Dr. Archie Dunn
Thisve-Kastorion: Town, Territorium and Loci of Maritime Traffic (report on fieldwork conducted in 2007)

The third season of the archaeological survey of Ancient Thisve and Byzantino-Frankish Kastorion, at modern Thisve in the Koinotita of Domvraina, lasted for 4 weeks, from August 15th to September 11th. It has become a collaboration between the British School (represented by Birmingham University) and the new 23rd Ephoreia of Byzantine Antiquities, directed by Dr E.Gerousi, in succession to the 1st E.B.A. But we continue to work with the 9th Ephoreia of Prehistoric and Classical Antiquities (directed by Dr V.Aravantinos), a partner organization in our collaboration in 2005 and 2006, in that, by virtue of the necessity and value of recording all visible in situ archaeological features (excluding rock-cut depressions) at this multi-period urban site, we are producing maps of the Pre-Classical, Classical-to-Hellenistic, and “Greco-Roman” monumental features (for which see below). The team from Birmingham (Photeini Kondyli and Caroline Sturdy – geophysicists; Kevin Colls and Chris Mavromatis – topographers; Chris Hewitson – Three-Dimensional Laser-Scanner operator; Marilyn Cassedy – volunteer assistant) continued to assist Archie Dunn with the topographical, architectural, and wider archaeological survey, all of which were facilitated by the Ephoreia of Byzantine Antiquities (E.B.A.)’s cleaning programme.
The continuing aims of fieldwork are (1) the production by the British team of the first accurate and comprehensive record of visible in situ remains including key examples of post-medieval economic installations, which illustrate pre-industrial agricultural processing and storage; (2) the use of remote sensing (geophysics) wherever conditions allow, to explore further the ground plans and immediate built environments of visible fragments of monuments; (3) assisting the E.B.A. with the study of the site’s great exposed bio-archaeological feature: the surviving example of the “curious heaps of shells” of “purple shellfish” (i.e. of the murex purpurea range) reported by the British ambassador Sir Thomas Wyse in the mid 19th century, the role of which in the origin and rise of medieval Kastorion is a longer-term theme of the project. These activities are designed to complement the evidence of the American School of Classical Studies’ unpublished surveys of the Plain of Thisve, core of its ancient chora/territorium, and of Thisve-Kastorion’s loci of maritime traffic. Integration of the three surveys, and the interpretation of the survey of the plain, are underpinned by a parallel palaeo-environmental survey organised in collaboration with members of the Greek Centre for Marine Research. Integrating the study of the urban site, of its clearly defined hinterland (the plain being a polje), of its several sites of extra-regional contact, and of the landscape-archaeological matrix, is designed to support a dialogue, from the perspective of Byzantine Studies, with other multi-period surveys’ approaches to the medieval millennium, both within and beyond Boeotia. The urban survey itself is also intended to inform the design, by the Ephoreias, of an urgently needed programme of conservation at the urban site. The 23rd E.B.A. began this season to investigate the definition of conservation areas within the village which overlies most of the site.
The British team’s archaeological survey continued to have several aspects:

(A) TOPOGRAPHICAL SURVEY
(1) interlinking all in situ features and relating them to modern topography, (2) connecting these finally to the Greek Army’s nearest trigonometric points, (3) completely replacing the Greek military 1:5000 series’ mapping of the Lower Acropolis (“Neokastro”), (integrating the geophysical surveys’ grids, (5) importing all readings into our scanned geo-referenced 1:5000 base map. Chris Mavromatis (Ph.D candidate, Birmingham University) was responsible for these activities, using an EDM, assisted by our volunteer.

(B) REGISTRATION OF SITES
Cleaning in 2006 in preparation for the Magnetometer survey in the extramural survey zone, and subsequent processing of the results of that survey, have led to the recognition of a spolia-built structure on an east-west orientation to the west of Locus 5, which becomes Locus 77. The surroundings of Locus 77 have been re-surveyed in 2007. The number 77, which was used in 2006, is here re-assigned.
Completing the search for visible traces of the line of the fortifications between the Upper and Lower Acropoles, and the process of recording all in situ remains around the Upper Acropolis (“Palaiokastro”) led to the discover of three more features:
Locus 78: a fragmentary socle of monumental coursed trapezoidal masonry extending outwards from the north side of the Upper Acropolis across the narrow valley that defines this side of it; apparently a “single skin” wall; Classical/Hellenistic. The exposed faces are more finished than those of Neokastro and the city wall. This feature will have to be incorporated into the site plan under the aegis of the E.B.A. in 2008.
Locus 79: several rectangular Greco-Roman spolia set on end on an east-west orientation on suitably even ground within Palaiokastro recall in key respects the foundations of Thisve-Kastorion’s Byzantino-Frankish churches; entered into the site plan.
Locus 80: a significant stretch of the western city wall (curtain wall and a tower front) was located north of the projecting tower discovered on the last day of fieldwork in 2006. It confirms the course taken by the western city wall between Neokastro and Palaiokastro and is in exactly the same masonry as the rest of the lower city’s walls; entered into the site plan.
Completion of inspection of the demarcated urban survey zone:
Locus 81: the Greco-Roman rock-cut necropolis which is situated just outside the line of the western city wall, on the southern slopes of Palaiokastro, was recorded topographically.

(C) ARCHITECTURAL SURVEY
This was effected, as in 2006, using a High-Definition Three-dimensional Laser-scanner, another EDM, and geophysical prospection, while photography, standardised descriptions of all elevations, and measurement of engaged but fully visible spolia, complete the record. For the more detailed survey of better-preserved features, and the production of ground plans, elevations, sections, top plans, volumetric survey, cut-aways, and models viewable from any angle, the scanner was used, but an EDM continued to be used to record ground plans within the total site plan, and, time allowing, selected elevations. Moreover Palaiokastro is so steep and in many areas so unstable, that a Laser Scanner could not be used there, so the first stage of recording of its archaeological features (ground plans) was done using an EDM.

Palaiokastro (Locus 23)
Palaiokastro’s multi-phase fortifications comprise a major object of study, so in this season, its minor in situ exposed features, intramural rock-cut depressions and platforms of various kinds, were not recorded. Each identified phase, or possible phase, of construction of the fortifications was recorded as a separate exercise in six topographic layers, from “Cyclopean” to “Byzantino-Frankish”. This replaces a crude, inaccurate, and incomplete, sketch plan published by F.Maier in 1958. Several phases of construction are apparent, but, as at another multi-phase monument (Agios Loukas), important details of the relationships between phases remain unclear. Nevertheless the ground plans, military architectural features, and physical scale, of the majority of phases of activity are now reasonably clear.
At two points a phase of construction characterised by massive coarsely hewn quarried pieces of limestone, some with the appearance of “stretchers” up to 90 cm in length, apparently define an earth embankment or “core” whose other face is buried within later phases. They are clearly distinct from, and not functioning as, modern agricultural terraces. This is provisionally called the “Cyclopean phase”, which was noted by prehistorians as a “Mycenean” phase.
Three Archaic-to-Classical and Classical-to-Hellenistic masonries are apparent on Palaiokastro itself (excluding new Locus 78) : (a) a roughly dressed polygonal masonry forming the front of an approximately rounded bastion; (b) a masonry of trapezoidal and rectangular blocks with smooth abutting facets and rough exposed facets; (c) a conventional ashlar masonry which is either immured within post-Roman work and is only visible in top plan or, where visible in elevation, has been lime mortar-bonded (although not certainly disassembled first). Ancient masonry (b) is stylistically identical to that of the Lower Acropolis and the intermediate city walls, and would almost certainly be the masonry of the sections of these walls, now removed, that stood on rock-cut platforms which ran up the slopes of Palaiokastro. In this most important phase of activity there is in fact no evidence of acropoles as such, but of a single enceinte, with projecting towers, which enclosed the heights known as Palaiokastro and Neokastro. Ancient masonry (c) survives on the west-facing side of Palaiokastro, but the detailed design of this phase of activity and its relationship to work produced in ancient masonry (b) are obscured by its encasing within post-Roman phases.
The next discernible phase of activity in the walls of Palaiokastro, as at many of the fortified urban sites of Greece, is Late Roman-to-Early Byzantine. A utilitarian opus incertum, combining undressed quarried limestone with small spolia and bonded with lime mortar, is used to create an enclosed citadel for which there is no earlier evidence. The western, relatively vulnerable, side is defended by a simple forewall without towers and an inner wall with rectangular towers which may be Classical or Hellenistic in origin, partly built in ancient masonry (c). Forewalls on vulnerable aspects are typical of 5th-to-6th-century fortresses in the Balkans. The southern side, which is poorly preserved, is also defended by a line of towers in the last stages of disintegration. The northern side is also poorly preserved. Fragments of the line of the enceinte (without preserved evidence of towers) are traceable, but its NE “return” is well-preserved: massive spolia-built lime mortar-bonded foundations, perhaps of a bastion which only projected on one axis. The short east side, the most accessible to stone robbers, is untraceable, but we hope that a volumetric survey of the site may reveal it in 2008.
Despite its enclosed design, this Late Roman-to-Early Byzantine phase of activity on Palaiokastro does not mean that the city walls were abandoned. But it is likely to be one of the phrouria (forts or fortresses) built, according to Procopius, by Justinian in Boeotia in the 550s at the sites of earthquake-damaged settlements.

One or two later phases of activity are apparent on the better-preserved western side of Palaiokastro: a lime mortar-bonded opus incertum which encases the Classical-to-Hellenistic ashlar phase as re-used in Late Antiquity; and at the SE corner of the citadel remains of a rectangular building characterised by engaged external piers (buttresses) in a diagnostic Middle Byzantine-to-Frankish masonry which overlies the Late Roman-to-Early Byzantine phase at this point.
On Palaiokastro standardised descriptions of elevations, architectural features, and types of construction, were completed, but measurements of elevations, volumetric survey in general, and relief modelling, could not be effected this season. The terrain is unsuitable for the 3-D Laser-scanner (and there was insufficient time to use an EDM for these activities). So all three activities will be carried out using a Differential GPS, which will of course be the most effective instrument.
The 3-D Laser-scanner was meanwhile employed at four monuments, sometimes in combination with geophysical survey and EDM survey. The Laser-scanner was used at

1. Locus 25, a multi-phase tower at the SE corner of Neokastro, to re-record the interior of the ground-level vaulted chamber, since the scans made here in 2006 produced no results due to a malfunction. The Byzantino-Frankish redesign and reconstruction of this Hellenistic tower is completely preserved in elevation. The tower was converted from being one entered from the wallwalk, but of unknown internal layout, to a freestanding tower with vaulted basement and fighting platform;

2. Locus 1: parts of this church of Middle Byzantine design and masonry were re-scanned to fill gaps detected in last year’s readings.

3. Locus 3 and Locus 74: after their cleaning by the E.B.A. scans were taken around and within Agios Loukas and its monumental northern annexe, bringing to ten the number required to capture visible features in their entirety. EDM survey of the ground plan was also completed after cleaning. An EDM was also used to record the elevation of the west wall of the naos, and imported architectural conventions and symbols were used to visualise rubble masonry, terra cotta, and plastered surfaces, and to distinguish them from spolia construction in the elevation. Entirely visible Greco-Roman architectural spolia were measured; standardised descriptions of elevations were supplemented; and a Resistivity survey (led by Photeini Kondyli, Ph.D candidate, Birmnigham University) was carried out all around the western and northern sides of the monument (the eastern and southern sides being under modern streets). However in the accessible areas indications of the building complex within which Agios Loukas can be presumed to have stood were unfortunately not detected. Agios Loukas is situated outside, but within a few meters of, the deducible course of the ancient city wall, and its site may be assumed not to have been narrowly confined by contemporary buildings. Cleaning of vegetation and of modern debris leaves fundamental questions about the origins, details of design per phase, and building history, of this multi-phase monument unanswerable for the present, but the essential first non-intrusive stage of study is almost complete.

4. Locus 24: the Hellenistic walls at Neokastro (the “Lower Acropolis”). These were surveyed (in ground plan and selectively in top plan) in 2006. In 2007 the 3-D Laser-scanner was used to record selected elevations and to make a volumetric survey of the upstanding sections. Maier published in 1958 only an idealised ground plan of the “original” design, which is neither accurate nor complete. Our aim has been to produce an accurate ground plan, and, to illustrate the mode of construction, accurate examples of top plans and external elevations.

Photeini Kondyli’s Resistivity survey was carried out within and around three other monuments: within and to the west of Agia Triada (Locus 4), the remains of a church of Middle Byzantine type which, like Agios Loukas, may be deduced to have stood very close to the line of the ancient city wall (although almost certainly within it); and on all sides of two contiguous monumental structures (Loci 5 and 77), 5 being Late Roman-to-Early Byzantine in terms of masonry, 77 being medieval in terms of masonry, and both being the remains of churches. The reports on these surveys are not complete. Preliminary results however at Agia Triada seem uninformative. There is probably too much rubble immediately below the modern earthen surface to enable buried architectural features of the church (in particular: central piers, and the north and west external walls) to be distinguishable. Around Loci 5 and 77 it is anticipated that the combination of the Magnetometer survey’s results of 2006 with the new Resistivity survey’s results, for which the orientation of transects was shifted through 90 degrees, will be instructive. In addition, under the aegis of the E.B.A., sherds were counted in alternate 1-meter squares of the total grid, to examine the correlation between the intensity of readings and sherd densities. The area of the Resistivity survey was bought by the state and is scheduled for the construction of a school.
Our collaboration’s third current aim, the study of samples from the preserved murex purpurea-processing site (Locus 60) has not progressed. Responsibility hopefully can be re-assigned to the Greek Centre for Marine Research, with which we have collaborated to carry out the parallel palaeoenvironmental survey (which will be resumed in 2008-2009).
Meanwhile the processing of the records and finds of Professor Gregory (Ohio State University)’s surveys of the contiguous plain and loci of maritime traffic is being taken forward by Gregory and by Professor Bill Caraher (N.Dakota University), and will be completed during 2008, at the same time as the Anglo-Greek colaboration’s projected Study Season and complementary activities. This will pave the way for the integration of the results of these two complementary surveys.

Fieldwork Planned for 2008
In collaboration with the 23rd EBA (represented by Dr E.Gerousi, Dr N.Kontogiannis, and Dr E.Daphi), with our American colleagues, with members of the Birmingham University Field Archaeology Unit and of the Centre for Byzantine, Ottoman, and Modern Greek Studies, Birmingham, with Dr Joanita Vroom (Sheffield University), and with Dr Margaret Atherden (University of York, St John), we will carry out a high-resolution GPS survey of the urban site both to create the micro-contoured matrix within which to set all planned and/or mapped features, and to complete the volumetric survey of features on the Upper Acropolis; complete photography and standardised descriptions of all features; revise the EDM survey if necessary and insert the positions of all post-medieval monumental features (about which we have been selective so far); record the multi-phase dam which traverses the Thisve Basin (reported by Strabo and Pausanias); complete the vegetational survey of the Basin; re-study the harbourside sites (on-shore and off-shore); use high-resolution GPS to better integrate the mapping of the rural and urban surveys; re-analyse the pottery from the rural survey; and assess the next phase of the palaeoenvironmental survey.

Fig.1: Archaeological topography of Ancient Thisve/Byzantino-Frankish Kastorion (the urban survey)