“Slow” Network Research? A Mixed-Methods Approach Towards Funeral Status Representation in the Late Urnfield Period

Journal of Archaeological Method and Theory https://doi.org/10.1007/s10816-025-09698-5 (2025) 32:33 RESEARCH “Slow” Network Research? A Mixed‑Methods Approach Towards Funeral Status Representation in the Late Urnfield Period Aline Deicke1,2 Accepted: 21 January 2025 © The Author(s) 2025 Abstract From its earliest stages on, the rise of computational approaches in the humanities—whether in archaeology, history, or digital humanities more generally—has been accompanied by discussions and critical reflections on the way in which datadriven research methods are informed by the representation of research objects as data structures. Various dimensions, challenges, and characteristics can be roughly divided into three intersecting aspects: the subjectivity of data, their complexity, and their size. Archaeological network analysis as a formal, quantitative method is situated firmly within the tension between these fields, and many authors focus on the application of network research to archaeological data while respecting their complex nature. This paper adds to this growing body of work by focusing on the specificities of a medium-sized data set that offers multiple perspectives on a complex question of social archaeology: the study of intersecting social identities and their materialisation in funeral assemblages, particularly of a collective identity of high status-individuals or “elites”, during the Late Urnfield Period. It offers a mixedmethods approach that centres quantitative results and qualitative contextualization across different scales, and minimises loss of information and context, while transparently disclosing its practices of data selection, pre-processing, and analysis. In doing so, it aims to make the reflective positionings of “slow data” and “slow technology” productive for a methodology of “slow networks”. Keywords Network analysis · Data in archaeology · Bronze Age · Mortuary analysis * Aline Deicke 1 Philipps-University Marburg, Marburg, Germany 2 Academy of Sciences and Literature, Mainz, Germany Vol.:(0123456789) 33 Page 2 of 51 A. Deicke Introduction1 In his seminal work on “Reassembling the Social”, Latour identifies “the nature of groups” as a first source of uncertainty for the study of sociology. As he states, “there exist many contradictory ways for actors to be given an identity” (2005: 22) and thus to negotiate their inclusion or exclusion from social aggregates. Evidently, this uncertainty extends into archaeology and may even be amplified, as archaeologists are unable to “follow the actors”, as Latour encourages. However, they can follow “the traces left behind by their activity of forming and dismantling groups” (Latour, 2005: 29)—though of course, these traces consist of only partially preserved material remains, the study of which presents many challenges. Several authors have sought to integrate Actor-Network Theory (ANT) with archaeological networks as a metaphor and as an analytical perspective (Blair, 2017; Knappett, 2005; van Oyen, 2016). Indeed, the ability to identify patterns of interaction between social actors and artefacts, and thus of group associations, without relying on a priori definitions is widely regarded as one of the strengths of network research in archaeology and beyond (Holland-Lulewicz, 2023; Newman, 2018; Terrell, 2013). As one of the main interactions through which social structure of the past may have been organised, the expression of group associations and identity in funeral assemblages has been the focus of numerous studies in recent years.2 Yet, identity or identities appear as “multiple and hybrid” (Díaz-Andreu & Lucy, 2005: 2), “fluid and relational” (Brück & Fontijn, 2013: 204), overlapping and conflicting. To map out these “contradictory ways” via formal, quantitative methods presents a challenge that stems from a fundamental tension within the field of computational archaeology: between the affordances and requirements of formal analysis, and the “shadowy nature of archaeological data” (Wylie, 2017: 204). Archaeological Data and Formal Analysis From its earliest stages, the rise of computational approaches in the humanities— whether in archaeology, history, or digital humanities more generally3—has been accompanied by discussions and critical reflections of the manner in which datadriven research methods are informed by the way research objects are represented as 1 This paper builds on, refocuses and revisits some of the research presented in my dissertation, originally published in German in 2021 (Deicke, 2021). Several of the analyses included in this paper have been revised to reflect an improved methodological approach. The author thanks the organizers and participants of the “Connected Past 2021” conference on archaeological network research for their helpful discussions and feedback. 2 In defining the term “social structure”, I follow Martina Löw in contrasting it with “social structures”. The latter construct the former through various interactions (Löw, 2016). 3 While undoubtedly, archaeological data has its very own set of particularities, in the following paragraphs, I want to extend the discussion to the field of digital humanities and digital history as well, as I believe these fields deal with many of the same challenges and have valuable insights to offer into the datafication of archaeological phenomena. “Slow” Network Research? A Mixed‑Methods Approach Towards… Page 3 of 51 33 data structures. Beyond “shadowy”, data have been called “messy”, “dirty”, “tidy”, “smart”, “characterful”, “big”, “small”, “slow”, “mindful”, “data with depth”, and more (Cooper & Green, 2016; Crawford, 2013; Huggett, 2015, 2022; Lemercier & Zalc, 2019; Lincoln, 2020; Schöch, 2013; Wickham, 2014), addressing the various dimensions, challenges, and characteristics of the “elemental building block from which information is derived” (Huggett, 2022: 98). For the purposes of this paper, these dimensions can be roughly divided into three intersecting aspects: the subjectivity of data, their complexity, and their size. The idea that data are not objective representations of (historical) reality has a long tradition; however, the available space does not permit a full discussion in this paper. Nevertheless, Bowker’s famous statement of raw data as an oxymoron (2005), followed by Drucker’s conception of data as “capta” (2011) certainly stand out as two of the commonly cited points of reference. As Huggett emphasises, data are “theory-laden, process-laden, and purpose-laden, created by different people, under different conditions, for different purposes, at different times” (2022: 103). This concept resonates with the phenomenon of “data cultures” as postulated by Acker & Clement (2019)— situating the elements of human agency and social practice as vital components of the modalities and circumstances involved in the encoding of cultural data. Against this background, the notion of “messy” or “dirty” d (...truncated)