The Ancient Maritime Dynamics (AMD) Project

• Abr-Nahrain
• Ägypten und Levante
• American Journal of Archaeology
• Anadolu Akdenizi Arkeoloji Haberleri / News of Archaeology from Anatolia’s Mediterranean Areas 
• Anatolian Studies
• Anatolica
• Ancient Near Eastern Studies
• Annual of the American School of Oriental Research
• Annual of the British School at Athens
• Annual Report of the Department of Antiquities Cyprus 
• Araştırma Sonuçları Toplantısı
• Archaeologia Cypria
• Archaeology and History in the Lebanon
• ‘Atiqot 
• Berytus
• Bulletin d’Archéologie et d’Architecture Libanaises
• Bulletin d’Études Orientales
• Bulletin of the American Schools of Oriental Research
• Bulletin of the Anglo-Israel Archaeological Society
• Centre d’Études Chyproites
• Egyptian Archaeology
• ENALIA 

• Hadashot Arkeologiyot, Excavations and Surveys in Israel 
• Hesperia
• Israel Exploration Journal
• Journal of Egyptian Archaeology
• Journal of Field Archaeology
• Journal of Near Eastern Studies
• Journal of Palestinian Archaeology
• Journal of Roman Archaeology 
• Journal of the American Research Center in Egypt
• Les Annales Archéologiques Arabes Syriennes
• Levant 
• Mediterranean Archaeology, Australian and New Zealand Journal for the Archaeology of the Mediterranean World
• Opuscula Atheniensia
• Orient, Reports of the Society for Near Eastern Studies in Japan
• Palestine Exploration Quarterly
• Progress in Underwater Science 
• Report of the Department of Antiquities Cyprus 
• Studies in Ancient Art and Civilization
• Syria
• Archéologie, Art et Histoire
• Tel Aviv, Journal of the Institute of Archaeology of Tel Aviv University 
• The International Journal of Nautical Archaeology 

• Ancient Akamas I. Settlement and Environment
• Journal of Roman Archaeology, Supplementary Series
• Caesarea Papers. Straton’s Tower, Herod’s Harbour, and Roman and Byzantine
• Egypt’s Sunken Treasures
• Istanbul: 8000 Years Brought to Daylight
• Le navi antiche di Pisa
• Ancient Shipwrecks of the Adriatic

• Proceedings of the International Symposium Cyprus and the Sea
• Res Maritimae, Cyprus and the Eastern Mediterranean from Prehistory to Late Antiquity
• Stories from the Hidden Harbor, Shipwrecks of Yenikapı
• The Ancient Harbour and Anchorage at Dor, Israel.
• The ‘Old Ships’ of the ‘New Gate’
• The Roman Port and Fishery of Cosa
• Transport Amphorae and Trade in the Eastern Mediterranean

Shapefiles with associated spreadsheets: Assemblages, Sources, and Polygons.

• Shapefile of Assemblages:  The AMD Project starts by plotting the locations of assemblages on the seafloor as a series of points arrayed across the Mediterranean Sea.  To avoid misrepresentations, not all of these collections on the seafloor are necessarily shipwrecks, nor are they labeled as such.  Instead, in this study each is labeled as an assemblage – this could be a large collection of material or a shipwreck, or a single item on the sea floor.  Site was avoided as, in some reports, a site is composed of numerous assemblages and single items, but not vice versa.  The assemblages’ points are stored in a spreadsheet in ArcGIS with fields recording their contents, name, approximate date, and bibliographic reference, among other data.  With these fields, it is possible to segregate between assemblages drawn from Parker’s dataset and those published elsewhere after 1990, or contrast assemblages with material from only one source to those with material from multiple sources.
• Shapefile of Sources:  Each of the sites in the Site Dataset has material originating from one or more locations in the Mediterranean region.  Some of these sources are indicated geographically, and could be specific, such as Corinth, slightly broader, like Campania, or particularly broad spatially, such as Gaul or Egypt.  Other sources may be roughly extrapolated from a typological style, such as Dressel 1B, Late Roman Type III, or African Red Slip ware.  Geographically-specific sources are ideal, but are not always available.  When possible a single dot is plotted at the source of the material.  When not possible, more dots are placed in the region representing the general source of the item.  Three dots are arrayed between Libya and Tunisia, for example, to represent the general area where North African type amphorae are produced.  Four dots are dispersed along the Adriatic coastline of Italy to represent the general source of Dressel 6 amphorae.  This information is under continual refinement, however, and if you would like to improve the resolution of our models, please feel free to contact us with more information.
• Shapefile of Mobility Polygons:  Once the assemblages’ locations on the seabed have been plotted, and the sources’ locations on land have been established as well, then the mobility polygon representing the likely area of the assemblage’s movement can be generated.  The source or sources of each item in the assemblage became vertices of the polygon, and the location of the assemblage on the seafloor becomes the final point along the polygon’s edge.

Color Coding

Assigning colors to the projected polygons, or to the subsequent rasters, is done both numerically and visually.  Among the fields within the Assemblages Shapefile should be four to record the color value for four of the five basins in the Mediterranean: the western basin, the Adriatic, the Aegean, and the eastern Mediterranean.  As mentioned in the Background section, these regions had particular colors: red, black, blue, and green. The Ionian Sea was not included because it has the lowest number of assemblages, and none of these assemblages carry items from the Ionian Sea itself.  The Ionian Sea, it seems, is a place of transshipment.

The numerical values for these fields are determined with the following variables:

• N = Total number of sources for the assemblage
• Xw = Total number western-Mediterranean sources for the assemblage
• Xad = Total number of Adriatic sources for the assemblage
• Xag = Total number of Aegean sources for the assemblage
• Xe = Total number of eastern-Mediterranean sources for the assemblage

And the constant 256, which is the highest color value in the RGB color scale within ArcGIS.

To calculate the color value for an assemblage with items from three sources in the western Mediterranean, use this equation:

(256/N) x Xw = Red color value in RGB color scale

(256/3) x 3 = 256

In the RGB color scale in ArcGIS, the red value should be 256, and green and blue should be zero.  This color is then applied to the appropriate polygon.

Similarly, the color value for an assemblage with items from 7 sources in the Aegean is determined:

(256/N) x Xag = Blue color value in the RGB color scale

(256/7) x 7 = 256 blue

In the RGB color scale, the blue value should be 256, and the red and green should be zero.  This color is then applied to the appropriate polygon.

The color value for an assemblage with items from five sources in total, but two sources are in the Aegean and three sources are in the eastern Mediterranean:

(256/5) x 2 = 102 blue

(256/5) x 3 = 153 green

In the RGB color scale, the blue value should be 102 and the green should be 153; red should be zero.  This color is then applied to the appropriate polygon.

Once the mobility polygons representing the most likely area of movement for each assemblage have been plotted, and the colors for each polygon have been determined, it is possible to conduct a variety of geo-spatial analyses and project the results.  The results can be generated in two forms: projections of superimposed and color-coded polygons, or rasters displaying varying densities of activity.  For a detailed discussion of these projections and rasters, please click here.

With the completion of the data collection in 2016 supported by the Honor Frost Foundation, it meant that the dataset of assemblages in the eastern Mediterranean basin was now substantially larger than that presented in Parker’s 1992 catalogue, and more up-to-date as well; it contained the results of new analytical investigations and refinements of ceramic typologies.  With this larger corpus of information, it also meant that the projection and modeling of patterns of activity in the eastern basin could be more meaningful statistically.  The first results using this expanded dataset were presented at the 2016 Connecting Cultures Conference in Lefkoşa, Cyprus, organized by Vasıf Şahoğlu, Müge Şevketoğlu, and Yiğit Erbil.  At this event, a straightforward hypothesis was investigated: If ancient writers’ maritime geographies emerged from the everyday use and valorization of maritime space, then the places recognized in these texts should coordinate spatially with concentrations of maritime activity emerging from these archaeological models.

What emerged through this testing was that some writers’ portrayals of maritime space, such as that by Pomponius Mela in the 1st century AD, oriented poorly with the archaeological data (Figure 1).  Other writers’ geographies, however,  coordinated more closely.  Pliny the Elder’s Lycian, Pamphylian, Cilician, and Phoenician Seas  overlap well with the concentrations of activity in the 1st century AD projected by the archaeological models (Figure 2), as does the geography of Saint Orosius from the 5th century AD (Figure 3).

The parallels that emerge from this comparison are promising, and suggest that these two independent sets of data are portraying similar phenomena.  Where there appears to be a greater density of maritime activity – a social phenomenon that creates a place from space, as argued by Lee and Ingold (2006), Ur (2009), Leary (2014), and Lucas (2014) – textual sources have a toponym for the place in question.  These maritime places, as loci of activity and elements of a formalized geography of the sea, thus have a value that transcends the social boundaries between sailors and the literate historians and clergy recording the world.  Nevertheless, some issues are still open to questions.  How much of this coordination is a coincidence, and how much is the result of the transfer of knowledge?  For a former naval commander like Pliny the Elder, with aides compiling data for his scholarship, it seems more likely that his geography represents the latter.  But, for Pomponius Mela, do the discrepancies represent a failing on his part to represent or record these maritime places accurately?  Or, are his models instead presenting one cartography of a space shared and perceived in many different ways?

Whereas the previous studies investigated constructions and use of maritime space, the workshop organized by Luca Zavagno, “Islands at the Frontier of the Empires in the Middle Ages” at Bilkent University, Ankara, in 2018 investigated edges.  Where are the boundaries of empires and communities?  How are they represented, how are they perceived, and how might they be identified today?  Within the AMD Project, these questions are particularly pertinent: Evident in models made as early as 2013 are gradients or frontiers between concentrations of activity.  One appears in the south eastern corner of the Aegean Sea, and a second appears just to the east of Sicily.  What is unclear, however, is the resiliency of these gradients – how long do they last?  Or, how dependent are they on geography or human activity?

Using a dataset of 449 assemblages containing only material from the western Mediterranean, polygons for each assemblage were projected then segregated on a centennial basis from the first to the seventh centuries AD.  Importantly, these assemblages were chosen because they each contain items only from the western Mediterranean basin, they are not assemblages only found within the western basin.  This means that assemblages throughout the Mediterranean Sea were used, although each only contained items from the western portion of the sea.  Variations in the density of the polygons were used to create a series of rasters illustrating concentrations of activity in each century.

The results of this analysis suggest that between the first and fifth centuries AD, maritime activity was much more likely to constrain the movement of materials originating within the western basin to that space, than to move the items farther eastwards (Figures 1 and 2).  For these centuries, a boundary or gradient appears to distinguish waters to the east and south of Sicily from the waters to the west.  As there is no environmental or technological impediment to movement farther eastwards, however, this gradient is much more likely a social construct.  Moreover, while the gradient’s longevity suggests that it was tied to the presence of the island, the gradient’s segregation of western maritime activity also suggests the presence of a distinct maritime community in the waters west of the island.  By the sixth century AD and later, however, this gradient seems to be much more permeable (Figure 3).

This perception that a western Mediterranean maritime community existed as a distinct social group may have reinforcement in the archaeological record.  Among the mosaics of navicularii and negotiatores at the Piazzale delle Corporazioni at Ostia, fourteen record the home cities of the individuals in question.  Of those fourteen, eight come from places west of Carthage.  Similarly, gravestones demonstrate that navicularii born around the western basin were buried in that region, whereas naukleroi born farther east were buried in a wider variety of locales.

There are numerous reasons to study the maritime landscape of the Amalfi coastline in Campania, Italy.  The rugged geology and deep river valleys have made the area’s settlements dependent on maritime activity for communication, movement, resources, trade, and revenue.  Similarly, the area between Salerno and Capri has been continually occupied since the Lower Paleolithic, and the Duchy of Amalfi was one of the four founding maritime republics of Italy, with Venice, Genoa, and Pisa.  The area is a rich palimpsest of private, commercial, public, sacred, and secular maritime spaces, and as much as the Bay of Salerno may be viewed as an isolated unit, it is equally integrated with the Tyrrhenian Sea and other spaces farther east and west.

In addition to these reasons, this survey began in 2016 prompted by preliminary results from the AMD project.  According to the modeling, much of the Tyrrhenian Sea to the south of this peninsula had strong inter-regional connections farther eastwards.  Yet, the waters north of this peninsula did not, suggesting that either the peninsula, or the bays to the north and south, represented a socially-constructed gradient between these two types of activity.  With the generous permission of the Soprintendenza Archeologia Belle Arti e Paesaggio di Salerno e Avellino, the Consorzio di Gestione Area Marina Protetta Punta Campanella, and in collaboration with the Centro di Cultura e Storia Amalfitana and the Centro Sub Costiera Amalfitana, our project is collecting data to model changes in this coastline’s maritime landscape and to test the AMD project’s conclusions.