Conference Report: Historical Network Research in Chinese Studies

Marilyn Levine's picture

Historical Network Research in Chinese Studies

This article is cross-posted from H-Soz-Kult (10.09.2021)

Veranstalter [Organizers]: Chen Song, Department of East Asian Studies, Bucknell University, Lewisburg; Henrike Rudolph, Department of East Asian Studies, University of Göttingen; Fairbank Center for Chinese Studies, Harvard University

Datum, Ort: 23.07.2021–31.07.2021, digital (Cambridge, MA)

Bericht von [Report from]: Marilyn Levine, Department of History, Central Washington University

Conference Website and Workshop Materials Website (Harvard Dataverse):

The conference, sponsored by the Fairbank Center for Chinese Studies (Harvard University), the China Institute (Bucknell University) and the Department of East Asian Studies (University of Göttingen), was held on two successive weekends. The organizers put together a flawless program that utilized the virtual format which enhanced the introduction of new digital history research and tools, and the engagement of ideas. The conference goal statement covered the origin and range of the conference: “Network analysis is a burgeoning field in East Asian digital humanities. In recent years, the digitization of source materials, the proliferation of databases, as well as the development of digital tools, have greatly facilitated the study of networks in Chinese studies. To promote interdisciplinary dialogue between network scholars specializing in different periods of China and beyond, Henrike Rudolph and Chen Song edited a special issue for the _Journal of Historical Network Research_. Based on stellar contributions to the special issue, this conference showcases recent scholarship that applies the methodology and technology of network analysis to the study of diverse topics in Chinese history.”

The structure of the conference was organized around three main areas. First the conference highlighted several articles in the upcoming China issue of the _Journal of Historical Network Research_ (JHNR). Presenters pre-posted materials for the virtual audience to read and were limited to ten-minute presentations and ten minutes for audience questions. The second activity included demonstrations of digital humanities projects that ranged from textual corpora to biographical databases and to online visualization platforms. The third activity was hands-on interactive workshops for digital humanities beginners and intermediate levels that included using the immense China Biographical Database (CBDB) for network visualization with Gephi;  a workshop on centrality and modularity analysis in Gephi and Visone (based on data from _Records of the Imperial Library in the Southern Song_ and _Comprehensive Survey of Song Dynasty Capital and Court-Rank Officials_), hierarchical clustering and heat maps with Orange data mining tool (using two datasets from the Chinese Biographical Database [CBD] and demographic data from Toussaint LOUA [1873]); and network simulation and ERGM (based on a dataset collected from _A New Account of the Tales of the World_). Each of these workshops centered on tools that audiences could download ahead of time and provided some datasets for practice. Finally, the conference began with a framing of some historical network issues by the organizers along with two keynote addresses that enhanced the intellectual value of the conference. The conference ended with a discussion session. The organizers put together the logistics of the conference with the support of Kwok-leong Tang (Fairbank Center for Chinese Studies, Harvard University), Yuan-heng Mao (Harvard University) as well as Tiange Wang (Harvard Graduate School of Design).

As someone who has participated in digital initiatives and conferences since the mid1990s, to me this conference was a true turning point, almost a generational shift that promoted interdisciplinarity and diverse methods, with creative interpretations from the digital projects. As some examples:

The two keynotes bookended the qualities of the conference. PETER K. BOL (Cambridge, MA) overviewed the dataset, methodologies, and network analyses, including some excellent tables on centralization and superlative visualizations which aligned with the advances he has promoted for many years on the arts of prosopography. Bol introduced theoretical underpinnings and some leading edge works of other Sinologists, several of whom he has mentored. His methods, ideas, and conclusions were elegantly supported by his articulate explanations – not always an easily accomplished objective in digital history. This presentation was a great keynote because in many ways it was an intellectual culmination of the broader conference as a generational passage. Indeed, the conference included several of Bol’s protégées, who succeeded in meeting the high expectations that I have seen him express throughout the years.

The ending keynote by ALINE DEICKE (Mainz) was a great selection, as she exemplified the generation who has been trained in digital scholarship. She discussed her graduate training and research projects. From a network analysis of the keywords and types of digital projects at the Historical Network Research 2021 Conference, to the hills of Troy, to the capabilities of network research publication demonstrated in visuals and tables – Deicke’s keynote address welcomed everyone to the digital field and encouraged collaboration. She demonstrated the potentials of intellectual relevance, interdisciplinarity, and international collaboration for digital research.

The motif of generation particularly is important in historical network research, which as a field is coming into its own. It is noteworthy that graduate students delivered two of the four workshops: XIONG HUEILAN (Leiden) and SHANG WENYI (Urbana-Champaign), both of whom also participated in panels, and one of the project presentations: ALEX MAYFIELD (Boston). Their workshops and the presentation were sophisticated and intuitive, and it bodes well for future students enrolled in courses by these scholars. Secondly, the maturation of the CBDB also exemplified the generational issue as it was shown by robust utilization as a model for several of the papers. Another demonstration of generation is demonstrated by the attendees of the conference. Overall, 679 individuals registered for the conference, with 399 individuals attending, and 153 individuals attending for 200 minutes or more. Optional registration data revealed there were 36 different countries and world regions represented whose viewers were online for more than 30 minutes. Of 177 individuals who gave their professional title, 51 identified as holding an academic post, while 90, (almost twice as many), identified as undergraduate, graduate, or postdoctoral individuals. I believe that the joining together of hands-on workshops, project demonstrations, and almost “lightening rounds“ for the panel presentations were a successful format for engaging our students and is perhaps the most significant accomplishment of this conference, alongside the truly international nature of the presenters.

Important advances in digital research were not just presented, but the presenters encouraged the audience to follow-up and utilize the resources. Resource typologies and methods included textual corpora, online visualization platforms, text analysis, and utilization of leading-edge network analysis techniques (see conference schedule below). Here are three examples:

1. The Early Chinese Periodicals Online Database (ECPO), presented by MATTHIAS ARNOLD (Heidelberg) and HENRIKE RUDOLPH (Göttingen) includes 308 publications that have 134 items, over 55,000 names, and 21 languages that allow researchers to utilize this resource as a dynamic data service. ECPO is located at the University of Heidelberg and has a broad membership of distinguished contributors. In this regard, one of the key elements of research design is interoperability as the presenters elucidated: “Agent records in ECPO can be linked to other datasets like the German Integrated Authority File (GND), OCLC’s Virtual International Authority File (VIAF), Wikidata, DBpedia, and the Chinese online encyclopedia Baidu Baike, and have the potential to reference other databases like the Modern China Biographical Database (MCBD) or even CBDB (until 1911).”

2. The Chinese Historical Christian Database (CHCD), presented by Alex Mayfield, is a resource that examines Christianity in China with a focus on the period between the arrival of Francis Xavier in 1552 to the establishment of the People’s Republic of China in 1949. The white paper, shared during the conference and archived at Harvard Dataverse, goes into detail on the philosophy, technical logistics and data structure of the database, with some great demonstrations of network visualizations of relationships among individuals, religious groups, and institutions. Among the design features are bridging the various religious denominations, allowing for multiple forms of belonging, controlling complex and fuzzy geographies, and ease of querying the database.

3. The panel presentation by ANNE CHAO (Houston) and LIU ZHANDONG (Houston) is one of the best intellectual usages of textual analysis I have seen.   Analyzing two giants of modern China, Liang Qichao (1873-1929) and Chen Duxiu (1879-1942), the authors presented a nuanced analysis of their usage of 30 nationalistic terms as they evolve over three key periods of their lives. The statistical and network analysis of these corpora is incredible as the authors show visualizations that include network modularity changes over the three time periods and illuminating comparative co-occurrence graphs on the global networks of the textual analyses. Their paper went into detail about the implications of these analyses and their pros/cons section in the presentation is insightful. The hallmark of this research team, however, is that they frame both statistical and network analysis alongside intelligent historical explanation. In addition, the visualizations were crisp and comprehensible.

One of the areas that could have gone more smoothly in the conference was the Q&A. Given the hundreds of participants, it would have been difficult to have raised hands. Chen and Rudolph and others answered numerous questions in the reply boxes, as well as the presenters answering orally, but there were too many questions to answer in the allotted times, both in the presentations and in the workshops. Q&A statistics ranged from 38 to 67 questions per workshop day. I would suggest a workshop process to be considered that has 3-4 segments (examining a dataset and why it is important; downloading the program; beginning to use the program; and interpreting the results) with the ability to sign up with small groups so that attendees can ask questions in person. Expecting the attendees to download or examine datasets ahead of time is not optimal in my view, because many might need the workshop to do the work or understand the information.

Likewise, the panel presentations might value in an expanded format with breakout sessions with 3-4 discussion leaders and a series of questions of what should have been most important in that presentation. It also would allow for more questions and cross interests to be explored.

Finally, I would suggest that the workshops and perhaps some of the panels that introduce new programs, in addition to the slideshows, also have a more compact 3–5-pages instruction on how to utilize the program – step by step, with appropriate and perhaps small screenshots so that attendees know they are correctly implementing the program (such handouts were available from some of the workshops).

In conclusion, whether one’s interest is in a discipline like digital history, or Eastern Jin (317–420) to Yuan dynasty (1271–1368) elite marriage and social interactions, or political party formation in the 20th century, or religious practice in the past few millennia, or even viewing the coming generation of young scholars, this conference served as a good opportunity to expand one’s horizons.

Conference overview:

Chen Song (Bucknell University, Lewisburg)/ Henrike Rudolph (University of Göttingen): Introduction to the Conference

Keynote speech

Peter K. Bol (Harvard University): Kinship and Collegiality in Chinese Literati Networks

Matthias Arnold (University of Heidelberg) / Henrike Rudolph (University of Göttingen): Project presentation: Early Chinese Periodicals Online Database (ECPO)

Michael Fuller (University of California, Irvine) / Wang Hongsu (China Biographical Database, Harvard University): Workshop: Network Analysis with CBDB Data

Panel discussion

Shang Wenyi (University of Illinois, Urbana Champaign): The Aristocratic Social Network in the Eastern Jin Dynasty (317–420 C.E)

Xiong Huei-Lan (Universiteit Leiden): Path toward the Top Leadership: A Network Analysis of the Civil Service System in the Early Southern Song (1131-1164)

Marcus Bingenheimer (Temple University, Philadelphia): Buddhist “Influencers“ in Late Ming

Xiong Huei-Lan (Universiteit Leiden) / Chen Song Bucknell University, Lewisburg): Workshop: Centrality and Modularity Analysis in Gephi and Visone

Panel Discussion

Anne Chao (Rice University, Houston) / Liu Zhandong (Baylor College of Medicine, Houston): Network of Words: A Co-Occurrence Analysis of Nation-Building Terms in the Writings of Liang Qichao and Chen Duxiu

Marilyn Levine (Central Washington University): Forged in Europe: Post-WWI Chinese Revolutionary Leaders in Europe

Henrike Rudolph ((University of Göttingen): Transforming Networks and Changing Duties: Tracing Structural Changes in the Jiusan Study Society (1949-1990)

Alex Mayfield (Boston University): Project Presentation: China Historical Christian Database

Marilyn Levine (Central Washington University): Workshop: Hierarchical Clustering and Heat Maps using Orange Data Mining Tool

Clovis Gladstone (University of Chicago) and Jeffrey Tharsen (University of Chicago): Project Presentation: TextPAIR Viewer (TPV) 1.0

Shang Wenyi (University of Illinois, UrbanaChampaign): Workshop: Network Simulation and ERGM

Keynote speech

Aline Deicke (Academy of Sciences and Literature, Mainz): Quo vadis, HNR? Chances and Challenges of Current Network Research in the Humanities

Concluding discussion

Tagungsbericht [Conference Report] Historical Network Research in Chinese Studies. 23.07.2021–31.07.2021, digital (Cambridge, MA), in: H-Soz-Kult 23.09.2021.