Historical network analysis: to the fifth anniversary of the profile international journal

Borodkin Leonid Doctor of History Corresponding Member of the Russian Academy of Sciences, Professor, Head of the Department for Historical Information Science at Lomonosov Moscow State University (MSU) 119991, Russia, Moskva oblast', g. Moscow, ul. Lomonosovskii Prospekt, 27-4 borodkin-izh@mail.ru Other publications by this author     Gasanov Arsenii Alanovich Postgraduate student, Department for Historical Information Science, Lomonosov Moscow State University 119992, Russia, g. Moscow, ul. Lomonosovskii Prosp., 27 korp. 4 quat@bk.ru

Danilov Evgenii Vladimirovich Postgraduate student, Department of Historical Information Science, Lomonosov Moscow State University 119992, Russia, g. Moscow, ul. Lomonosovskii Prosp., 27 korp. 4 random.pc.user@gmail.com

The proposed notes are intended to acquaint readers of "Historical Informatics" with the tasks and content of the international scientific journal Journal of Historical Network Research, existing since 2017 ^[1].

The announcement of the journal notes ^[2] that while interdisciplinary studies of network social relationships/interactions have led to an impressive amount of work in the social and political sciences, as well as increasingly among historians, there has not yet been an international channel of publications devoted to the study of networks in their historical context. This has put scientists interested in historical network research at an extreme disadvantage and has led to the fact that they have long been accustomed to publishing their research papers in "non-historical" journals. The situation for historians interested in network research is further complicated by academic and cultural idiosyncrasy, since most of the groundbreaking and recent research on historical networks in the English-speaking world has been conducted by historical sociologists, not social historians. Thus, they remained largely outside the scope of the traditional academic history departments. Naturally, this has also influenced the channels for publishing research in this area; journals preferred by "networkers", such as Social Networks and the American Journal of Sociology, pay great attention to methodological and theoretical aspects. In short, there has not yet been a platform for international publications devoted to the study of networks (social and other) with specific historical content.

The Journal of Historical Network Research (JHNR) seeks to fill this gap. Its purpose is to publish original works in which theoretical and methodological approaches to social network analysis are applied to historical research, to promote epistemological and theoretical understanding of social network analysis in historical, social and political sciences, as well as to promote empirical research in the field of social networks characterizing historical social interactions. The journal aims to develop interaction between various fields of historical research, social and political sciences, as well as various research traditions and disciplines, while strengthening the dialogue between network research and "traditional" historical research. JHNR serves as a meeting place for the traditional hermeneutics of historical research and its accompanying emphasis on contextualization and criticism of historical sources (as is done in traditional academic historical journals), on the one hand, and theoretical and/or sometimes overly technical discussions of methodological and technological issues (which prevails in publications focused on "pure" or sociological network research) - on the other ^[2].

JHNR Magazine is an open access online publication hosted by the University of Luxembourg. Some issues of the journal are thematic – for example, reflecting the experience of using network analysis methods in the study of antiquity or the history of China. For social media researchers, this publication format has a number of advantages. Unlike traditional printed publications, network illustrations (graphs) are not subject to restrictions on the size or nature of visualization. Data collections or databases may be published together with research papers, and visual representations of networks are not limited to static illustrations, but may also include dynamic/animated graphs and/or timelines, as well as three-dimensional representations of the network.

Almost all articles of the journal contain visualization of constructed and analyzed networks represented by graphs. More often authors use Gephi for this purpose (https://gephi.org ) is an open source software package for visualization and analysis of related data. At the same time, other network analysis tools are used in the journal articles, for example, NodeXL, Ucinet, etc.

The editorial and advisory boards of the journal consist of scientists who have previously published in the field of historical network research, covering all major historical periods, from antiquity and the Middle Ages to modern and modern history. The Advisory Board consists of well-known scientists, experts both in the field of historical sciences and related disciplines (for example, archaeology, social, political and economic sciences, digital humanities and computer science).

The idea of writing this review appeared during the course "Network analysis in historical research", which was read to graduate students and undergraduates of the Department of Historical Informatics of the Faculty of History of Moscow State University in the spring semester of 2021/2022. The following set of brief reviews of a number of articles published in the journal JHNR is intended to give an idea of the topics of publications in this journal, the formulation of network analysis tasks in concrete historical research. The first two reviews were written by A.A. Hasanov, the next two by E.V. Danilov.

Informal networks of the elite of the early Roman Empire: analysis of the letters of Pliny the Younger

The article by Fabin Hermerodt, published in the 4th volume of JHNR, is devoted to social relations in the society of the early Roman Empire ^[3] Using the example of Pliny the Younger, a Roman statesman and public figure and senator, the author traces various types of relationships in Roman society, as well as the process of their folding into complex and contradictory intricacies. The source here is Pliny's letters, quotes from which reinforce the author's narrative.

The article includes three case studies. The first plot concerns the marriage of one of Pliny's clients to the daughter of his friend and patron, where the kinship of houses and the improvement of political positions definitely come to the fore. A network formed by friendly and family ties, as a result, connects Pliny and the Emperor Trajan ^{[3, pp.255-259]}. The second plot examines the case when Pliny defended the daughter of his deceased patron in court. What the author finds interesting in this case is that Pliny's motive here is not to help the girl herself, who is practically not mentioned in the letters, but a debt to the person who once helped the senator. Another important point is that Pliny's opponent in court is also a senator, and therefore they are forbidden to engage in open hostility ^{[3, pp.263-268]}. The third plot is connected with the confrontation between Pliny and his opponent, Marcus Aquillius Regulus, who tried to catch Pliny in court by linking him with people who fell into disgrace as political opponents of the emperor ^{[3, p.269]}.

This study demonstrates the complexity of intersecting friendly, family and political relations among the elite of the early Roman Empire. The author demonstrates that such connections can both advance a person in his career and give him more influence, and create a threat. As the author points out, the network formed around Pliny the Younger included hundreds of people of various origins, and the use of network analysis allows you to visualize the identified complex systems of relationships, extremely difficult to perceive if you do not use the methodology of network analysis.

Such studies are important for understanding the functioning of the public institutions of ancient Rome. Social, political and historical aspects are practically inseparable in this case, and therefore special interdisciplinary approaches are required, such as, in particular, the analysis of social networks. It is worth noting, however, that Pliny the Younger was an extraordinary and outstanding man, and therefore it is unlikely that his example can be transferred to Roman society as a whole. The question of whether Pliny was a typical representative of Roman society is also raised by the author, but does not receive an answer, since the study is limited to the source – Pliny's letters have been preserved and are quite a unique monument of their time.

An important conclusion is also that Pliny's various friends, relatives and acquaintances have their own connections among themselves, and are not scattered clusters united by one figure ^{[3, p.269]}. This can be interpreted as an indication that Roman society during the early Empire was a lively and open social system where connections between individuals played a very important role. Interesting in this case is the further study of the role of social ties in history, its dynamics, differences in different societies.

Informal political communications in Late Republican Rome

The article by Cristina Rosillo-Lopez ^[4] examines personal informal relationships between senators of the late Roman Republic using network analysis methods. Such relations formed a large part of the political life of republican Rome, since personal dialogue was the main way of exchanging information between Roman politicians, and relatively short distances made such communication possible ^{[4, pp.91-92]}.

         Roman politicians often relied on their friends and acquaintances, which was especially evident in their election campaigns, when senators tried to spread their influence in all possible ways – through friends, relatives, acquaintances, friends of friends or debtors ^{[4, p.95]}. Awareness and temporary alliances that senators informally concluded to resolve certain issues (without being associated with permanent organizations or parties) were key factors of success in the political struggle in the Roman Republic ^{[4, pp.92-93]}. Thus, it makes sense to consider the numerous personal relationships between senators from the point of view of network analysis, since this can allow a deeper understanding of the mechanisms of the ancient Roman political system.

         The author presents in the article a number of case studies on several personalities, for which a network of informal connections is traced and corresponding graphs are built. They are Marcus Caelius Rufus, Titus Pomponius Atticus and Marcus Tullius Cicero. The case of Cicero is the most interesting, since his example shows the mechanism of isolation and, consequently, loss of influence by a senator in a crisis situation (the civil war in Rome in 49-45 BC). Cicero was the center of many discussions in January-May 49 BC (although, as the author points out, the centrality of Cicero's figure is put doubted by the fact that the data are based on his letters and speeches), but later he was excluded from all major discussions in the future, which significantly shook his status ^{[4, pp.104-109]}.

         The author points out certain difficulties in conducting network analysis, since it is unclear how to determine the degree of interaction of certain characters and the significance of individual connections for the entire network as a whole ^{[4, p.94]}. We will also note another problem mentioned by the author – the problem of sources, but it would be interesting to reveal it to a greater extent. How complete are the data on senators' meetings? Probably, some of them took place completely secretly. It is also likely that they will be more complete for some senators than for others, which complicates the creation of an overall picture. This makes it clear the author's approach to the study as a set of stories dedicated to specific personalities, but it could be fundamentally interesting to scale the study to a larger volume and create larger and more complete networks.

         Rosillo-Lopez's research allows us to look at the political system of the late Roman Republic from a new angle – network analysis shows its specific nature and reveals some of the principles of its work. It is also important that the peculiarities of a particular political system are emphasized, and not the stereotypes inherent in the modern system are applied.

Understanding the Coastal Trade Community in England in Early Modern Times

In this article, Lianna Brinkley applies network analysis to the study of coastal trade in Tudor England ^[5]. A large number of works in the historiographical tradition are devoted to the study of trade, trade relations and their impact on the economic, social, and political state of a particular society/public education. They use a number of methods, ranging from mathematical and economic to traditional methods of text analysis. L. Brinkley offers an original methodology for creating a prosopographic database of merchants and seafarers and using it to build networks of relations between them - in order to study trading communities and the development of coastal trade in England in the second half of the XVI century.

As the main source for creating such a database, the work uses port books containing data about the ship, its captain, merchant and cargo transported (and some other data). The author examines these data in the optics of network analysis at three levels.

The first level is a study of the commercial nature of coastal trade, in which the emphasis is placed on the study of the network of trading communities, the integration of its members at the intra-urban and interregional levels, as well as economic and geographical factors of this integration (separation into conditionally “rich” and “poor”, “countrymen” and “nonresidents”). The second level is the study of the strategy of the relationship between merchants and captains of merchant ships and the building of a common network of contacts between them (which, in turn, allows for a more detailed study of the integration processes in coastal trade). The third level is the study of the influence of the social, economic, and political statuses of individual members of the community (and its dynamics) on the functioning and transformation of merchant networks.

This article is of great interest due to the fact that the application of the network analysis methodology opens up new opportunities for studying such a complex phenomenon as coastal trade in Tudor England. Studying it as a network of relationships between merchants and other personalities involved in this process opens up new perspectives for the study of economic and social history. Moreover, it contains the potential to study trade relations in the light of institutional theory. In many ways, the very nature of network analysis somehow contributes to the fact that the phenomenon or process being studied appears as a network of interactions and relationships of its members, on the basis of which it is possible to study the institutionalization of these phenomena or processes.

The contribution of this study to the study of the history of England is difficult to overestimate. In addition to the direct study of employment in the field of coastal trade, the socio-economic status of merchants (which can be attributed to the field of business history), the use of network analysis methods allows us to formulate much broader conclusions about the impact of processes related to coastal trade on English society in the second half of the XVI century. The conclusions obtained could hardly be obtained without using network analysis methods.

Geospatial networks of the East German Opposition in 1975-1989/90

Kimmo Elo's article ^[6] examines such a complex socio-political phenomenon as the opposition in the GDR during the last 15 years of the country's existence through the prism of network analysis.

The main source used in the study is a corpus of photographs depicting German oppositionists. The author of the article offers a two-level system of analysis of the East German opposition. The first level is geospatial: the place where the photo was created gives a picture of the boundaries of the physical presence of the oppositionists; the second level is personal: the people present in one photo form a personal network, and the corpus of photographs allows you to build a network cluster that characterizes the relationships within the German opposition.

Such a two-level system of analysis allows us to build a comprehensive picture of the functioning of the opposition environment in the GDR: to identify the boundaries of their geographical presence, to determine the main locations of the opposition, to identify opposition “cores”, mediators, individual regional groups and their locations. In addition, the study of this network in dynamics allows us to study how, for example, the structure of relationships within the opposition network is affected by the activities of the security services (Stasi).

Research by K. Elo is a noteworthy example of how network analysis can be applied in historical and political studies. As the author himself notes, it is not revolutionary in its substantive conclusions - many of the provisions set out in the article are already known in the historiographical tradition. However, such a study offers quite interesting optics, in which the spatial and network nature of socio-political phenomena and processes becomes much more obvious and visual than in a more conventional narrative.

However, such a methodology is not without drawbacks. The application of network analysis strongly depends on the quality of the source database. The formalization of connections in the network is largely due to the quality of data and their availability, and in historical research this problem becomes even more urgent. The metadata of the photos of the corpus used in the study is not complete - the place where the photos were created and the information about the people depicted on them is not always amenable to attribution. The author, however, is aware of these problems, which he reports in the article.

Moreover, the source-oriented nature of network analysis may contain flaws in the study of phenomena in which documenting information about the state of individuals, groups, organizations is not mandatory. The degree of representativeness of the system of relations within the opposition, which is offered by the corpus of photographs, may not be as high as expected by researchers. The photos, presumably, were largely taken on the personal initiative of the oppositionists. In addition, this was probably hindered by the activities of the special services, in which any documented confirmation of participation in opposition activities assumed a risk for the oppositionists themselves.

The article is of particular interest as an example of research using the methodology of network analysis, which is an addition to fundamental historical research.

 The network structure of the conspiracy of July 20 , 1944

Let us briefly consider Joerg Raab's review published in JHNR magazine of Linda von Keyserling-Rebein's book about the assassination attempt on Hitler on July 20, 1944 and the coup attempt to overthrow the Nazi regime and restore the constitutional German state ^[7]. In the title of this book there is a question: "only a narrow circle?" (literally: "only a small clique?"). Thus, the author emphasizes an important aspect of the topic under consideration – was the circle of participants in the upcoming coup really narrow? The research is based on the effective use of concepts and methods of network analysis.

L. von Keyserling-Rebine made a report on this topic at the XV conference of the Association "History and Computer", held in Zvenigorod in 2016. The report aroused great interest both in the content aspect and in the methodological aspect.

Before proceeding to the material of the review of this extraordinary book, let us briefly describe the historical context of the author under consideration. Following the assassination attempt on Hitler at the Wolfschanze headquarters in East Prussia, a coup was launched in Berlin a few hours later with the aim of overthrowing the Nazi regime. Almost everything was prepared, but the plan failed. Hitler survived the assassination attempt, receiving only a few wounds. Many people were persecuted for participating in the coup as participants, supporters or proxies. About 200 of them were executed as a result ^[8]. The official statement of the Reich leadership soon followed that the conspirators were only a "very small clique" that had a strong influence on the public. Speaking about the events of July 20, it is customary to mention Colonel Klaus von Stauffenberg and only a dozen other officers, although historians are well aware that this coup was a collective action in which very different people and groups from different strata of society were involved.

L. von Kaiserling-Rehbein analyzed (using quantitative and qualitative methods) a set of sources on the subject of the study, including reports on Gestapo investigations, indictments, sentences, trial records and other sources from the archives of the Reich. Network visualizations showed that in fact the Gestapo investigators knew about a large network of conspirators – unlike the official version about a "very small clique". The network structures of the constructed connection graphs clearly and in detail showed how different groups of conspirators — for example, army officers, diplomats, social Democrats or trade union members - were connected to each other, what was the structure of intra—group ties.

Analyzing the specifics of the study by L. von Kaiserling-Rehbein, the reviewer of the book, Joerg Raab notes that even 75 years after the assassination attempt on Hitler on July 20, 1944, historians did not come to a coordinated interpretation of the motives and actions of the participants in this event ^{[7, p.118]}.

However, what unites most of the works in the established historiography is a clear emphasis on individual actors and their motives (or on individual groups of actors). As a rule, these works ignore that actors are embedded in networks of relationships that influence their behavior and perception of the situation. In addition to focusing on individual participants in the conspiracy, the emphasis in this thematic historiography is also on the analysis of society's reaction to the events of July 20 and the military component of the coup attempt. At the same time, not enough attention is paid to the issue of the participation of civil servants, representatives of clerical circles, trade unions, as well as politicians; the complex network behind the preparation for the assassination of Hitler and the organization of the coup d'etat, as well as its evolution, remains in the shadows.

Based on available sources (mainly investigation materials), L. von Kaiserling-Rehbein builds a network structure "July 20", including 132 participants who consciously supported the coup plan, and analyzes this network using network analysis methods, also using a hermeneutic approach. Quantitative measurements of the network structure reveal, for example, the important role of reserve officers as intermediaries between conflicting interests, ideological positions characterizing clusters of military and civilian actors. The results obtained show that in no case can we talk about a "small clique", and that the fixation on Stauffenberg as the main participant in the coup attempt does not correspond to historical reality. In his analysis, L. von Kaiserling-Rebein demonstrates that the core of the network included 19 people,

How is this composition and structure of network relations interpreted by the author? In his review, Y. Raab notes that, contrary to the popular point of view, L. von Kaiserling-Rebein does not see here any signs of a democratic-pluralistic orientation of the network. We are talking about the unifying desire of the network participants to avoid the catastrophic defeat of Nazi Germany, to which the course of the war led, to put an end to this war, the fascist regime, to turn Germany into a constitutional state ^{[7, p.120]}.

J. Raab highlights another important aspect of L. von Kaiserling-Rehbein's research related to the "secret" or hidden nature of the formation and functioning of the analyzed network, associated with a compromise between security and efficiency. The establishment of network contacts is absolutely necessary to achieve the common goals of the network structure, which required coordination of actions - despite the fact that each meeting carried the risk of participants being discovered in the conditions of Gestapo surveillance. This problem was solved, for example, through the use of "legitimate" professional contacts, as well as conspiratorial secrecy from accomplices. According to Y. Raaba, the July 20 network offers a very interesting concrete example of analyzing a network containing not only explicit, but also hidden connections. Such studies, as a rule, deal with criminal or terrorist networks, but so far there has been no research on the organization of a network in a totalitarian state, whose participants are the highest circles of the military, politicians, government officials, etc. administration ^{[7, p.121]}.

Concluding the review, J. Raab notes that the study by L. von Kaiserling-Rebein is "an exemplary work on historical network analysis."

* * *

Instead of a conclusion, we will give a judgment on the methodology of historical network analysis expressed in an article by V. Brocard, E. Kestner and K. Rollinger, published in No. 4 of the journal JHNR (2020) ^[9].

The authors emphasize that the analysis of social networks does not replace the traditional methods of source criticism, which is the main principle in the work of the historian. Nevertheless, the analysis of social networks offers an additional methodological and theoretical approach that opens up new perspectives in the field of research, which is mainly dominated by more traditional prosopographic studies, and at the same time provides a powerful tool for analyzing and visualizing social and political connections in historical societies. These advantages are especially noticeable when network analysis is combined with traditional historical research tools. For this approach, the name "Historical Network Analysis" (or "Historical Network Research") gradually became established, which distinguishes it both from a purely metaphorical or theoretical concept of the network, and from network analysis that is not related to the historical context, but is influenced by sociology. The application of historical network analysis is not limited to specific epochs or specific thematic issues. It provides directions and starting points for further interpretations, as well as offers new perspectives ^{[9, pp. vi-vii]}.