Mapping Medieval Conflicts: Braveheart, Facebook and Karate

1

Braveheart, Facebook and Karate.

The new project „Mapping Medieval Conflicts“ at the Austrian Academy of Sciences (OEAW)

analyses conflicts in the Middle Ages with the help of digital tools of network analysis

Johannes Preiser-Kapeller, Institute for Medieval Research/Division of Byzantine Research,

Austrian Academy of Sciences

Website: http://oeaw.academia.edu/MappingMedievalConflict

For some observers, the armed conflicts of our day, in which religious fanatics kill members of

other faiths and politicians seek boundary changes by relying on centuries-old territorial

claims seem to come from the “dark Middle Ages”. Sometimes, the active power of religious

or national group identities, which motivate individuals to unspeakable acts of violence, is

incomprehensible for “modern” people. In fact, “Middle Ages”, during which warriors under

the banner of Islam or as crusaders with papal blessing spread death and devastation, in many

ways serve as a reference point - for some even as an idealized period of the heyday of the

“true” faith.

On the other hand, the memory of the beginning of WW I 100 years ago provides opportunity

to reflect on the impact of individual decisions and actions for the outbreak of conflict. In his

masterful book “The Sleepwalkers - How Europe Went to War in 1914”, the Australian

historian Christopher Clark paints a new, more complex picture of the unfolding of the world

war especially on the basis of detailed descriptions of the activities of individuals or small

groups - from the assassins of the "Black Hand" and the bullets of Gavrilo Princip in Sarajevo

1914 up to the politicians and chancelleries of the great powers.

The focus on the individual (as a hero) is also typical when in Hollywood Middle Ages is told:

in Mel Gibson's “Braveheart” (1995), the Scottish farmer William Wallace at end of the 13th

century through his personal vendetta initiates a general insurrection against the English

occupiers, which affects the whole country (in fact, the rebellion of the minor nobleman

Wallace was originally only one of several and the role of the nobility was far more significant

than narrated in the movie).

2

The individual or the society?

Actually between these two poles - overall group identities or individual activity - one of the

great debates of conflict theory and the social sciences in general takes place: Do overarching

ideologies and distinctions (by class, race, nation or religion) control the outbreak and course

of conflicts and set human communities in opposition to each other (such as the “inevitable”

class struggle of the Marxist doctrine) - the "macro-level", which determines the dynamics of

conflict “top-down”? Or shall we understand (armed) conflicts as the sum and the result of

individual decisions and acts of violence, sometimes committed between people who for a

long time had lived close to each other (as for example, during the civil wars in disintegrating

Yugoslavia) - the “micro-level”, from which the conflict dynamics emerges “bottom –up”? (cf.

Kalyvas, 2009).

For modern and medieval societies, both levels must be closely interwoven with each other

to reach a proper analysis of conflicts. The all-embracing Roman Church, its institutions and

doctrines motivated thousands of Western Europeans from the 11th century onwards to

undertake an “armed pilgrimage” to Jerusalem. However, the individual Crusaders took these

hazards on themselves for various reasons, if one looks on their immediate social environment

and the various political, economic or family ties, which influenced the possibilities and limits

of their actions. On the other hand, political structures without a monopoly for violence of a

central state authority offered wide spaces for individual conflict resolution, as through the

legal institution of the feud. In its final escalation level as a “blood feud” it allowed for the

violent enforcement of one´s own claims. Often, the feud was distinguished from “proper”

war only by the number of combatants involved. In fact, the transition between feud and war

was flowing when entire noble families, knightly alliances or cities were entangled in such

conflicts. In order to maintain such larger coalitions, overarching institutions and identities

played a role again. Also for historians, such complex mixed constellations of individual and

group conflict resolution are often confusing and difficult to unravel.

Social networks and their analysis

Social network analysis provides instruments to systematically collect such linkages from the

individual level up to that of entire state communities as well as to visualize and to analyse

them on a mathematical basis. Since the network concept has become omnipresent, in

3

particular through "social networks" such as Facebook, it also has experienced a boom in

historical sciences. Yet, the beginnings of historical network analysis in German-speaking

scholarly community can already be found in the 1970s (with the works of Wolfgang

Reinhard), and of sociological network theory in general even in the 1930s (with the

sociogramms of Jacob Levy Moreno). But especially the development of modern software

allowed the collection and analysis of large amounts of data and led to a further spread of

these approaches. Nevertheless, the actual number of studies that actually apply the

methodology of network analysis is still limited. The reluctance of historians to use the

instruments of network analysis can also be connected with the conceptual and terminological

gap between humanities and formal sciences. At the same time, the user-friendliness of

modern software tools seduces some to use them as “black boxes” to produce a variety of

visualizations and statistics without proper evaluation of the underlying concepts.

Network analysis is based on the assumption that social ties are not only relevant, but “that

they are organized in a significant way that for example this or that individual has an

interesting position due to her or his connections” (Cl. Lemercier, 2012) or that relationships

in groups occur in certain structural patterns. To detect these patterns, social network data is

collected and stored in the form of matrices (“tables” of rows and columns for each individual,

which are linked together) or of network graph (with “nodes”, that is, individuals or other

connected entities, and “edges” or “links” that represent with lines the relationships between

the nodes) (see fig. 1).

Fig. 1: The same network as graph with nodes and edges (left) and as matrix with rows and

columns (right) (J. Preiser-Kapeller, IMAFO/ÖAW, 2014)

4

Matrices and graphs serve both for the visualisation of graphs and as basis for further

quantitative analyses of the structures of the linkages (based on matrix algebra and graph

theory) (fig. 2). Thereby, differences in the “centrality” of individual nodes can be determined

due to the number of their connections or their favourable position between disconnected

groups of nodes. Other methods allow the identification of clusters and cliques as groups of

nodes that are more closely intertwined with each other than with the rest of the network,

and can represent different factions, for instance (see below). Finally, networks can be

analysed in their entirety with regard to the density and “resilience” of the web of relations

or the (un)equal distribution of central network positions among nodes.

Fig. 2: Some basic concepts of quantitative network analysis (from: Dave Gray, The Connected

Company. O'Reilly & Associates 2012)

Such network models can also be created for individual time slices to represent changes in the

structure of relationships. By combining network analysis with instruments of Geographical

Information Systems (GIS), also the spatial dimension of social relations can be made visible.

Thus, network analysis provides a framework to capture social phenomena from the micro

level of the individual and his "social environment" via the meso-level of groups to the macro-

level of entire communities and to survey types and numbers of the relationships and

interactions relevant for the (more or less strong) cohesion of such social formations.

5

Thousand documents are sometimes not enough

Data for the creation of such network graph has to be extracted from medieval sources with

much more difficulties than information about the relations between Facebook users, which

can be stored automatically and made available as "Big Data" for further digital analysis. The

repositories of past social connections and interactions are chronicles and in particular

hundreds, sometimes thousands of written records on legal transactions, administrative acts

and conflicts. These texts (in this project in Latin, Greek or the respective vernacular language)

must first be sighted, deciphered, translated and evaluated before their information content

can be used for further investigation. Therefore, the project "Mapping Medieval Conflicts" at

the Institute for Medieval Studies of the Austrian Academy of Sciences is closely coupled with

several internationally connected long-term programmes for the edition and analysis of

medieval texts (“Monumenta Germaniae Historica”, “Regesta Imperii”, “Prosopographisches

Lexikon der Palaiologenzeit”).

Fig. 3: A social network which has become a charter: the charter of alliance of the Mailberger

coalition of more than 200 knights, lords and representatives of the estates of the realm

against Frederick III, 1451/1452 (Haus-, Hof- und Staatsarchiv Wien)

Yet, even in the best cases with thousands of documents we know for sure that our

information is not complete. Written sources provide only a certain part of the spectrum of

social relations for a limited group. As for any other historical study, the researcher must be

sure that the data basis is sufficient "to demonstrate general structures and developments"

6

(M. Burkhardt, 2009) - in the case of network analysis, that significant and characteristic

patterns can be reconstructed, especially for those types of relationships (kinship, allegiance,

economic interaction, etc.), which are essential for the problem in hand. Wolfgang Reinhard,

the pioneer of the German historical network research was quite optimistic in this respect:

“The selection (of relationship types) in the sources is based on certain rules, for which the

values and norms of the historical society were of crucial importance from which the sources

originate”. (W. Reinhard, 1979). In many cases, only the systematic collection and

presentation of the relations recorded in a stock of sources allows us to discover the big gaps,

but also areas of denser evidence, which offer themselves for further structural analysis.

Conflicts, egos and triads

Even if fragmentary tradition does not allow the use of quantitative methods or only to a

limited extent, it is worthwhile to take systematically in the focus the social connections

between individuals and groups as the context of their actions, especially when it comes to

conflicts. Every single actor was embedded in an abundance of relationships which he or she

had received by birth (e. g. kinship) or that he or she actively established and maintained (e.

g. the membership in a fraternity). These links could be connected with different positions in

more or less formalized and institutionalized systems of order (as the patron of followers or

as a follower of a higher ranking patron, for instance) and could play an essential role for the

identity and overall social position of an individual (e. g. the integration into the networks of

peers as a confirmation of the noble status). In the case of conflict, an “ego network” thus

could serve as a resource (support from relatives, friends, allies, patrons), but also limit the

room for manoeuvre, due to obligations as a follower of another patron, for instance.

Particularly problematic was the integration in competing networks in the case of conflict

between two patrons or friends.

In such situations, conflict could act as structure-changing force, as the German medievalist

Robert Gramsch has shown for the German Kingdom of the 13th century. He resorted back to

the relative simple principle of “structural balance” back, which was developed in the 1950s

by the psychologists Fritz Heider (born in Vienna in 1896): Heider focused on three

interconnected actors (the “triad”) as the smallest network formation. These actors can

maintain positive (e.g., friendship) or negative (e. g., conflict) relationships with each other. If

all actors are friends with each other, the triad is considered “balanced”. If one actor is friends

7

with the other two, but these are in conflict, then this balance is disturbed, which leads to

“cognitive stress” for the actor friend to both disputing actors. She or he now has two options:

either to try to reconcile the two rivals, or to choose a side and to end the positive relationship

with one of the other two. In both cases, the structural balance is restored. According to

Heider, unbalanced triads in social networks are unstable and tend to change into a balanced

state (fig. 4).

Fig. 4: Balanced and not balanced triads in social networks in the theory of Fritz Heider (from:

Gramsch, 2013)

Robert Gramsch was able to verify the validity of this principle for the network of alliances and

conflicts between the princes of the German Kingdom of the 13th century and to show how

these simple mechanisms frequently caused rearrangements in the structure of the different

factions. Via manifold linkages, decisions of individual actors thus led to changes in the

political system as a whole (fig. 5).

Fig. 5: Groupings and conflict lines in the network of the German princes in the year 1232 (from:

Gramsch, 2013)

8

The power of the group

This is just one example of the entanglement between the micro and macro level. As indicated

above, both individual and overarching social identities require mutual affirmation and

demonstration in the encounter and interaction with others (during worship and other

religious meetings, in the ceremonial at the ruler´s court, etc.). One of the most important

scholars of medieval political symbolism, Gert Althoff, states: “power had to be brought to

view in the Middle Ages. (...). Via ceremonial and ritual practices, obligations were assumed,

relationships presented, rights recognized and much more. Exercise of power took place

obviously very much in such acts. Its character was expressed nowhere more directly than in

the frequent interaction of the powerful in public”. Thus visible became also the “social

capital” of the individual (to speak with Pierre Bourdieu), that is “the totality of actual and

potential resources which are connected with the possession of a durable network of more or

less connected institutionalized relationships of mutual knowing or acknowledging”. Also for

Bourdieu, these “resources based on membership in a group” for their reproduction require

regular interaction, by which mutual recognition is reinforced (P. Bourdieu, Die verborgenen

Mechanismen der Macht 1992). Equally, the identity of groups and communities is confirmed

and becomes manifest, is reproduced, but may also be modified by such acts. At the micro-

level, these elements of identity can serve as additional entry points for the establishment of

networks and the taking of sides in a conflict beyond immediate realities of life such as kinship

or neighbourhood. Sometimes they even demand a solidarity overarching those basic links

(and neighbours kill neighbours). The cohesion and therefore the assertiveness of a group in

turn depends on the potential of their networks to enforce individual support and loyalty.

Civil war in Byzantium: coincidences, networks and Karate

As in many parts of Europe, Africa and Asia, also for the Byzantine Empire the “calamitous 14th

century” (as it was called by Barbara Tuchman in her classic “A Distant Mirror”) was a time of

crisis and conflict. External enemies, climate change associated with natural disasters, and

(since the middle of the 14th century) the plague threatened the existence of the empire, while

the imperial family and the elite weakened themselves in internal conflicts (fig. 6).

9

Fig. 6: Years of internal unrest in the Byzantine Empire per decade, 1200-1400 (J. Preiser-

Kapeller, IMAFO/ÖAW, 2014)

The first of these civil wars in the years 1321 to 1328 owed its outbreak (similar to the First

World War, or, at least according to Mel Gibson, the revolt of the Braveheart) to the deeds of

one man: Andronikos III Palaiologos, grandson of the reigning emperor Andronikos II, and like

his father Michael IX formally co-emperor of Byzantium. The 23-year old younger Andronikos

had fallen in love with a beautiful noblewoman in the capital of Constantinople; but the Lady

granted her favours also to other young aristocrats. The jealous emperor's grandson hired

assassins who in the night should lay an ambush for the rivals near the house of his beloved.

But the first who fell into their hands and lost his life was the brother of Andronikos III, Manuel.

This news was too much for their ailing father Michael IX, who died in October 1320. The

equally shocked grandfather disinherited Andronikos III and nominated another grandson as

his successor.

But Andronikos III had found his own followers especially among the younger representatives

of the Byzantine aristocracy, who were extremely dissatisfied with the regime of the elder

Andronikos who had ruled for almost 40 years with rather limited success. Skilfully, some of

the closest friends of Andronikos III, including the rich 25 years old nobleman Ioannes

Kantakuzenos, acquired important command posts in the provinces of the empire, which they

used as a base to build up resistance against the old Emperor in Constantinople. In the spring

of 1321, the younger Andronikos and his followers declared war on Andronikos II and

demanded the reinstatement of the heir to the throne. The representatives of the elite had to

take sides now, and the fracture lines were drawn between the generations within the

10

individual families. A telling example is the family of Theodoros Metochites, chief minister and

a close confidant of the old emperor. If we visualize his ego-network, we find him embedded

in a dense network of followers and friends around Andronikos II, to whom he remained loyal

in the Civil War. His sons Nikephoros and Demetrios, however, were far less connected to the

“old” generation of the regime and changed to the camp of the younger Andronikos (fig. 7).

Fig. 7: The Ego-Network of the Byzantine aristocrat Theodoros Metochites in the year 1321,

with ties of kinship (red lines), friendship (green) and allegiance (blue) (J. Preiser-Kapeller,

IMAFO/ÖAW, 2014)

Fig. 8: Theodoros Metochites as founder of the church of the Chora Monastery, mosaic in

Istanbul from the 14th century (wikimedia.commons)

11

The polarization between the camps of the old and the young Andronikos, which we observe

for one family on the example of Metochites, affected the entire Byzantine elite, as a look at

a visualisation of the network of allegiance in this group (141 individuals) for the year 1321

clearly shows (fig. 9).

Fig. 9: The network of allegiance in the Byzantine elite in 1321 (141 individuals (J. Preiser-

Kapeller, IMAFO/ÖAW, 2014)

If one visualises the ties of friendship and of kinship between these persons, it becomes laos

evident that this conflict in many cases overlaid often long-standing networks and (according

to the theory of "structural balance" of Heider so) enforced a “reorganization” of many triads

(fig. 10).

Fig. 10: Networks of kinship (left) and of friendship (right) between members of the Byzantine

elite in 1321 (J. Preiser-Kapeller, IMAFO/ÖAW, 2014)

12

To assess the cumulative, partly mutually opposing force of these linkages on the structure of

a group, various methods for determining the degree of polarization or fragmentation of a

network exist. The physicist M. Newman developed algorithms, which identify sub-clusters

which are internally structurally stronger intertwined than with the rest of the network, and

thus could mark potential lines of breakage. In one test, he applied these algorithms to

network data of a karate club in the United States, which had been observed for a longer

period of time by social scientists and had broken up due to the polarization of the friendship

relations between two popular instructors. The Newman algorithm ordered the nodes in a

network graph for the period immediately prior to the breakup into two groups, which (with

the exception of one node) were identical with the two new clubs that had emerged after the

breakup of the karate club (fig. 11).

Fig. 11: Identification of clusters and potential lines of breakage in the friendship network of a

Karate Club (from: Newman, 2010)

If we apply this algorithm on a network model of all ties of kinship, friendship and allegiance

in the Byzantine elite for the year 1321, the actual fragmentation of this group becomes

visible. In addition to two larger clusters of nodes that are assigned to one of two emperors,

the method identifies smaller clusters that do not clearly belong to one camp as that of the

Metochites family. Potential lines of divisions and cores for the emergence of factions in a

systematic synopsis of the relevant relationships thus become visible already for a time before

the actual outbreak of the conflict (fig. 12).

13

Fig. 12: Identification of clusters in the network of the Byzantine aristocracy with the help of

the Newman-algorithm: clusters of Andronikos II (light blue), Andronikos III (dark blue), the

Palaiologoi (dark green), Kaballarioi (yellow), Tornikes (light green) and Metochites (red) (J.

Preiser-Kapeller, IMAFO/ÖAW, 2014)

While this division in two parties rearranged the “social space” of the realm, also the

remaining Byzantine territories and cities in the southern Balkans, Greece and western Asia

Minor were captured by the competing networks of allegiance of the two Andronikoi. This can

be demonstrated with a combination of the link between the various actors with geographical

information about the places of their residence or their travels in the years of the civil war

from 1321 to 1328 in a comparison for the two groups (fig. 13 and 14).

14

Fig. 13: The connections between localities due to ties of allegiance in the ego-network of

Andronikos II, 1321-1328 (J. Preiser-Kapeller, IMAFO/ÖAW, 2014)

Fig. 14: The connections between localities due to ties of allegiance in the ego-network of

Andronikos III, 1321-1328 (J. Preiser-Kapeller, IMAFO/ÖAW, 2014)

However, the close social and familial connections between representatives of the two parties

also contributed to a de-escalation of the conflict; several times, these channels were used to

achieve temporary settlements between Andronikos II and his grandson, which included a

15

renewed recognition of the younger Andronikos as (co)emperor and a division of power

between the two. At the same time, the fundamental rift between the generations of the elite

could only be removed when in 1328 the old emperor was forced to abdicate and the

representatives of the “ancien régime” were removed from the positions of power (Theodoros

Metochites, for instance, became a monk). If one analyses the faction of Andronikos III

quantitatively, it was indeed smaller in number, but the structure of the interconnections

between its members was closer “knitted” and thus more resilient in crises periods of the

conflict. The following of the old Emperor was more numerous, however, in its entirety - with

the exception of the above-outlined core of the “old guard” with Theodoros Metochites –

more loosely linked with regard to its network patterns; soon defections to the side of the

younger Emperor occurred (the sons of Metochites as one example) (fig. 15).

Fig. 15: Some quantitative measures for the clusters identified in the network of the Byzantine

elite (J. Preiser-Kapeller, IMAFO/ÖAW, 2014)

The mapping and analysis of these networks thus provides valuable new insights into the

structural background to the course and outcome of this conflict. In the immediate social

environment of each actor as in the leading group of the state as a whole, different cohesive

and centrifugal forces of entanglement worked with and against each other, which

significantly affected both individual decisions as well as the identity and success of groups.

Mapping Medieval Conflicts – the aims

The main aim of the project is to survey, to map and to analyse the different categories of

links, their importance and their interplay (e.g. between kinship and status bound or ethno-

religious solidarities – “To the Christian Nobility of the German Nation”) for the outbreak and

dynamics of conflict for individual actors, groups and larger communities in their temporal and

spatial dimension. Thus, also a comparative perspective for similar phenomena in other eras

up to the present time is established.

Cluster Number of nodes Density Path length Clustering Coefficient Degree centralisation Betweenness centralisation

Andronikos II 67 0.041 1.985 0.378 0.973 0.982

Andronikos III 37 0.099 2.204 0.524 0.747 0.82

Palaiologoi 16 0.75 1.283 0.83 0.133 0.253

Tornikes 6 0.73 1.267 0.767 0.4 0.4

Metochites 5 1 1 1 0 0

Kaballarios 8 0.64 1.357 0.782 0.476 0.367

Total network 139 0.039 2.183 0.544 0.8 0.83

16

From a technical point of view, the aims of MEDCON are:

• The adaptation and combination of a set of software tools which facilitates the

relational survey of medieval sources and the visualisation and quantitative analysis of social

and spatial networks. We use an open source database application named “OpenATLAS”,

developed by Stefan Eichert (who is also part of the team) in Vienna for the survey and storage

of data on archaeological and historical-geographical evidence

(http://www.openatlas.eu/conc/). OpenATLAS is an open source-Software, which uses

PostgreSQL with PostGIS as backend and therefore is compatible with nearly all GIS-

applications. For the structuring of data and meta-data classes and attributes of CIDOC-CRM

(Dublin Core) are used, with guarantees a high degree of connectivity with other data bases

• The development of case studies demonstrating a “best practice” of the application

and evaluation of tools of network analysis for medieval history (distribution as open data)

• The creation of an online platform for the exploration of data, methods and results by

the wider public (open access)

A generalizable work flow from data input on the basis of medieval sources to the creation,

visualisation and analysis of social and spatial network models and their web-based

publication and presentation will be established (see fig. 17 below).

In order to demonstrate this in detail, the project will focus on the analysis of political

networks and conflict among power elites across medieval Europe with five case studies (fig.

16):

• Fluctuations between opposing parties in the struggle for the German throne 1198-

1208 (Andrea Rzihacek, Renate Spreitzer)

• Coalitions in the war of Emperor Sigismund against Duke Frederick IV of Tyrol (Günter

Katzler)

• Emperor Frederick III and the League of the Mailberger coalition in 1451/52 (Kornelia

Holzner-Tobisch)

• Factions and alliances in the fight of Maximilian I for Burgundy (Sonja Dünnebeil)

• Political factions in 14th cent. Byzantium (Johannes Preiser-Kapeller)

17

Fig. 16: The areas of the five case studies of the project „Mapping Medieval Conflict“

“Mapping Medieval Conflicts” will evaluate the explanatory power of network tools for

phenomena of political conflict in medieval societies. The project uses the relational

structuring provided by modern software not simply as instrument for the organisation of

data, but as heuristic tool for the reconstruction and analysis of the relational character of

social phenomena of the past. Thus, also the additional benefit of digital tools beyond data

collection and their potential to allow for new research questions and analytical results will be

demonstrated.

At the same time, the comparative analysis of these conflicts in their dynamics from the micro

to the macro level establishes the Middle Ages as a reference point for the study of similar

phenomena in history and present times without the aforementioned clichés of a dark and

bloody era.

18

Fig. 17: The workflow for the project „Mapping Medieval Conflicts“

19

Selected bibliography and web links:

G. Althoff, Die Macht der Rituale. Symbolik und Herrschaft im Mittelalter. Darmstadt 2003.

Th. Bonacker (ed.), Sozialwissenschaftliche Konflikttheorien. Eine Einführung. Wiesbaden

2008.

M. Burkhardt, Der hansische Bergenhandel im Spätmittelalter: Handel – Kaufleute –

Netzwerke. Vienna – Cologne 2009.

Robert Gramsch, Das Reich als Netzwerk der Fürsten. Politische Strukturen unter dem

Doppelkönigtum Friedrichs II. und Heinrichs (VII.) 1225-1235. Ostfildern 2013.

J. Habermann, Verbündete Vasallen: Die Netzwerke von Grafen und Herren am Nordwestharz

im Spannungsgefüge zwischen rivalisierenden Fürstgewalten (ca. 1250-1400). Norderstedt

2011.

E. Jullien, Netzwerkanalyse in der Mediävistik. Probleme und Perspektiven im Umgang mit

mittelalterlichen Quellen. Vierteljahrschrift für Sozial- und Wirtschaftsgeschichte 100/2

(2013), 135–153.

Ch. Kadushin, Understanding Social Networks. Theories, Concepts, and Findings. Oxford 2012.

St. Kalyvas, Conflict, in: P. Hedström – P. Bearman (eds.), The Oxford Handbook of Analytical

Sociology. Oxford 2009, 592-615.

Cl. Lemercier, Formale Methoden der Netzwerkanalyse in den Geschichtswissenschaften:

Warum und Wie?, in: Müller, Albert; Neurath, Wolfgang (eds.) (2012), Historische

Netzwerkanalysen (Österreichische Zeitschrift für Geschichtswissenschaften 23/1). Innsbruck

– Vienna – Bozen, 16–41.

M. E. J. Newman, Networks. An Introduction. Oxford 2010.

J. Preiser-Kapeller, Complex historical dynamics of crisis: the case of Byzantium, in: S. Jalkotzy-

Deger - A. Suppan (eds.), Krise und Transformation. Vienna 2012, 69-127.

J. Preiser-Kapeller, Networks of border zones – multiplex relations of power, religion and

economy in South-eastern Europe, 1250-1453 CE, in: Proceedings of the 39th Annual

Conference of Computer Applications and Quantitative Methods in Archaeology, "Revive the

Past" (CAA) in Beijing, China. Amsterdam 2012, 381–393 (online: http://oeaw.academia.edu/

JohannesPreiserKapeller/Papers/).

20

W. Reinhard, Freunde und Kreaturen. "Verflechtung" als Konzept zur Erforschung historischer

Führungsgruppen. Römische Oligarchie um 1600. Munich 1979.

N. Tackett, The Destruction of the Medieval Chinese Aristocracy. Cambridge, Mass. – London

2014.

H. Vollmer, The Sociology of Disruption, Disaster and Social Change: Punctuated Cooperation.

Cambridge 2013.

H. C. White, Identity and Control. How Social Formations emerge. 2nd ed., Princeton – Oxford

2008.

The project: http://oeaw.academia.edu/MappingMedievalConflict

The go!digital-Programme of OEAW: http://www.oeaw.ac.at/oesterreichische-akademie-der-

wissenschaften/news/article/die-neuen-dimensionen-des-forschens/

The Institute for Medieval Research of OEAW: http://www.oeaw.ac.at/imafo/

The Interdisciplinary Working Group “Digital Middle Ages” at IMAFO:

http://www.imafonet.at/dma/

The OpenATLAS-software of Stefan Eichert: http://www.openatlas.eu/conc/

Further examples for the visualisation of medieval networks:

http://oeaw.academia.edu/TopographiesofEntanglements

Platform for historical network analysis: http://historicalnetworkresearch.org/