The Power of Networks

Prospects of Historical Network Research

Florian Kerschbaumer, Linda von Keyserlingk-Rehbein, Martin Stark, Marten Düring, Florian Kerschbaumer, Linda von Keyserlingk-Rehbein, Martin Stark, Marten Düring

• Publisher: Taylor & Francis
• Year: 2020
• eBook ISBN: 9781351744997
• Edition: 1
• Topic: History
• Subtopic: World History
• Index: History

The Power of Networks describes a typology of network-based research practices in the historical disciplines, ranging from the use of quantitative network analysis in cultural, economic, social or political history or religious studies, to novel approaches in the Digital Humanities.

Network data visualisations and calculations have proven to be useful tools for the analysis of mostly textual sources containing relational information, offering new perspectives on complex historical phenomena. Including case studies from antiquity to contemporary history, the book provides a clear demonstration of the opportunities historical network research (HNR) provides for historical studies. The examples presented within the pages of this volume are arranged in a way to highlight three central typological pillars of HNR: (re-)construction and analysis of historical networks; computational extraction of network data and infrastructures for data collection and exploration.

The Power of Networks outlines the history and current state of research in HNR and points towards future research frontiers in the wake of new digital technologies. As such, the book should be essential reading for academics, students and practitioners with an interest in digital humanities, history, archaeology and religion.

1
Introduction

Almost four decades ago, the historian Wolfgang Reinhard was the first to recognise the potential added value of network analysis methods for social historiography in German-speaking countries. Using the term “entanglement analysis”, he imported theoretical and formal concepts developed within the emerging field of social network analysis and tried to apply them to research in early modern history. The resulting “entanglement history” produced some valuable work on social entanglement (“Verflechtung”) in Upper German cities and the early modern Papal State, but it failed to become a recognised method in historical studies. Instead, the impetus for historical network research (HNR) came from sciences that lend themselves more to quantitative approaches. Accordingly, some of the most widely received empirical network analyses on the basis of historical data were produced by American historical sociologists. Their research often focused on political and social upheavals and the resulting social tensions. In recent years, network approaches in history have been rediscovered and adapted by historians of all eras for their research objectives. The resulting research landscape is very heterogeneous, and its development is highly dynamic. Methods range from the mere adaptation of the theoretical concepts of SNA to the application of complex mathematical models. Accordingly, there is still no consensus on how social science network methods can be adequately applied to historical topics and sources and what added value network theory and network analysis methods can really offer to historical research.

The editors of this anthology were directly involved in this rediscovery of network approaches in history through their own doctoral theses. They met for the first time in 2008 at a social science-orientated methodological summer school for social network analysis at Trier University in Germany. The idea soon emerged that they should regularly exchange ideas on the specific challenges that arise in adapting network analysis theories and methods to history. Starting in 2009, this led to the establishment of a series of workshops on historical network research, initially in German, which took place every six months, with a first event in Cologne. These events in Germany, Austria and Switzerland, mostly organised by local third parties, offered newcomers the opportunity to discuss their project ideas and learn about other projects and solutions. Separate method training was subsequently added. In 2019, the thirteenth workshop in this series, “Networks Across Time and Space: Methodological Challenges and Theoretical Concerns of Network Research in the Humanities”, took place at the Academy of Sciences and Literature in Mainz, Germany. For the first time, the workshop was held entirely in English, and the programme also took into account the needs of advanced users, in an attempt to reflect the existing research landscape. These events were and continue to be supplemented by the Historical Network Research website, which provides a bibliography, a calendar of events, researcher profiles and other resources and the email newsletter of the same name.

In 2013, the European Network for Digital Methods in the Arts and Humanities (NeDiMAH) gave us the opportunity to organise a first international conference on historical network research in Hamburg. This was followed by conferences in Belgium (Ghent 2014), Portugal (Lisbon 2015), Finland (Turku 2017) and the Czech Republic (Brno 2018), run by local organisers. From 2013 onwards, we were also able to hold sessions on historical networks at the International Sunbelt Conferences of the International Network for Social Network Analysis (INSNA), and from 2014 on at the corresponding European Regional Conferences (EUSN), together with various colleagues, which drew a great response. We would like to thank all our colleagues who have accompanied us on this journey over the past few years.

On the basis of all these events and activities, we published the German “Hand-buch Historische Netzwerkforschung” (Handbook on Historical Network Research) together with our Cologne-based colleague Ulrich Eumann in 2016. This handbook was a first attempt to bring together existing knowledge about the emerging field of historical network research. The aim was not to canonise this knowledge but rather to look back on what has been achieved so far and establish a foundation for further development. Questions that have arisen in recent years, from our own and external research projects, were addressed: What are the consequences of heterogeneous sources? What do centrality measures and network visualisations ultimately tell us? The universality of the network approach is accompanied by diverse possibilities for implementation and evaluation, which can be adapted to the specifics of questions. This means that no two projects are alike. Readers of the handbook are thus given an overview of the potentials, problems and challenges of historical network research, enabling them to design their own projects on this basis.

The year 2017 saw the publication of the first issue of the online, open access Journal of Historical Network Research. The idea behind the journal as an international publication solely devoted to the study of networks (social or otherwise) from a specifically historical perspective is to bring together research that seems relevant but is currently dispersed, thereby contributing to the consolidation of the field. The journal also promotes an exchange between different areas of historical research (in the broadest sense), the (digital) humanities, social and computer sciences and various research traditions and disciplines, as well as attempting to strengthen the dialogue between network research and traditional historical research.

This anthology follows the fundamental ideas behind the handbook and the journal, with a different layout. It is intended to give English-speaking readers an overview of the field of historical network research. Since a mere repetition would serve no purpose, the anthology is not structured like a handbook. Instead, selected thematically sorted research essays are used to introduce the reader to essential pillars of historical network research in a first attempt at consolidation. In addition, a glossary enables readers with little or no specific prior knowledge to orient themselves in this new and exciting field of research.

The enormous popularity of networks as a central object in historical research¹ has led to a rapidly growing interest in data-driven historical network research (HNR) across the historical disciplines.² The concept of analysing individuals not only in an isolated state but embedded in a larger social context with their resulting interdependent relationships represents an essential part of both social sciences and humanities.³ As a result, systematic research on the influence of social structures on agency and the impact of agency on social structures is crucial for the field of history and may be of particular value for network-oriented perspectives. As Claire Lemercier has pointed out, social network analysis (SNA) is not a neutral process; it enables researchers to perceive their objects of research from a new perspective.⁴ SNA is inherently biased towards a worldview that attributes explanatory power to social relationships. The gradual appropriation of SNA methods from the social sciences into historical research workflows has been accompanied by three distinct areas of tension.

The first synergistic field concerns the ongoing popularity of the term “network”⁵ and its metaphorical use within the academic world as well as within other societal domains. The versatile implementation of the term may not always correspond to the accuracy and adequacy⁶ called for by Max Weber. Wolfgang Reinhard, the earliest adopters of SNA for historical research in Germany, observed that the term network has turned, along with discourse, into another empty phrase.⁷ Nevertheless, recent research (including the chapters in this book) contributes fundamentally to critically reflect on and to sharpen the definition of the network concept in historical research by taking on a non-metaphoric approach to build a sound methodological and theoretical basis for HNR.

A second synergistic dimension is the strained relationship between quantitative and hermeneutical approaches in historical research.⁸ Quantitative methods have a long tradition within this field. Several innovative initiatives stemming from areas like cliometrics and new institutional economics (NIE) have produced a variety of results, including the Nobel-Prize-winning work of Robert Fogel and Douglas North.⁹ By linking source-based data collection and quantitative analysis with traditional source criticism and reflective interpretation, many projects in HNR effectively bridge the gap between qualitative and quantitative approaches.

In recent decades, digitization has created a third synergistic dimension.¹⁰ Increasing use of digitized sources and digital tools in the humanities has brought about profound changes, and the impact of digitization – from the search for materials to the analysis of sources to the communication of research results – in the field of history can be observed today.¹¹ This widely discussed approach initially emerged in the middle of the 20th century¹² and became known as “humanities computing”, later as “digital humanities” from which later emerged the term “digital history”. These processes not only raise awareness of the substantially growing digital toolbox available to historians; they also point towards new research methods and different ways of thinking about and perceiving research content. Digital history involves large-scale collections of digitized sources, ever-growing metadata catalogues and numerous efforts to interlink existing repositories. The data models used for these endeav-ours are not always relevant or even useful for network analysis research questions (nor should they be). Nevertheless, free access to a variety of datasets opens up exciting new opportunities, especially when combined with collaborative approaches such as crowdsourcing¹³ or semi-automatic processes for the extraction of actors and relationships using methods applied in computer linguistics.¹⁴ Moreover, different methods for visualizing networks are continuously providing new ways of interactively exploring and communicating findings.¹⁵

Despite all these technical achievements, the premise put forward by Konrad Jarausch and Kenneth Hardy still remains valid: “hardware and software are only tools, not ends in themselves. In historical research the intellectual question must always be dominant”.¹⁶

The impact of the earlier-mentioned synergistic fields has turned HNR into a research area that is characterised by a high degree of self-reflection. With this book we hope to initiate a broader discussion about the applicability of social network analysis and digital humanities methodologies in historical research. To this end we have sought to represent the wide spectrum of network-based history research.

About this book

This edited volume presents best practices of HNR as a complement to existing methodological and theoretical manuals. This includes approaches inspired by and adopted from quantitative social network analysis but also more recent efforts to apply network analysis principles to the computational as well as crowd-based creation and curation of research data. We hope that this book will serve as a source of inspiration and a template for new network-based research projects. To guide readers through the diversity of approaches we suggest the following typology of research practices in historical network research:

1. (Re-)construction of historical networks and their analysis
2. Computational extraction of network data from large corpora
3. Infrastructures for data collection and exploration

We will use this typology to break down the case studies presented here:

Computational extraction of network data

Today most applications of network analysis principles in the historical disciplines are inspired by the methodological and theoretical toolkits offered by social network analysis as practiced in the quantitative social sciences. From a historical perspective, the perhaps most influent...

Table of contents

• Cover
• Half Title
• Series Page
• Title Page
• Copyright Page
• Contents
• List of figures
• List of tables
• List of contributors
• Acknowledgments
• 1 Introduction
• 2 (Re-)construction of historical networks and their analysis
• 3 Computational extraction of network data from large corpora
• 4 Infrastructures for data collection and exploration
• 5 Outlook
• 6 Glossary
• Index