship to the fountains, but also to blame the need for such resurfacing upon them as well! The unstoppable flow of the aqueduct added an extra element, indeed a catalyst, to the erosional environment of the street. Like spit on a whetstone, the water honed the edges of the forming ruts. Moreover, compared to the flow inside the aqueduct, the gradient of Pompeian streets could generate a torrent, carrying abrasive elements down the paths of least resistance, the ruts (Figure 45). 19This image of the street environment also explains the significant iron staining associated with all forms of wearing (Figure 46). Can it be that the famous wheel ruts are only partially wheel made? 20
Still, by most accounts 21, the aqueduct – like much of Pompeii – went out of service after the earthquake of AD 62. So, the final 17 years had a reduced environment for erosion. However, the removal of debris and arrival of materials for new construction must have made up for this in the sheer volume of vehicles. These facts must be taken into account when we look back to the final state of Via Consolare. Since it was planned to have two stepping-stones at its southern end (Figure 47), 22it must have been finished before the watershed year of AD 62. Conversely, Via Consolare’s other relationships give it a likely post quem of the mid 40’s AD.
The picture presented by this evidence is one of a dynamic city acting and reacting to stimuli both internal and external. It is easy, in general, to imagine that Pompeii was once a living city, bustling with the activities of daily life: cargo being brought in, goods being transformed and consumed, and waste being taken out. The importance of the present study, however, is to give a sense of both specificity and familiarity to this generalized urban existence. As a street was paved, worn out, and repaved its condition created a new patterns of traffic in the ancient city. Each new pattern impacted the daily existence of a Pompeian beyond merely how they got into and out of the city. All one needs to do is reminisce on the disruptions that road construction and detours cause in our daily lives, especially if one lives or works on the road being repaired. And, while the archaeologist may not be impressed to learn that street construction caused detour, the specific, datable, and tangible effects this detour caused within the city that is the benchmark for the study of Roman urbanism is impressive.
- Because coins have to exist before they are deposited, they have a terminus post quem relationship to their context. Conversely, because wearing on curbstones or paving stones can only occur after those stones are set, wearing gives a terminus ante quem date to the material on which it is enscribed. ↩
- Thus, two walls that abut one another are temporally distinct in relation to each other, but are joined in a singular chronological opposition to a unified layer of plastering that covers both walls. ↩
- This is equivalent to a drain being cut by a line of curbstones, showing the interruption of water flow. ↩
- This is equivalent to the same drain cutting into and crossing two different soil deposits. ↩
- The north curb is disturbed and remade in opus incertum. ↩
- These may have been moving here, possibly from Via delle Terme, rather than existing where they were found. A more romantic notion is that these paving stones are found here because the carts that were bringing them to the construction are were caught here during the eruption and deposited their load when the carts decayed. ↩
- The curved ruts extend from the deeper straight ruts on Vico di Mercurio onto Via Consolare tying both streets by the driving activity. Moreover, the lava curbstones at the intersection, different from the tufa or Sarno curbstones on all sides, are worn by this same turning event suggesting that these lava curbstones are part of the Via Consolare repaving. ↩
- The directionally diagnostic wearing may have been possible for even earlier phases, but not likely later. In fact, the northbound wearing evidence may have been from repair carts taking the paving stones from the south to double the efforts of those that must have been arriving from the north and working south. ↩
- The Herculaneum gate is by far the largest gate in the city, at approximately 4.1m in width. ↩
- This pattern may be even older and the stone reveals at least the latest and very likely earlier phases. ↩
- There is another wearless triangle at the intersection of Vico Narciso and Via Consolare. Because Vico Narciso is a dead end it would carry very little traffic even though – unlike Via delle Terme or Vico del Farmacista – it was open to traffic at the same time as Via Consolare. ↩
- At very least, the number of carts that passed through the area (if any) was insufficient to leave any ruts. ↩
- As I argued in my 1999 unpublished Master’s Thesis, one reason why narrowing stones were set up was to cause carts to make a wider arc in turning. ↩
- It is well known that the Aediles were generally responsible for the maintenance of streets. See Robinson, O.F. 1992. Ancient Rome: City Planning and Administration. London: Routledge. pp. 62-72, 79-80. ↩
- Statius describes (with some panegyric zeal) the construction of Via Domitiana. See Statius, Silvae IV 40-55. ↩
- In fact, it is my belief that the street was only metaled because of the amount of water expected in the streets once the aqueduct was constructed. Indeed the streets are the drainage system of the city. See Jansen, Gemma C.M. 2000. “Systems for the Disposal of Waste and Excreta in Roman Cities. The Situation in Pompeii, Herculaneum, and Ostia”. In Sordes Urbis, edited by Xavier Dupre Raventos and Josep-Anton Remola, pp.37-50. Rome: L’Erma di Bretschneider. ↩
- Moreover, the orientation of the iron staining on the curbstones of Via Consolare (as well as newly raised and stained curbstones on Vico del Fauno) suggests the curbs were getting even higher. ↩
- Without a complete underground sewer system, Pompeians had to rely on lapilli filled soak-aways and sluice drains into the street. ↩
- The 5% grade of Via del Vesuvio / Via Stabiana is double that of the open channel Roman aqueduct. ↩
- Because the abutments of the polygonal paving stones often align with the axis of the street, they facilitate the formation of ruts. At other sites, especially in North Africa, the road builders use a reticulated paving pattern that places only rectangular paving stones at diagonal angles to the street axis, thereby leaving no line in which wheel rims to fall. ↩
- The excavations of the Anglo-American Project in Pompeii have shown a dramatic reduction and even removal of piped water features at this time. See: http://www.brad.ac.uk/acad/archsci/field_proj/anampomp/index.html ↩
- In fact, this would have been the only place to cross Via Consolare with a dry foot. ↩