Luke Madson and Jesse James, Session III 2022 (May 30 - June 24)

Excavation Summary:

This excavation took place from May 30 to June 17, 2022, during the third excavation session of the season. It continued work done in the same trench in the second session (May 2022) recorded by Kaia Brose and David Picker-Kille, for which see [prior field report]. Jesse James and Luke Madson worked as trench recorders under the supervision of Chris Pfaff (Director) and Manolis Papadakis (Assistant to the Associate Director). James Herbst (Architect) and Ioulia Tzonou (Associate Director) also offered guidance on our complex and sometimes bizarre stratigraphy and Michael Ierardi assisted with the identification of our coins. In our trench, Argyris Tsirikis was our Pickman (newly appointed) and worked hard in consultation and collaboration with Athanasios Notis (Foreman); Argyris was supported by our Shovelman Agamemnon Karbouniaris, our good-natured Barrowman Sotiris Raftopoulos, and our eagle-eyed Sieve Operator Ilias Soli (Hekuran Coli), who also picked occasionally.

Area Description:

The excavation area consisted of a rectangular trench and was designated NET 16C, that is north east of the Theater, Trench 16C. The coordinates for the area when initially opened were: 35.0 E to 41.0 E and 1387.5 N to 1385.0 N. At the time we began our rotation, the trench had three main architectural features: Wall 27 which runs slightly off N-S axis; Wall 28 running east from Wall 27 and partially embedded in the southern scarp; and Water Pipe 3, running N-S and dividing NET 16 C from NET 16 B. The excavations were carried out largely in relation to Wall 27; from May 30th to June 2nd we worked E of the wall, moving west of the wall from June 2nd to June 16th. On June 10th the architectural features of the Vaulted Drain began to emerge and on June 16th the Amphora Deposit and Branch Drain were excavated.

Phases:

Hellenistic:

We found no evidence of activities in this area during the Hellenistic period (but note that two Hellenistic Sikyonian coins were found in contexts that are dated by pottery to the 1st century CE).

Early Roman:

We uncovered two apparent drain structures that we have dated preliminarily to the 1st century BCE: a “Vaulted Drain” (Structure 96 in iDig) and a “Branch Drain” (Structure 97), the latter containing the amphora deposit below, with significant remains of approximately ten amphoras. The Vaulted Drain, and possibly the Branch Drain, was likely built shortly after foundation of the Roman colony in 44 BCE and its centuriation at approximately the same time. The Vaulted Drain aligns with a major N/S road of the Roman era uncovered in Trench 16 B (directly adjacent to the W), a road whose width may originally have extended into our trench, although we found no direct, independent evidence of it.

Vaulted Drain
The most significant structure excavated in Trench 16C during the June session was the Vaulted Drain (Structure 96), a stone structure running N-S at the west end of Trench 16C, below Water Pipe 3. As so far exposed in the trench, it consists of a wall, partial arched ceiling, and apparent floor layer. 1.74m of the drain’s N-S length has been excavated. The floor’s elevation is approximately 57.89masi (ca. 3.4m below current topsoil), and the arch’s highest interior point is ca. 1.25m above the floor. The wall of the drain (which is the western wall of the visible structure) consists of a main lower course of large, moderately worked rectangular poros stones (the largest stone is 0.79m tall and 1.22m wide) topped with a course of smaller worked poros stones (ca. 0.2m tall), possibly with mortar between them. At its top this wall curves into the arch of the ceiling consisting of smaller unworked stones (rounded, hard limestone and conglomerate, ca. 0.12–0.35m in length) and bonded into a vault structure with a rough pinkish mortar embedded with small pebbles (0.001 to 0.008m dia.). There is an apparent floor layer consisting partly of worked rectangular stones and partly of soil. The stones run beneath the bottom course of wall stones, indicating that they were set there deliberately, as part of the construction of the Vaulted Drain. Further investigation of these stones and what lies beneath them is needed. They are of different sizes and their top surfaces are now set at slightly oblique angles and elevations, making an irregular floor surface. The larger floor stone measures approximately 0.7m N-S by 0.8m E-W (visible exposed surface); the smaller stone, directly to the N, is 0.31 by 0.33m. The soil around the stones was not compacted to a hard surface, possibly indicating that other floor stones were previously in place and were subsequently removed. An alternative interpretation is that the “floor” stones are not a floor at all but served another function.

The existing arch of the ceiling of the Vaulted Drain appears to peak ca. 0.25m east of the wall surface. Assuming a symmetrical arch, and that we have the highest point of the arch (which seems correct), we would expect an eastern wall to the Vaulted Drain ca. 0.5m from the existing wall. But we found no direct evidence of such an eastern wall: no large worked rectangular stones (no worked stones at all), no other large stones that appeared part of the same structure as the western wall and arch, and no inclusions of mortar matching the mortar of the arch. This evidence suggests one of two possibilities: either the eastern half of the Vaulted Drain was fully disassembled in antiquity (see dating discussion below), removing all traces of the disassembly within the area of our trench; or the drain is significantly wider than appears to be indicated by the remains of the arched ceiling.

The outside top of the Vaulted Drain’s arched ceiling is covered with a reddish, perhaps clayey soil, and Water Pipe 3 sits ca. 0.4m above the top of that ceiling, also in reddish soil. During excavation, that 0.4m depth of soil appeared to be in two layers, which suggests that the pipe may have been laid some time after the Vaulted Drain was built. But the pipe appears to be centered directly over the Vaulted Drain, which could indicate that it was laid at the same time. If that is the case, it may be that this 0.4m of reddish soil was placed deliberately both as a kind of sealing layer over the Vaulted Drain and as a bedding layer for Water Pipe 3.

Branch Drain
Slightly to the east of the Vaulted Drain, at the north side of the trench, we discovered what we have called the “Branch Drain” (Structure 97) running roughly SE to NW. It becomes visible in its path from the E (at 36.95m in the easting in our trench, elevation 58.3masl) at a height of ca. 0.5m above the floor level of the Vaulted Drain. At this point the Branch Drain is 0.25m wide, and widens to ca. 0.4m by the time it exits the trench to the N (at 35.25m in the easting, elevation 57.95masl), with a total exposed length of 1.7m and drop of 0.25m. At its westernmost visible point the Branch Drain is less than 1.0m from the floor of the Vaulted Drain. Some stones still in situ between the SW curve of the Branch Drain and the Vaulted Drain may originally have been part of the Branch Drain walls.

We only partially excavated the Branch Drain, even within the boundaries of our trench. We excavated as far E as the west side of Wall 27 (which is a much later wall, for which see below, under “Byzantine”), and have not uncovered the north or south sides of the stones that form its walls. We have also not found a point at which the Branch Drain joins the Vaulted Drain, although we presume such a joint slightly north of the boundary of our trench. Hence the description and interpretation here are highly provisional.

The Branch Drain appears to consist of two walls or sides built of unworked, dry-stacked stones. At the moment of writing there appear to be three or four courses of stones in these walls, but more excavation is needed to confirm what remains of the entire structure. This Branch Drain also has a partial tile floor. At its eastern visible limit the drain is oriented mostly E-W, but curves more toward the northwest as it proceeds west toward the Vaulted Drain. The tile floor slopes down visibly, and the tiles give way to soil after two visible overlapping tile courses. The top tile shows 0.4m of visible length and 0.33m of visible width, and is 0.03m thick. The Branch Drain’s stone sides also slope downward. (Some of this slope of the sides of the drain could be a result of inadvertently removing structural stones during the excavation process. We attempted not to remove any stones that were parts of an obvious structure, but some large stones were excavated in this context and they might originally have been built into the drain sides).

Amphora Deposit
One of the major breathtaking finds from the trench occurred on the final day of excavation. A deposit of perhaps eleven amphoras and one mortar, mixed with non-amphora potsherds, some large stones (ca. 0.2 to 0.4m long), and soil, was found in and above the Branch Drain. This deposit, roughly in the shape of a steep wedge, was approximately 0.7m wide (N-S), 1.6m long (E-W), and 1.7m deep at its highest, easternmost point. We were not able to determine definitively whether this deposit was placed into a man-made trench, but its position directly above the Branch Drain indicates that the deposit was made into and onto that man-made drain (the soil within and around the amphoras and stones may have been deposited by natural erosion). All the amphoras appeared to have been lying on their sides (none deliberately placed upright). We did not collect elevation points on individual amphoras in situ, but the highest was found at approximately 59.35masl. All of the amphoras were broken, but the completeness of the remains suggests that none had been moved more than once or twice between the end of its functional life and its final deposition here, and that therefore their deposition was intentional and expected to be final. At least two amphoras (C-2022-7 and C-2022-20) appear to have been deposited as complete vessels, although parts of them (the mouth of C-2022-7 and one longitudinal half of C-2022-20) remain in situ in the trench scarp as of the time of writing. Preliminary analysis indicates that the amphoras date to between the 1st century BCE and the 1st century CE.

As study and cataloging of the amphora remains is ongoing, and additional ceramic material remains underneath Wall 27, this is a preliminary catalog of the approximately ten amphoras from this location:

C-2022-7 (Amphora 1); Dressel 2-4 with dipinto, 1st cent BC to 1st cent CE;
C-2022-8 (Amphora 2): Dressel 21-22 (resinous substance on interior), 1st cent. CE (cf. C-80-180)
C-2022-9 (Amphora 3): Spanish Dressel 2-4, 1st cent BC to 1st cent. CE (fragments originally labeled C11 [Amphora 5] and C16 [Amphora 9] have been determined to be part of C-22-09)
C-2022-10 (Amphora 4); Dressel 6A(?), 1st cent. BCE to 1st cent. CE
C11 (Amphora 5); see C-2022-09
C-2022-12 (Amphora 6): pompeii vii amphora, end of 1st cent. BCE to 1st cent. CE
C-2022-13 (Amphora 7): thin-walled sandy fabric amphora toe
C-2022-14 (Amphora 8); pompeii vii amphora
C-2022-15 (spouted mortar): late 1st cent. BC to early 1st cent CE (cf. C-2004-12)
C16 (Amphora 9); see C-2022-09
C-2022-17 (Amphora 10): Pompeii vii
C-2022-18 (Amphora 11): Pseudo-Coan
C19 (Amphora 12); see C-2022-15
C-2022-20 (Amphora 13 = C19 [fragments originally labeled C19 (Amphora 12) have been determined to be part of C-2022-20)

Water Pipe 3
Water Pipe 3 (Structure 95) runs N-S directly above the Vaulted Drain, centered at 34.645m east, with a top elevation of approximately 59.96m above sea level. Within Trench 16C the pipe slopes slightly from N to S, with a drop of 0.023m over a length of 1.585m (this is surprising because the general slope of the landscape here is gradual from S to N). Its diameter varies from 0.107m to 0.127m (the greater width is at the junctions) with approximately 3 segments (ca. 1.6m) currently exposed in situ. As noted above, the apparent continuity of red clayey soil from the top of the Vaulted Drain up to the bottom of Water Pipe 3, along with the similar orientation of the two structures and Water Pipe 3’s position approximately centered over the Vaulted Drain, suggest that Water Pipe 3 was laid down close in time to the construction of the Vaulted Drain. As explained in detail below, that was likely between 44 BCE and the mid-1st century CE.


Dating
The Vaulted Drain, Water Pipe 3, and the Branch Drain were likely constructed between 44 BCE and the middle of the 1st century CE. The founding of the Roman colony in 44 BCE provides the terminus post quem. Two separate sets of evidence provide the same terminus ante quem: the dating of the amphora deposit in the Branch Drain, and the pottery deposited around Water Pipe 3.

The manufacture and use of the amphoras and the mortar have been preliminarily dated to the period from the 1st century BCE to the 1st century CE. A single coin was found in the amphora deposit (Coin 2022-440) and dates to between 40 and 30 BCE, shortly after the founding of the colony. The non-amphora pottery within the amphora deposit dates to the second half of the 1st century CE. These dates provide a terminus post quem for when the Branch Drain went out of use (although not a precise date as the amphoras were likely used for a significant duration after their manufacture). It is unclear how the amphoras came to be in the Branch Drain (deliberate human action? Mudslide?); it is also unclear whether they were all placed there at one time or over a long period of time. Because no material in the amphora deposit dates after the end of the 1st century CE, we conclude that the deposit was in the Branch Drain by that time, and therefore that the Branch Drain was out of use by ca. 100 CE (although the Vaulted Drain may have continued to function). Because it was defunct by the end of the 1st century CE, the Branch Drain was likely built somewhat earlier, probably at or before the mid-1st century CE.

Pottery found in the soil directly surrounding Water Pipe 3 (in both Trench 16C, Context 68 and Trench 16B, Context 111), indicates that the pipe was laid in the 1st century CE.

As discussed above, stratigraphy indicates that the Vaulted Drain was built either before or at the same time as Water Pipe 3. And because the Branch Drain is apparently ancillary to the Vaulted Drain, it is likely that the Vaulted Drain was built before or at the same time as the Branch Drain (this hypothesis should be clarified in next year’s campaign, when the conjectured meeting point slightly to the north can be explored). These two structures therefore converge on a terminus ante quem for the Vaulted Drain of the mid-1st century CE.

On present evidence it is difficult to settle on a date for the Vaulted Drain more specific than between 44 BCE and ca. 50 CE. On one hand, the Vaulted Drain appears to be a major infrastructure and planning project aligned with the Roman road. This suggests that it was part of the original centuriation of the colony and therefore was built soon after 44 BCE. On the other hand, Water Pipe 3 may have been laid at the same time that the Vaulted Drain was built. But this points to a 1st century CE date, decades after the founding of the colony.


 
Late Roman (4th to 6th cent.):

Disassembly of the Vaulted Drain
The Vaulted Drain’s fragmentary state within trench 16C--no eastern wall and incomplete arched ceiling--and the lack of remains from the eastern wall and the eastern part of the arch, suggest that at some point the Vaulted Drain stopped being used and that part of it was deliberately removed, i.e., robbed out. The deep deposit of loose sandy fill within the Vaulted Drain (in particular from Contexts 84, 91, and 92) appears to date to the Late Roman period from both pottery and coins (e.g. Coin 2022-403 dates to 347–48 CE). The entire deposit was of similar texture and soil type, with no apparent stratigraphy between layers, suggesting that it was deposited within a short period of time. Because we failed to number the buckets of pottery as they were excavated from Context 84 (a fill whose height was 0.84m containing 34 kg of pottery), we cannot now distinguish between pottery from the top of 84 and that from the bottom, to determine if there is in fact any discernible chronological distinction between the top and bottom layers. Yet Contexts 91 and 92, both beneath Context 84 and just above the Vaulted Drain’s floor layer, also contain a mixture of early Roman and late Roman pottery and therefore indicate that they were part of the same deposit as Context 84. Hence it appears that the Vaulted Drain was partially disassembled, and filled, in the Late Roman period.

Bronze ring: One find of particular note was a bronze finger ring (MF 2022-39), located quite close to the Vaulted Drain’s floor in Context 91; while this ring has yet to be conserved, its basic form is that of Type 1A finger ring with a setting (Davidson 1952: 228). It is most similar to Davidson No. 1819 (Plate 102/MF 7176; cf. Davidson No. 1818). As this ring form seems to be common in Hellenistic, Roman, and Byzantine contexts (see Davidson 1952: 232, indicating a gap from the 5th to 10th centuries CE), it is consistent with a late Roman dating for the deposit.

Domitian coin: Another noteworthy find from the fill in the Vaulted Drain was Coin 395, a bronze assarion of Domitian, minted in Corinth between 85 and 87 CE. On the reverse is depicted a tetrastyle temple on Acrocorinth, seen in perspective from the left corner. This coin, with the obverse head of Domitian facing right, is an example of a hitherto unknown die combination (all published examples with this reverse pair it with an obverse head facing left). This coin, found in Context 84, is chronologically consistent with the wide chronological variety of the pottery found in that deposit.

Road
Unlike in Context trench 16B, immediately adjacent to the west, we found no definite road surfaces dating as early as the Late Roman period. To the west of Wall 27 the lowest clear road surface was Context 71 or 72, both of which still contained Byzantine green glazed monochrome and other Byzantine pottery dating to the 12th century. Similar layers of road may lay east of Wall 27 but remain to be excavated.

The disassembly of the Vaulted Drain and the lack of Late Roman road surfaces suggest that the area between Water Pipe 3 and Wall 27 was not used as a road surface during the Late Roman period. There may have been a Late Roman road which was either intentionally removed or naturally washed out. Perhaps a flash flood or a partial collapse of the Vaulted Drain caused the east side of the Roman Road to subside in the Late Roman period. Then inhabitants may have taken the opportunity to partly disassemble the Vaulted Drain, fill it in, and then use the disturbed area as something other than a road.

Byzantine:

We have found no features or objects datable to the roughly 600 years from Late Roman to the 12th century CE. 12th century features include Walls 27 and 28, apparent road surfaces to the W of Wall 27, and levels of fill to the E of Wall 27.

Road
In the Byzantine Period a road ran N to S along the western side of the trench, to the W of Wall 27. The width of this road may continue East of Wall 27 but this remains to be seen. While the earliest apparent layer or fill of the road contained some 6th Century CE Late Roman pottery, no layer appears to date earlier than the 12th cent. CE. The manner of road construction appears informal: rather than any sort of paved surfaces, the stratigraphy revealed a series of hard packed earth surfaces with occasional inclusions and/or potholes, though since there was no formal construction we cannot say how many road layers there were or the thickness of a given layer. The precise width of the road is similarly not yet secure. At some point in the 12th Century, a cut was made for a foundation trench for Wall 27 (below).

Wall 27
Wall 27 appears to be continuous with a wall segment in Trench 18C to the south (although the two segments may not be in perfect alignment with each other). Its construction dates to the 12th Century CE based on pottery in the foundation trench (context no. 48). This wall (length 2.25 m running the entire width of the trench; width varies from .60 to .66 m; height 1.09 m at maximum surviving height) is characterized by 2 large worked blocks (block (1) width .71 m; height .44 meters; thickness .36 m; block (2) width .62 m; height .70 m; width .32 m; apparent spolia from another structure) in the east face which sit on 2 or 3 courses of at least partly worked stone blocks. There are a few worked smaller square blocks in the wall placed irregularly. The west face is mostly made of irregular unworked cobbles and larger stones 6 or 7 courses high. The fill that went up and over the remaining portion of Wall 27 also dates to the Byzantine Period (Pottery NPD), suggesting the wall went out of use later in the Byzantine or Post Byzantine Periods. Notably, in the fill directly beneath the lowest course of stones on the west side of Wall 27, an intact Roman unguentarium was found (C 2022 6); when dated, this object will establish a terminus post quem for the construction of the foundation trench and wall. Wall 27 provides a 12th century eastern limit for the width of the road after it was constructed. Whether the wall was cut into the middle of an existing Byzantine road, or built against the side of that road remains to be seen.

Wall 28
Wall 28 runs along the south edge of the eastern side of the trench (length 0.41 m; exposed thickness 0.35 m in W to 0.23 m in E; height 0.15 - 0.20 m). The stones and tile on top are 0.10 to 0.15 m in width forming a sort of capping cours. Wall 28 is an enigma and remains only partially exposed as the scarp encloses the south-facing side. There does not appear to be a foundation trench on the north facing side but the soil level that is at the level abutting the wall and that was in use with the wall with the first course of stones dates to the 12 century CE (pottery). The date of this fill and the construction of the wall probably date to the 12th or later. Wall 28 appears later than Wall 27 based on two features: (i) its base is at a higher elevation than the base of Wall 28 (suggesting deposition of earth after the construction of Wall 27 and before that of Wall 28), and (ii) Wall 28 butts to the E face of Wall 27 rather than being bonded or integrated into the larger wall’s stonework. Wall 28, as it remains now, consists of three or four courses of irregular cobbles and small flat stones and tile pieces. As with Wall 27, the fill which covered the wall dates to the Byzantine Period (Pottery NPD), suggesting this wall went out of use later on in Byzantine or Post Byzantine Period.

Suggestions For Future Excavators:

A great deal of further excavation is needed to clarify this trench. A parallel trench to our south might be opened to better account for the south-facing side of Wall 28 and any possible foundation trench. Such southern exploration would clarify the purpose of Wall 28 as it relates to Wall 27 and whether it functions as an internal wall to a larger structure. Similarly, such investigation would continue to clarify the courses of Wall 27, Water Pipe 3, and the Vaulted Drain, and the course of the Branch Drain as it lies in relation to its possible source to the (south?) east. A similar parallel trench might be opened to the north as well, in order to better clarify the relationship between the Branch Drain and Vaulted Drain which may join just beyond our north scarp. Digging both north and south would also clarify the disassembly of the Vaulted Drain, and whether this occurred throughout the Vaulted Drain structure. The Amphora Deposit may continue to the NE as evidenced by the remains of Amphora 1 still in the scarp to the north and Amphora 13 underneath Wall 27 to the east. Additional amphora finds from next year’s campaign will need to be collected in relation to the amphoras we excavated to complete their conservation and show us their level of preservation. The road layers and deposits underneath, like the Amphora Deposit, may be better clarified with the removal of Wall 27.