Dromography (Historic Routes Research)

To what extent long-distance transport in north-western Europe changed after the Roman period is generally unknown. Few historical sources are available and existing archaeological records are unclear and sometimes conflicting. Traditionally, research on the long-distance exchange of goods mostly has focussed on the spatial analyses of luxery goods such as jewellery, weapons and religious artefacts. Relatively little attention has been paid to the spatial modelling of common exchange networks and transport routes.
In this study we used a dendroarchaeological approach to model long-distance transport of oak (a common good) to the Roman and early-medieval Netherlands. By combining established and newly-derived  provenances of imported timbers with data on Roman and early-medieval route networks, we were able to reconstruct: (a) Roman and early-medieval exchange networks of imported timbers, (b) changing transport routes and (c) spatially shifting frequent-travel zones. The findings were compared with distribution patterns of other commodities for daily use: pottery and stone household goods.
Results show that in the early and middle-Roman periods (12 BC – AD 270) timbers were imported from the German Rhineland, the Ardennes and the Scheldt region. We have no evidence for wood import to the current Netherlands during the late-Roman period and first phase of the Early Middle Ages (AD 270 – 525). In the following centuries, between AD 525 – 900, oak again was brought to the current Netherlands, this time exclusively originating from the German Rhineland. This pattern significantly changed during the last phase of the Early Middle Ages (AD 900 – 1050) when timbers were derived from the Ardennes only. We used these patterns to calculate changes in long-distance transport routes and frequent-travel zones in the research area. Through our analyses existing data on Roman and early-medieval route networks could be expanded and improved. The calculated wood-transport patterns agree well with the distribution of imported pottery and  (other) household goods in these periods.

ABSTRACT
Mobility is one of the key aspects of 21st century’s globalization. It may sound inappropriate
to use the latter word when referring to past civilizations, however, it is indisputable that in
past societies as well several groups of people were constantly on the move (merchants,
pilgrims, craftsmen, soldiers). Because of their relevance to the understanding and
explanation of many aspects of human behavior in the past, mobility studies are central to
archaeological research. In addition, they gain increasing importance in view of the
contemporary challenges posed by our society’s ever-increasing mobility. Mobility must be
supported by roads, routes, paths, tracks. In this sense we believe that “the history of roads is
not the study of the dead past, but of the living present” (Taylor 1979). This consideration
represents the core concept of the session and our goal is mainly to explore mobility through
the routes that served as its backbone. In this session we especially wish to invite
contributions that consider three main issues. First, we focus on mobility as a network of
connected routes, also in relation to the spreading of goods, ideas, military conquest and
domination. Second, we consider the materiality of routes, how did they make them, what
does remain in the archaeological record. Third, we think that the recovering of ancient
routes in relation to contemporary mobility can help the sustainable development of a given
territory for, both, tourism and inclusive societies. The focus of the session is mainly
methodological and theoretical and it is thus open to any chronological period.

For this case study of connectivity among sites in the regions of the historical Armenian kingdom of Vaspurakan (around Lake Van) in the period between the 9th and 11th century CE, I first mapped all sites for which larger scale building activity is documented (in written sources or by monumental evidence) in this period on the basis of the following study: J. M. Thierry, Monuments arméniens du Vaspurakan. Paris 1989.
I then constructed a model for the network between this sites by connecting every site with all neighbouring ones within a radius of 50 km.
For the emerging network, I weighted the strength of connections between sites indirectly proportional to the geographical distance between them (strength of ties = 1/number of km).
For this network model, I determined the centrality measure of betweenness (quantifying the ratio of shortest connections between nodes a node is part of and indicating the significance of a node as « intermediary » between otherwise not directly connected nodes or groups of nodes)
Furthermore, I applied the Newman grouping algorithm in order to detect clusters of nodes more densely connected among each other than with the rest of the network.
The network visualised here thus constitutes a first approach towards the modelling of connectivity within the natural and built environment of an early medieval polity in historical Armenia and is part of a larger ongoing study on the emergence and development of the Kingdom of Vaspurakan within the matrix of politics, economy, geography and ecology.
For an overview of similar applications see also:
Barthélemy, M., Spatial Networks. Physics Reports 499 (2011) 1-101.
Conolly, J./Lake, M., Geographical Information Systems in Archaeology (Cambridge Manuals in Archaeology). Cambridge: Cambridge 2006.
Gorenflo, L. J./Bell, Th. L., Network Analysis and the Study of past regional Organization, in: Trombold, Charles D. (ed.), Ancient road networks and settlement hierarchies in the New World. Cambridge 1991, 80-98.
Rodrigue, J.-P., with Comtois, Cl./Slack, B., The Geography of Transport Systems. 3rd ed., London – New York 2013.
This study, which is work in progress, was made possible on the basis of a fellowship of the Alexander S. Onassis Public Benefit Foundation for research at the Institute for Historical Research of the National Hellenic Research Foundation in Athens.

In the manuscript of the Greek sailor Michael of Rhodes, who served in the Venetian fleet from 1401 to 1443, the author also provides a list of names and places of origin of many of the oarsmen working on a ship which sailed from Venice to Jaffa and back between May 9th and August 15th 1414 under the command of Francesco Querini (cf. Stahl, Alan M., Michael of Rhodes: Mariner in Service to Venice, in: Long, Pamela O. (ed.) (2009), The Book of Michael of Rhodes. A Fifteenth-Century Maritime Manuscript. Volume 3: Studies. Cambridge, Mass. – London, pp. 35–98, esp. 42-43).
If we combine this data into a network model, we can visualise how the ship connects the places of origin of its crew with the localities on its route from Venice to the East (fig. 1)
Michael of Rhodes´ data also allows us to visualise the relative significance of localities on the basis of the respective number of oarsmen coming from each of them (fig. 2). The largest numbers came from Venetian possessions and other sites in Dalmatia and Albania as well as from further inland of the Western Balkans, but also from the Italian hinterland of Venice, Hungary and Germany, as well as from the Eastern and Western Mediterranean (for this phenomenon cf. also Doumerc, Bernard (2007), Cosmopolitanism on Board Venetian Ships (Fourteenth-Fifteenth Centuries). Medieval Encounters 13, pp. 78-95).
We can visualise this assemblage of individuals from all over the Mediterranean in Venice for the purpose of this journey also on a map (nodes again scaled according to the number of oarsmen originating from there) (fig. 3).
Finally, this social network of the ship of 1414 is of course a mobile one (fig. 4), so that this assemblage of people and their places of origin connects to all ports on its route from Venice to Jaffa (fig. 5), establishing a complex web of individual entanglements across the entire Mediterranean; the ship thus emerges as a “heterotopia”, a real place in which various societies and cultural backgrounds of the time were “simultaneously represented, contested, and inverted” (for the adaption of this concept of Michel Foucault cf. Van de Noort, Robert (2011), North Sea Archaeologies: a Maritime Biography, 10,000 BC to AD 1500. Oxford, pp. 33-34).
All graphs were created by Johannes Preiser-Kapeller with the help of the software ORA*; for further questions please contact Johannes.Preiser-Kapeller@oeaw.ac.at (do not use any material without the consent of the author!).

The original data for this model stems from: Ciolek, T. Matthew, 2005. Georeferenced data set (Series 1 - Routes): Trade routes in the Ottoman Empire 1300-1600 CE. OWTRAD Dromographic Digital Data Archives (ODDDA). Old World Trade Routes (OWTRAD) Project. Canberra: www.ciolek.com - Asia Pacific Research Online. (www.ciolek.com/OWTRAD/DATA/tmcTRm1300.html)
This dataset was significantly enlarged and modified by me in order to map the most important maritime and terrestrial routes across the Eastern Mediterranean in the period 1300-1500 AD, with a focus on the Balkans and Asia Minor, the core regions of the Byzantine as well as of the Ottoman Empire.
This dataset was then turned into a topological network model, in which localities serve a nodes and the routes between them as links.
In order to integrate “real space” into the model, the strength of links between nodes was weighted according to their geographical distance (the larger the distance the bigger the cost of interaction and the weaker the connection - for terrestrial routes, a cost factor of five has been applied in comparison with maritime routes).
On this basis, several measurements of network centrality have been calculated in order to quantify the relative significance of sites and routes within the model for the Eastern Mediterranean. In addition, two analyses for diffusion and expansion related to historical developments (the spread of the Black Death from the Crimea and the expansion of the Ottoman state from Northwestern Asia Minor) have been executed.
This study, which is work in progress, was made possible on the basis of a fellowship of the Alexander S. Onassis Public Benefit Foundation for research at the Institute for Historical Research of the National Hellenic Research Foundation in Athens.

Free download: http://digitalcommons.asphs.net/bsphs/vol40/iss1/2/
Free download: Villuga Gazetteer of the 16th-century Iberian Peninsula:
http://www.isu.edu/history/hibberdgazetteer.shtml
It is frequently difficult to write a geographically-integrated history of Portugal and Spain prior to the late nineteenth century because researchers often lack a means by which to identify the locations of historical places on the basis of the geographic coordinates of their modern counterparts. This article presents a free, downloadable digital gazetteer, which the authors have founded on a 1546 traveler's guide to 139 major routes in " Spain " (as the Iberian Peninsula was then known), in which the author, Juan Pedro Villuga, listed each stopping place along each route and indicated the approximate distance between each of them. The authors explain the technical problems of gazetteer and database design and of associating historic placenames with their modern equivalents and geographic coordinates of longitude and latitude, and of visualizing them in geographic information systems (GIS). In the gazetteer, the authors link the sixteenth-century Iberian place names with more modern ones, which were largely solidified in the modern country of Spain by the territorial organization in 1834. More importantly, the authors offer suggestions about how this type of research infrastructure can be used to address various historical problems in Portuguese and Spanish history, which are related to geographic space, topography, and distance.

Data from: F. Hild – H. Hellenkemper, Kilikien und Isaurien (Tabula Imperii Byzantini 5). Vienna 1990, esp. pp. 128-142; H. Hellenkemper – F. Hild, Lykien und Pamphylien (Tabula Imperii Byzantini 8). Vienna 2004, esp. pp. 244-293;  G. Graßhoff – F. Mittenhuber (eds.), Untersuchungen zum Stadiasmos von Patara. Modellierung und Analyse eines antiken geographischen Streckennetzes. Bern 2009, esp. pp. 104-121.
The following network visualisations and calculations are based on the scheme for the systems of routes at land and sea in the Byzantine period as depicted in the above mentioned volumes, regardless of the relative significance of the respective routes in various periods of Late Antiquity and Byzantine history. At the same time, it neither takes into account the actual distance (and travel costs) between localities nor the connections via sea routes; therefore, the model is only a first rough approximation towards a more accurate model of the Byzantine transport system in its dynamics through centuries (cf. also Graßhoff – Mittenhuber, 2009, for a much more sophisticated model for Lycia). Interesting for this study is especially the modification of centrality measures if sea routes are added to the network of land routes.
Three centrality measures have been calculated; nodes in the graphs are scaled according to their relative centralities in this regard (centrality measures are of course only valid within the extract of the total route network of Asia Minor integrated into the network):
Closeness; closeness centrality measures the length of all pathes between a node an all other nodes. The more central a node is the lower its total distance to all other nodes. Closeness can also be used as a measure of how fast it would take to spread resources or information from a node to all other nodes.
Betweenness; betweenness centrality measures the extent to which a node lies on paths between other nodes and indicates the relative significance of a node as “intermediary” within a network due to its position on many (or few) possible shortest routes between other nodes.
Eigenvector; eigenvector centrality is a measure of "indirect" centrality and indicates, if a node is connected to more or less central other nodes within the network.
For a similar study cf.  L. Isaksen, The application of network analysis to ancient transport geography: a case study of Roman Baetica. Digital Medievalist, 4 (2008) (http://eprints.soton.ac.uk/204515/)
For more information, contact: Johannes.Preiser-Kapeller@oeaw.ac.at 
All network graphs were created by the author with the help of the software package ORA*.