Archaeology is a cumulative discipline, reliant on the establishment of regional chronologies and typologies to provide context for understanding past human communities (Nicholson et al. Reference Nicholson, Kansa, Gupta and Fernandez2023; Wobst and Keene Reference Wobst, Arthur, Gero, Lacy and Blakey1983). Archaeology is also unique in its epistemology, which entails the generation of “long data” through the controlled destruction of the primary record to produce a deep history of what it means to be human (Arbesman Reference Arbesman2013; Huggett Reference Huggett and Andrew2015:13; Shryock et al. Reference Shryock, Smail, Earle, Feeley-Harnik, Fernández-Armesto, Gamble and McMahon2011). The simultaneously destructive and incremental nature of archaeological knowledge increases the disciplinary benefits of the open research movement, which calls for the open access, use, modification, and sharing of knowledge. Recognizing this potential, archaeological work on open research practices has targeted specific components of openness, such as open data (Marwick and Pilaar Birch Reference Marwick and Birch2018), open methods (Marwick Reference Marwick2017; Palmisano and Titolo Reference Palmisano and Titolo2024), open source software engineering (Batist and Roe Reference Batist and Roe2024), preprints (Marwick Reference Marwick2020; Queffelec et al. Reference Queffelec, Maureille, Arzarello, Blasco, Crandell, Doyon and Halcrow2023), data reuse (Atici et al. Reference Atici, Kansa, Lev-Tov and Kansa2013; Faniel et al. Reference Faniel, Kansa, Kansa, Barrera-Gomez and Yakel2013), replicability (Marwick Reference Marwick2017), reproducibility (Farahani Reference Farahani2024; Karoune and Plomp Reference Karoune and Plomp2022; Marwick Reference Marwick2025), digital infrastructure (Huggett Reference Huggett2012, Reference Huggett and Andrew2015), and the deployment of guiding principles (Kansa et al. Reference Kansa, Gupta and Martinez2025; Lien-Talks Reference Lien-Talks2024; Nicholson et al. Reference Nicholson, Kansa, Gupta and Fernandez2023). These studies demonstrate that despite the advantages of openness for archaeology, adoption of open research within the discipline is at best scattershot and at worst begrudging or nonexistent.
Qualitative research that takes a fine-grained approach to individual attitudes and practices has the potential to resolve the paradox of why archaeologists lag behind existing guidelines for best practices despite the many benefits of openness for disciplinary scholarship. To date, however, there are only a handful of qualitative archaeological studies, and they are focused on data reuse. For example. Faniel et alia (Reference Faniel, Kansa, Kansa, Barrera-Gomez and Yakel2013) showed that archaeologists’ nascent data reuse practices were hampered by a lack of methodological standardization and the need for more detailed dataset provenance information. Similarly, Lien-Talks’s (Reference Lien-Talks2024) investigation of the alignment between the FAIR principles and current data reuse practices in biomolecular archaeology revealed the underuse of persistent identifiers and data management plans, variability in metadata requirements and preferred file formats, and a need for improved documentation, license specifications, and data citation.
The current study complements this growing body of scholarship by using semi-structured interviews to generate qualitative data and answer key questions about open research practices in archaeology. How do archaeologists define and understand open research? What are the perceived costs and risks of openness for practitioners? How does data diversity affect the adoption of open research practices? How do research evaluation systems and resources structure decisions about open research? Finally, what kinds of support could expand disciplinary participation in open research?
This analysis is part of the project “Cultivating an Open Research Culture in the Humanities and Social Sciences,” which is funded by the Irish National Open Research Forum. The project examines the principles and best practices of open research in the humanities and social sciences, the alignment between the principles of open research and research assessment criteria, and the rewards and recognition systems supporting open research. Archaeology was selected as a case study to interrogate the dynamics of open research in an interdisciplinary field that draws methods and theory broadly from across the sciences, social sciences, and humanities.
In addition to archaeology presenting a compelling data point for information studies scholars investigating open research dynamics, our study has important ramifications for archaeologists. Kansa (Reference Kansa2012:507) contends, “One would be hard pressed to overstate the ethical and policy case for greater data transparency in archaeology.” Identifying the obstacles that archaeologists perceive as preventing their implementation of open research practices offers insights for practitioners and institutions about how to expand participation within our discipline. Through developing an understanding of how, when, and why practitioners adopt or avoid open research practices, this article contributes to ongoing conversations about openness, knowledge production, and epistemology within archaeology.
Background
The Open Definition explains that “knowledge is open if anyone is free to access, use, modify, and share it—subject, at most, to measures that preserve provenance and openness” (Open Knowledge Foundation 2015). Open Archaeology incorporates the pillars of open access, open data, and open methods to support data stewardship, transparency, and public involvement (see Marwick et al. [Reference Marwick, Barton, Bates, Elizabeth Bollwerk, Carter and Conrad2017:8–11] for definitions of each pillar). The existing literature on openness in archaeology can be sorted into three major categories: state-of-the-field overviews, how-to guides, and case studies of practice.
The most accessible review of disciplinary open research practices comes from a 2012 special issue of World Archaeology, which aimed to bring conversations about open research to a wider audience. The forum explored topics such as long-term data preservation, open source software, public archaeology, digital archaeology, citizen science initiatives, semantics, ontology, and open ideology (Lake Reference Lake2012). Attention to these issues was instigated by a sense of “data deluge” in a field that had rapidly become reliant on technology and “born digital” data (Bevan Reference Bevan2015). A welter of conflicting opinions surrounded the advent of “Web 2.0” (Limp Reference Limp, Kansa and Kansa2011).Footnote 1 Imagined futures of the research dynamism and public engagement that would be facilitated by an emerging universe of “linked data” (Beck and Neylon Reference Beck and Neylon2012; Morgan and Eve Reference Morgan and Eve2012) were tempered by recognitions of existing constraints, including critical mass requirements, infrastructural limitations, adoption costs, the restrictions of professional incentive structures, and the difficulty of generating semantic interoperability (Limp Reference Limp, Kansa and Kansa2011). Digital curation specialists voiced concerns about the underappreciated demands of preservation and archiving in light of the speed and unpredictability of technological innovation (Jeffrey Reference Jeffrey2012). Similar reservations were expressed about the fixation on new technologies at the expense of the collective development of shared standards for data content, documentation, and ontology (Huggett Reference Huggett2012, Reference Huggett and Andrew2015). Practitioners also recognized that painstakingly digitized archaeological datasets were underused and that “the research impacts of shared data are usually more speculative than realized” (Kansa Reference Kansa2010:15; see also Huggett Reference Huggett and Andrew2015).
Fifteen years later, archaeology has still not been able to substantively move the needle on open research. As Nicholson et alia (Reference Nicholson, Kansa, Gupta and Fernandez2023:64) decry, “Only a small fraction of the data from recent projects—and an even tinier proportion of data from legacy projects—are now publicly findable or accessible, let alone interoperable or reusable.” Yet, even though the discipline is lagging behind best practices guidelines (e.g., Beck and Neylon Reference Beck and Neylon2012; Kansa et al. Reference Kansa, Gupta and Martinez2025; Karoune and Plomp Reference Karoune and Plomp2022; Marwick et al. Reference Marwick, Barton, Bates, Elizabeth Bollwerk, Carter and Conrad2017), a dedicated cohort has continued pushing for change by crafting how-to guides explaining the ins and outs of open research practices. The earliest primers, which introduced the Open Context platform (Kansa Reference Kansa2010; Kansa et al. Reference Kansa, Kansa, Burton and Stankowski2010), have been followed by summaries of open science principles (Marwick et al. Reference Marwick, Barton, Bates, Elizabeth Bollwerk, Carter and Conrad2017), overviews of data sharing and citation (Marwick and Pilaar Birch Reference Marwick and Birch2018), introductions to preprint repositories (Marwick Reference Marwick2020), examples of reproducible workflows (Karoune and Plomp Reference Karoune and Plomp2022), and the development of Peer Community in Archaeology, a platform for preprint peer review (Queffelec et al. Reference Queffelec, Maureille, Arzarello, Blasco, Crandell, Doyon and Halcrow2023). Archaeologists have also outlined open data principles such as FAIR (“Findable, Accessible, Interoperable, Reusable”) and CARE (“Collective benefit, Authority to Control, Responsibility, Ethics”); for further details see Carroll et alia (Reference Carroll, Garba, Figueroa-Rodríguez, Holbrook, Lovett, Materechera and Parsons2020) for CARE and Wilkinson et alia (Reference Wilkinson, Dumontier, Aalbersberg, Appleton, Axton, Baak and Blomberg2016) for FAIR. These principles offer definitions of key terms, identify disciplinary benefits, outline institutional mandates and capacities, evaluate current practices, and chart actionable changes (Gupta et al. Reference Gupta, Martindale, Supernant and Elvidge2023; Kansa et al. Reference Kansa, Gupta and Martinez2025; Lien-Talks Reference Lien-Talks2024; Nicholson et al. Reference Nicholson, Kansa, Gupta and Fernandez2023). A number of short publications likewise have introduced open databases that clean, standardize, and consolidate specific types of archaeological data such as stable isotope values or radiocarbon dates (Bird et al. Reference Bird, Miranda, Linden, Erick Robinson, Nicholson and Capriles2022; Loftus et al. Reference Loftus, Mitchell and Ramsey2019; Plomp et al. Reference Plomp, Stantis, James, Cheung, Snoeck, Kootker and Kharobi2022).
In tandem with the increase in targeted how-to guides, archaeologists have published case studies to illustrate the process of creating and using open research. Morgan and Eve (Reference Morgan and Eve2012) described the successful incorporation of digital archaeological outreach in the Prescot Street project, which combined dedicated data-entry staff, hybrid paper and digital recording, blogging, interim reports, and a project website to provide real-time, publicly accessible updates on excavation progress and interpretations. Atici et alia (Reference Atici, Kansa, Lev-Tov and Kansa2013) grappled with the realities of open archaeological data by asking three zooarchaeologists to analyze an “orphaned” faunal dataset and comparing their interpretations. Cook et alia (Reference Cook, Çakirlar, Goddard, DeMuth and Wells2018) presented three examples of integrating data reuse and digital literacy into teaching, providing recommendations and reflections about the process. Marwick (Reference Marwick2017) outlined the core principles animating computational reproducibility and charted a step-by-step example, providing the reproducibility “blueprint” underlying an existing publication. Other case studies examined open research practices in specific journals (Marwick Reference Marwick2025; Palmisano and Titolo Reference Palmisano and Titolo2024) or national contexts (Solheim et al. Reference Solheim, Roalkvam, Ødegaard, Loftsgarden, Damlien, Vatsvåg Nielsen and Sand-Eriksen2025).
Methods
We conducted semi-structured interviews with 28 respondents to assess academic archaeologists’ perspectives on and experiences with open research (Table 1). We recruited participants via email, institutional channels, and professional networks through a combination of purposive and snowball sampling. Respondents were drawn from the higher education sector, although some also had experience working in contract archaeology. They were employed in Ireland (n = 15; 54%), the United Kingdom (n = 11; 39%), and the United States (n = 2; 7%) and represented a range of career stages, including PhD students (n = 10; 36%), postdocs (n = 7; 25%), and early career, mid-career, and senior faculty (n = 11; 39%). Although our US sample is small, it expands the available perspectives for graduate students and facilitates the recognition of concerns that characterize this career stage regardless of national background.
Career Stages, Country of Employment, and Numeric IDs of Respondents.
Table 1 Long description
Numeric ID entries are paired with a career stage and a country of employment. IDs 1, 2, 3, 6, 7, 8, 9, 10, 11, 12, 13, 14, 16, 17, and 27 are in Ireland, including one PhD student, three postdocs, and eleven faculty. IDs 5 and 15 are PhD students in the United States. IDs 18 through 26 and 28 through 29 are in the United Kingdom, including seven PhD students and four postdocs. Faculty appear only in Ireland, while postdocs appear in Ireland and the United Kingdom. PhD students appear in Ireland, the United States, and the United Kingdom.
Note: One respondent initially included in the research design was dropped from the study, but the original numbering system for anonymized IDs was maintained. As a result, there is no Respondent 4.
To target academic archaeologists, we initially contacted potential respondents using their email addresses on their professional webpages and then followed up on their suggestions of others to contact (snowball sampling). Zoom interviews took place between August 2024 and January 2025. Each interview lasted from 45 to 60 minutes and was audio-recorded with participants’ consent and transcribed verbatim while anonymizing identifiers in advance of data analysis. All recruitment and interviews were conducted by the second and third authors (who are not archaeologists), and respondents’ identities remained anonymized when the first author (who is an archaeologist) analyzed the transcriptions. This structure provided space for respondents to speak openly without being influenced by the involvement of another archaeologist.
The general interview structure is provided in Table 2. It offered a flexibility that allowed the interviewers to explore particular questions while maintaining a structured analytical framework. At the beginning of each interview, the interviewer reiterated that the project was not advocating for or against open research but was surveying researchers’ opinions and experiences.
Guidelines for Interview Questions.
Table 2 Long description
Three columns list Number, Category, and Question(s) for six interview prompts about Open Research. Row 1, Meaning and valence, asks what Open Research brings to mind, what it means or entails, whether it is good or bad, and possible unintended consequences. Row 2, Experience, asks about personal experience and includes follow ups for yes or no, whether it was required or a choice, whether it is more labor intensive, and what challenges were encountered. Row 3, Value, asks whether there is a difference in value placed on traditional research practices and publishing choices versus Open Research practices and open access publishing, with a follow up about changing academic values. Row 4, Support and training, asks what support should be put in place and includes follow ups about training for those with or without Open Research experience. Row 5, If research is humanities focused, prompts discussion of open data and a proposal that all data be open, including whether that suggests making notes available. Row 6, If early career researcher, prompts concerns about publications and open data, including where to store data, standards for metadata, training needed, whether openness is for researchers or the public, and dangers of opening everything to unqualified opinion.
Note: Questions 5 and 6 were contingent on the disciplinary affiliations and the career stage of the respondent.
After all interviews were transcribed, we analyzed the data using standard techniques for thematic analysis (Braun and Clarke Reference Braun and Clarke2022; Ryan and Bernard Reference Ryan and Bernard2003; Yin Reference Yin2016). We did not use preexisting frameworks or categories in the analysis; instead, the themes emerged inductively from the interview data. Thematic analysis involved four key components. First, researchers familiarized themselves with the interview data by reading and rereading the transcripts, identifying themes that recurred in conversations about archaeology and open research, and following the process outlined by Yin (Reference Yin2016:196) for Level 1 codes. Next, authors grouped themes into seven broader meta-themes, equivalent to Yin’s Level 2 codes (see Supplementary Material 1). Once themes and meta-themes were identified, each transcript was subjected to a close reading. Quotes pertaining to a specific theme were copied into a parent document for that theme. When a quote referenced multiple themes, it was categorized under multiple themes. After this initial coding process was complete, the fourth stage of analysis entailed rereading the repository of quotes under each theme to examine patterns, linkages, coherence, and dissonance.
Respondent quotations in the results and discussion sections were lightly edited—discourse markers and verbal repetitions were removed—to increase their clarity. Parenthetical numbers in parentheses after quotes indicate the anonymized numeric IDs of respondents, which are categorized by career stage and nation of employment in Table 1. We used the pronoun “they” for all respondents.
Results
How Do Archaeologists Define and Understand Open Research?
When asked to define open research, respondents foregrounded accessibility and availability of data, publications, and methods as key components. Forty-three percent highlighted accessibility to the public, and 25% foregrounded availability to researchers within and beyond academia. Openness was framed as a complex and multistage undertaking with many possible permutations: “It’s an all-encompassing term, and it’s not just making your academic paper accessible, but also the data [and] the descriptive programming that comes with it. In this sense, it’s kind of this whole pipeline, where everything is open” (#25).
Scholars underscored their ethical obligations, with open research identified as a pathway toward “democratizing knowledge” (#2), and “a great leveler” that offset institutional inequities and countered the “bad rep” of gatekeeping in academia (#11). Respondents also highlighted the link between openness and funding, explaining that “if you’re publicly funded . . . there’s a responsibility and a right to make that data open” (#29) and arguing “there’s no point in doing research and recognizing that our research is funded by taxpayers if it’s not open” (#17). Open data were also perceived as a form of reciprocity: “A lot of my work has actually been based on not just generating data, but implementing already available data and already available databases, and whenever I publish something, you know, I put all the data out there as well” (#11).
On the whole, respondents exhibited positive attitudes toward open research, explaining that it “makes things easier” (#19, #24) and “just makes sense” (#17). There was, however, some debate about whether archaeologists valued open research for moral reasons or because of more self-serving interests. One respondent claimed that “a lot of archaeologists would subscribe to the view that the past belongs to, to everybody and so . . . there’s almost a moral imperative to make that data available to everybody for reuse” (#17). Others, however, pointed to the connection between open research, accruing citations, and the performance of a particular kind of scholarship:
That’s why most of us become researchers anyway; it’s because . . . we want to be famous, so we want our research to be read by as many people as possible, and to get it out there as much as possible (#2).
It’s almost a cliché now, that you have to mention something like [collaboration or public impact] in a talk, right? Just to show that you’re a good person that does collaborative research (#3).
Archaeologists also expressed a keen awareness of the relationship between research funding and the responsibility to make research open. Receiving public financial support (framed variously as public, government, national, state, or taxpayer funding) required opening up the results of that research. One postdoc explained, “There’s also a civic responsibility as well, in that if we’re getting funding from national funding agencies to carry out the research, essentially that’s research that’s paid for by the taxpayer. . . . So it’s not to be this extractive thing that we just take money from them, and then we hide our research results behind paywalls” (#2).
Another professor underscored that “the public have the right to see what their money’s doing . . . and have access to the results. So that’s a no brainer” (#10). Although public funding was described as a “privilege” by multiple respondents, there was also a tension concerning whether the openness of publicly funded research was really a choice. Some described openness as a “civic responsibility”(#2) or “the ethical or moral thing to do”(#25), whereas others characterized it as “mandated”(#10), positioning the public as having a “right” (#10, #29) to taxpayer-funded data.
Regardless of whether open data was framed as a choice or an obligation, respondents identified numerous benefits for research, researchers, accessibility, and inclusivity. Openness was understood to facilitate cumulative scholarship, make interpretation more robust, and provide the transparency that ensured researchers would not “reinvent the wheel” (#5). This transparency was also beneficial for public communication. One professor stated, “If we want people to care about the past, . . . we have a responsibility to make good information and reasoned conclusions available so people see how it is done” (#27). Another postdoc emphasized that having a database of high-risk archaeological sites readily available made it easy for their research team to coordinate with local governments to prevent looting. For individual scholars, open research allowed for efficient access to scholarly work while providing positive citational benefits, both for authors who made their own work open and for scholars who wanted to credit the producers of open data.
What Are the Perceived Costs and Risks of Open Research?
Costs
In addition to these benefits, respondents also identified three categories of costs associated with open research (Supplementary Material 2). First, interviewees emphasized the amount of time required to conduct rigorous archaeological analyses, clean “messy” data, and prepare data for publication:
It’s an additional level of hassle that a lot of people maybe don’t have time for (#9).
The multiplier effect in terms of the amount of time and energy [poorly curated data] costs everybody else—whoever wants to enter that data—is so large (#17).
Open research was one of many responsibilities competing for interviewees’ time and attention. Such conflicting demands were evident at even the earliest career stages; one PhD student justified their reluctance to make their data open by explaining, “It is just . . . doing it. A PhD is busy” (#1), and a faculty member contended, “It’s a full-time job almost to keep up with your peers’ research” (#8).
Respondents also highlighted the financial costs associated with open research, identifying two main expenses: article processing charges (APCs: one-time fees covering the editing, formatting, and hosting needed to make publications open access) and fees associated with data storage and hosting. Although estimates of APCs ranged from a few hundred dollars to 12,000 euros, there was broad agreement that open access was “prohibitively expensive” (#6; Supplementary Material 3), a phenomenon linked to large, for-profit publishers. One postdoc historicized these dynamics: “It’s the [19]80s when you see this shift in journals and impact factor and so on. It’s the commoditization of research, and I wasn’t alive then, nor part of it, but we’re all complicit” (#25).
Regardless of perceived cost, most felt that archaeological funding could be better spent elsewhere. When asked about openness mandates as part of grant funding, one PhD student asserted that “people would rather spend that money on doing more work, getting more data, digging more sites” (#1). A postdoctoral scholar was more acerbic: “We have cleaning ladies and dinner staff and casualized zero hours lecturers who are being very, very seriously underpaid, and they sometimes can’t make ends meet . . . and here we are, spending literally thousands on making open access work which will not be accessible . . . to the vast majority of the population anyway, and which is already accessible to those people who need it” (#27).
Several respondents also underscored the expense of storing 3D and photographic data. One professor indicated that “having infrastructure to make [raw data available for catalogs and databases] is difficult too. Someone’s paying for it somewhere” (#27).
The third cost associated with open research was labor. This work included collecting, cleaning, organizing, sharing, and curating data; developing explanations for how data were collected and analyzed; cleaning and explaining code; seeking permissions; developing and managing open research training; and creating and maintaining websites. Cleaning data was framed as a particularly thankless task:
That’s sometimes why people are afraid to . . . look at these collections that aren’t as well documented, because there’s so much translating and so much dirty work they have to do before they can actually do what they’re interested in (#15).
It can end up being a lot of work that you’re doing that just sort of becomes invisible (#23).
Respondents also stressed the time and energy required to learn the skills essential for making research open, including building and managing databases, using platforms like GitHub, becoming familiarized with data formats and compatibility, and understanding General Data Protection Regulation (GDPR) requirements.
Risks
Data scooping, or “having someone else claim priority, usually through publishing, to a research idea or result you yourself have been working on” (Laine Reference Laine2017:1), was the most frequently identified open research risk. Some respondents framed this as an actual risk, whereas others identified it as a perceived risk that engendered territoriality:
I’ve known cases where people that were reviewers on papers that had all the supplementary data . . . decided to say something about it in their own paper before the other one came out (#6).
It seems to me a lot of time fear-based, so, “I am afraid if I share that, then it will help others to advance, that other people can piggyback off my hard work, and they will get some kind of benefit from it that I might consider unfair because I did the work” . . . it’s quite conservative, it’s traditional, it’s old school. It’s very much that thing of—“this topic, or this subject, this research is mine” (#17).
Concerns about data scooping were evenly divided among archaeologists at different career stages, suggesting that this perceived risk animates open research decisions for many practitioners.
Another commonly identified risk was the misinterpretation of results through access to raw data. Within academia, the phenomenon was framed as a strategy related to jockeying for power and prestige. One postdoc noted, “There’s unethical people in every field” (#6), and a PhD student expressed resignation: “There’s always going to be bad actors in any field, so you can’t really do much about that” (#19). Archaeologists described a similar double bind when it came to the general public accessing data. On the one hand, making data open could help combat misinformation and push back against the conspiratorial claim that “Big Archaeology” (#3, #26) was hiding information. Some respondents thus positioned public communication and transparency as a core disciplinary responsibility:
Maybe never before has it been more important for archaeologists to be publicly engaged and open about what they do and how they do it. . . . The amount of pseudoarchaeology that’s out in the world . . . potentially has negative impacts on how we think about human history (#3).
You are preaching to the choir constantly, when the reality is we should really be talking to people who, for example, don’t believe in science, or people who think that archaeology or other sciences are useless, or they are manipulated by politics and things like that (#5).
On the other hand, respondents recognized that archaeological data could be leveraged to promote pseudoarchaeological narratives. One postdoc put it this way: “We’ve been living in a post-expert landscape for a decade now” (#26). Public reinterpretations of archaeological data were often inherently political, with multiple faculty members pointing to archaeogenetic data as being particularly vulnerable to wilful misinterpretation. In the words of one professor, “I might get the interpretation of these sculptures wrong, but no one dies and no one will be beaten up over it. . . . [aDNA data] should be kept away from the public because it is actually dangerous. . . . And the danger with the ancient DNA studies is that they, as scientists, use the ancient DNA to make cultural statements and that’s extremely dangerous, that’s a big issue, that’s a terrible issue” (#13).
Other reported risks were discipline-specific, such as the potential for archaeological sites to be looted if geospatial information were made public and the ethical responsibility to be cognizant of the needs of descendant or local communities affected by archaeological research.
How Do Data Affect the Adoption of Open Research Practices?
Standardization requirements for data and metadata in archaeology were perceived as either limited in scope or nonexistent. As one postdoc framed the situation, “Everyone’s got their own recording system, and it’s idiosyncratic” (#6). Interviewees pointed to the diversity of subfields within archaeology as an impediment to standardization, because “there is no one size fits all” (#5) or “boilerplate standard” (#6) for open data. In this regard, archaeology was contrasted with STEM disciplines such as chemistry, geology, biomolecular studies, genetics, genomics, and geospatial analysis where open data and code were perceived as mandatory, with robust “industry standards” (#8) and supporting infrastructures. One postdoc explained,
In my experience with working with chemists and geologists, when it comes to open research they are much better primed and more comfortable with making data available, with data being standardized across the board, with standardization of what you report as data. . . . People in archaeology, from the more traditional humanities backgrounds, maybe it’s more of a steep learning curve, because we have very diverse types of datasets. I’m not sure that people are naturally accustomed to making [data] available and sharing it (#2).
When identifying obstacles to standardization, multiple respondents described how the same chronological periods can refer to different timespans in different regions; for example, “The Bronze Age in England starts at a different time to the Bronze Age in Northern Pakistan starts at a different time to the Bronze Age in Wales” (#29; see also Faniel et al. Reference Faniel, Kansa, Kansa, Barrera-Gomez and Yakel2013:301). Others pointed to disciplinary diversity, explaining how different specializations—including archaeometallurgy, environmental archaeology, zooarchaeology, bioarchaeology, geoarchaeology, geospatial analysis, and radiocarbon dating—have distinct data requirements. One professor described this variety as “bewildering,” noting how “if you ask a hundred archaeologists what datasets to work with, they might come up with a hundred different answers” (#9).
There could also be marked variability in the quality and quantity of open data within a single specialization. As one PhD student outlined, “Your entire project could be on analyzing the skeleton of just one individual, or it could be analyzing the scale of materials from vastly different cemeteries across the landscape, and that’s just within one single method” (#18).
A faculty member (#10) summarized this dilemma as needing to build databases that were broadly usable while still making data entry simple enough to ensure compliance. Broadly usable databases, however, were few and far between. Two archaeologists at different career stages used the metaphor of translation to describe the process of using or interpreting archaeological data. A PhD student described the need for translation:
Even in the field, a lot of historical archaeologists will dig in inches, but prehistoric sites [are usually excavated in layers]. . . . There’s a lack of standardization that does impact things; we have extra hurdles. When we try to establish comparability, when we try to just even think about different sites, there’s all this translating that you need to do, and preparation, before you even are able to engage in some of that higher level theory (#15).
Another postdoc took umbrage with funders’ perception that archaeological data could be uploaded as a “big static resource,” noting,
You haven’t added any value to the data by digitizing it. If you’re doing that, you may as well just [make a] pdf, scan those paper records, and just stick them up. . . . So it’s sort of its translation, the translation of the archive, and the translation of the information into the digital platform. . . . It’s not just creating digital data for digital data’s sake, or taking paper records and just making them digital with no kind of translation (#29).
A PhD student likewise outlined the minutiae that needed to be documented before interpreting archaeological science data: “There is so much knowledge that goes into ‘What is the material like? How does the machine work? What are the conditions of the machine on that particular day?’ So there are a lot of variables that could affect the results, right? It’s not just like you throw it there, copy, paste, done” (#5).
The ubiquity of what might be termed “translational labor” was one reason archaeologists were pessimistic about the implementation of standards. Attempts at standardization were time consuming and poorly incentivized, and they could produce “patchy” (#8, #11) outputs. One professor asked, “What infrastructure is available to us that can bolt all of these things in” (#16)? Not all respondents viewed standardization as appropriate, foregrounding its proclivity to gloss over error ranges and other methodological caveats.
How Do Research Evaluation Systems and Resources Structure Decisions about Open Research?
Respondents identified two components of research evaluation systems as instrumental in structuring decisions about open research, particularly when it came to publishing: (1) standards for hiring and promotion and (2) the agency afforded by job security at later career stages.
First, interviewees emphasized the impact of hiring and promotion standards on choices about time budgets and research outputs. Institutions were perceived as valuing prestige and citations more than open research, community outreach, citizen science, or impact. One postdoc bluntly explained, “If the job market doesn’t value [public research or community engagement], then people won’t do them, or the people who do them won’t survive” (#3). Instead, control over data and resultant publications were seen as a “linchpin” (#9) and “springboard” (#17) for obtaining an academic position. Another postdoctoral researcher asserted, “If you want to make a career in academia, you need to publish as much as you can, and get in as much money as you can” (#7).
This relentless focus on productivity also affected students’ understanding of the discipline. Describing a field project led by a pre-tenure colleague, one PhD student recounted the constant pressure to be producing articles: “It was a busy season, and we were exhausted, but we also had to be continuing our writing so we’re not behind. Like my colleague . . . was preparing materials because he’s trying to go on a tenure-track. And he’s like, ‘Yeah, I need to get this article out.’ . . . There’s always a hustle (#15).
Universities and funding agencies were “complicit” (#13) in intensifying these demands, and academics were “cynical” (#2) about outcomes, with students and faculty highlighting the negative ramifications of publishing pressures on scholarship. One professor described how, early in their career, their team would “drip [data] out so that we had more to keep going with our papers.” They also said, “I know lots of colleagues who still do that, who build up a big core dataset and do not share it, because, basically, you can make a career out of it” (#14).
These publishing pressures varied with career stage. Junior scholars were required to be ruthlessly strategic and understand the trade-offs inherent in their professional choices. Postdocs depicted early career scholarship as a zero-sum game:
If you’re on the academic pathway, anything that’s not an academic book or a journal article, well, you can you can do it if you want, but no one cares. It’s not really part of any assessment and in short, . . . if something doesn’t help you with your career, then effectively it hinders it, because you’re using up time that you could be using on something else (#28).
If my next position depends on me publishing in high-impact journals, then I’m not willing to sacrifice my own livelihood for the sake of open access publishing (#2).
Faculty acknowledged the link between job security and professional agency. One senior professor explained, “If you’re a junior staff member, and you’re on the promotion track, you’re always thinking about the metrics . . . whereas I don’t need to think about the metrics” (#13). Another acknowledged that pressures have increased in recent years, noting “the volume that is now required . . . simply wasn’t around before” (#10).
Respondents also emphasized the impact of job security and career stage on professional decisions. Interviewees perceived professional expectations as having shifted, with early career scholars today needing to do more work or different kinds of work to succeed. This perception ranged from the acknowledgment that research pressure for past cohorts “wasn’t as bad as it is today” (#5) to the certainty expressed by one postdoc that “it’s just kind of a joke . . . that people who went for jobs in the past would not be competitive today” (#25). This understanding was voiced by respondents in all career stages, with one professor noting that their own PhD students “have to be much more canny and well informed than I would have been when I was doing my thesis” (#27). Open research was portrayed as one burden disproportionately borne by more recent cohorts. One professor described a “Marie Curie generation” comprised of “scholars in Europe, at least in the last 10 to 20 years, who have gone through EU funding, where they’ve been trained up, where they’ve been encouraged to adopt a lot of these [open research policies]” (#11).
Finally, resource availability was seen as a major driver affecting decisions about openness. Scholars in more stable positions, particularly those with permanent contracts or large grants, had greater access to the institutional resources required to support open research. Variability in resources also deepened geopolitical inequalities, with open research products and practices more accessible to scholars in the wealthier Global North.Footnote 2 Several intersecting phenomena factored into this disparity, including the concentration of grant funding in the Global North, the exorbitant APCs, the high costs of library journal subscriptions, varying national attitudes toward open research, and the ethnocentrism of commonly deployed open research guidelines such as the FAIR principles. Recognition of such dynamics was bound up, however, with the sense that perpetuating them was a prerequisite for academic success. One postdoc bitterly noted, “I’m a hypocrite myself in this sense—I aim for these bigger journals and I know that paying 10,000 pounds to get into Nature, for example, is morally bankrupt because [that’s] more money than some people on Earth, like in the Global South, can obviously afford to spend to publish in these journals” (#25).
What Kinds of Support or Training Would Expand Participation in Open Research?
Two recurring themes emerged in archaeologists’ discussions of how to expand disciplinary participation in open research. First, one-third of respondents recommended developing workshops on open research or related topics such as data management and standardization. More explicit training around open research practices was perceived to be essential across PhD students, postdocs, and faculty. Second, archaeologists at earlier career stages showed a surprising affinity for deploying mandates as a tool for changing disciplinary behavior. This strategy was suggested by 29% of our respondents, including half of the PhD respondents, one-quarter of the postdoctoral respondents, and one faculty member. Most commonly, mandatory publication of data and scholarly work was recommended as an integral component of funding awards or journal publications.
Discussion
Investigations of open research in archaeology confront a central paradox: Why is there limited implementation of open research practices despite their many disciplinary benefits? Archaeologists were initially slow to make their research open (Beck and Neylon Reference Beck and Neylon2012:490; Jeffrey Reference Jeffrey2012:554). Even though the internet has since infused every aspect of contemporary life and many of the promises of Web 2.0 have been realized, little has changed to make archaeological data (Marwick and Pilaar Birch Reference Marwick and Birch2018; Palmisano and Titolo Reference Palmisano and Titolo2024), methods (Farahani Reference Farahani2024; Marwick Reference Marwick2017, Reference Marwick2025), or publications (Marwick Reference Marwick2020) more open. How, then, to explain this paradox?
Our interviews demonstrate that academic archaeologists understand open research to be beneficial and feel an ethical compulsion to make their research open. Many interviewees, however, lamented their insufficient familiarity with the open research landscape and the dearth of technical training necessary to generate portable data or open code. Open access publishing requires a command of arcane bureaucracy, including understanding different forms of copyright, appreciating transformative agreements,Footnote 3 and negotiating institutional supports. Marwick (Reference Marwick2020:164) notes, “These are not practices that are taught as part of a typical university education in archaeology and must be picked up piecemeal, as needed.”
This “piecemeal” approach was also characteristic of respondents’ open research practices, which tended to be ad hoc and reactive. PhD students often initially encountered openness through using open data or methods in their own projects, whereas postdocs and faculty first confronted demands for open outputs when submitting grant applications. Indeed, funding bodies were key institutions enforcing compliance with open research, requiring that applicants demonstrate at least a minimal familiarity with standard practices. Compulsory compliance, however, is not the same thing as an embrace of openness.
Funding mandates overlooked the unique character of archaeological data that, regardless of density, are “patchy, with levels of uncertainty and variable expert opinions that are hugely challenging” (Bevan Reference Bevan2015:1477). This “patchiness” presents major challenges for the standardization and interoperability required by truly open data. Because data collection and recording are idiosyncratic—specific to scholars or projects rather than to subfields or the discipline as a whole—“one-size-fits-all solutions will fail” (Kansa Reference Kansa2012:509). This methodological diversity includes variability in recording systems, technological equipment, file types, and software packages (Jeffrey Reference Jeffrey2012:557). Even when archaeologists share similar levels of experience and methodological training, during data reuse practitioners must make individual choices about data cleaning, data aggregation, and data disaggregation in which “seemingly small choices can lead to vastly different interpretive results” (Atici et al. Reference Atici, Kansa, Lev-Tov and Kansa2013:669)
Our study suggests that interoperability is not perceived as essential to all archaeological data but instead is understood to be an accessory feature tacked on afterward at considerable cost, especially when doing so requires learning new skills such as database management or open source data analysis software. Archaeologists who wish to make their research open must often backtrack and invest significant “translational labor”: “Integrating, cleaning, and adequately documenting such large and complex datasets requires a great deal of effort and experience with data” (Kansa Reference Kansa2012:510). Open data can also incur cumulative costs, because “for long-term access, digital data actually need constant management” (Jeffrey Reference Jeffrey2012:555).
In addition to the time and labor costs associated with open data and methods, the financial costs of open access were flagged repeatedly by our respondents. Another recurring motif was the incompatibility of grant timelines with the pace of archaeological research, which could take years to progress from excavation to analysis to publication. One faculty member complained, “You publish after your project is finished, after your budget is over, when you’ve no resources to actually publish open access” (#9). In addition to balancing their budgets, respondents were concerned about the risks associated with open data and methods, especially the potential to get scooped or to have other academics or the public misinterpret their results. Systematic work on data scooping in archaeology has yet to be conducted, so respondents’ anxieties seem to be rooted in personal experiences (as in the earlier quote from respondent #6) or anecdotal accounts of the practice from other researchers.
Sector-wide political and economic forces also played a role in decisions about open research practices. Key pressures shaping our respondents’ behavior were the standards for hiring and professional advancement in academia. The institutional focus on deliverables led peer-reviewed publications to be regarded as ends in themselves, rather than as platforms for sharing information with fellow scholars and the public. One PhD student explained, “The intention behind publishing is not sharing what you’ve done . . . ; it’s more like filling the quota” (#5). In a disciplinary demonstration of Goodhart’s Law—“when a measure becomes a target, it ceases to be a good measure”—respondents recognized this value system as antithetical to both rigorous research and public communication of scholarship. Troublingly, Cook et alia (Reference Cook, Çakirlar, Goddard, DeMuth and Wells2018:149) suggest that this ideology is inculcated early in archaeological careers.
Our findings partially reflect the geographic composition of our interview sample; in the United Kingdom and Ireland, regional funding bodies, such as UK Research and Innovation (UKRI) and the EU Framework Programme for Research and Innovation, play a central role in driving researcher compliance with open access mandates, and in Europe, open data are strongly encouraged. In other countries where major funding bodies employ less stringent criteria—for example, the US National Science Foundation requires only data management plans—researcher motivations for implementing open practices may differ. Research comparing national trajectories in open research practices in archaeology would help elucidate such distinctions. Despite national differences in funding and institutional mandates, many of the core factors shaping respondent attitudes and behaviors, including the growing diversity of archaeological data, escalating bureaucratic workloads, a hyper-competitive job market, and limited availability of research funding, are the result of sector-wide shifts in archaeology and higher education. We thus suspect that the tensions we identify between the perceived benefits of openness but the slow pace of implementation likely apply in archaeology more broadly, beyond Ireland and the United Kingdom.
The traditional value system of academic archaeology, which is perceived as rewarding peer-reviewed publications and external funding above all else, also presented particular obstacles for archaeologists without job security and for practitioners in the Global South, who lacked the institutional resources required to access paywalled articles or fund gold open access (see also Ma Reference Ma2022, Reference Ma2025). Bevan (Reference Bevan2015:1480) underscores, “It is often Western institutions that retain best access to [digital] information, and it is often researchers who have grown up (educationally speaking) in data-, resource- and technique-rich Western archaeological settings who will be academically best placed to exploit it.” An uncomfortable corollary is that the same forces that agitate against archaeology being open also work to render it demographically homogeneous, flattening our interpretations of the past (Heath-Stout Reference Heath-Stout2025).
Conclusion
Understanding how open research is meaningful to and within archaeology is far from a simple cost-benefit analysis. Developing such an understanding requires interrogating individual and collective risks and rewards, research evaluation systems, resource distribution, and the methods and epistemology of the field. Indeed, another way to frame this question might be to ask, echoing Farahani (Reference Farahani2024:393), “What is the point of archaeology”? If the point is to understand the past and make that understanding accessible to the public, then open data, open methods, and open access serve those goals. What this study and others have demonstrated, however, is that the benefits of open research are offset by a range of insidious costs and obstacles. The diverse range of data accrued by a field that examines everything from oyster shells to architecture presents particular problems for developing shared standards of recording and dissemination. The resultant diversity in training means that the baseline skills and core knowledge necessary for open data and methods are unevenly distributed. Archaeologists are certainly capable of acquiring such skillsets; practitioners learn how to use complicated devices (total stations, 3D scanners, magnetometers, drones) and software packages (ArcGIS, OxCal) as a matter of course. The distinction is that such skill sets are incentivized: archaeologists cannot collect the data they need to understand the human past without them.
Although archaeology is unique in its methodological and theoretical diversity, it also shares much with other disciplines. Beatson et alia (Reference Beatson, Tharapos, O’Connell, De Lange, Carr and Copeland2022:717) have recently denounced the erosion of “academic citizenship,” the “collegial contribution of faculty to the culture, community, and civic purpose of a university and society in general.” This retreat from service work has been instigated by the restructuring of universities under corporate models (Graeber Reference Graeber2018) that cultivate instrumentalist attitudes among faculty and would-be faculty, who are acutely aware of the metrics used to evaluate performance. Open research is one form of academic citizenship, but as Hostler (Reference Hostler2023) points out, open research practices and the time and energy that must be diverted from other activities to implement them are largely invisible in academic workload models.
Taking the larger sectorial context into consideration shifts the question from “Why do most archaeologists not engage in open research, despite the benefits” to “Why do some archaeologists engage in open research, despite the costs?” To make academic archaeology open will require more than developing coding workshops or committing to green open access,Footnote 4 although these would be commendable first steps. It will require a reevaluation of what we want the point of universities to be. If universities are to be one of the main fora through which we understand the past and make our interpretations accessible to a broader public, we must advocate for incentive structures that support, reward, and prioritize making our research open.
Acknowledgments
We thank all interview respondents who were generous with their time and shared their perspectives on open research, as well as Fiachra Ó Coileáin (Gaeltacht University College Dublin) for the Irish-language translation of the abstract and keywords. Constructive feedback from three reviewers improved this article. This research received a low-risk ethics approval from the Human Research Ethics Committee—Humanities (HREC-HS) at University College Dublin on February 28, 2024 (Research Ethics Reference Number HS-C-24-110-Ma). Respondents gave written consent for review and signature before starting interviews.
Funding Statement
This work was supported by Ireland’s National Open Research Forum (NORF) under the project “Cultivating an Open Research Culture in the Humanities and Social Sciences” as part of the 2023 Open Research Fund. NORF is funded by the Higher Education Authority (HEA) on behalf of the Department of Further and Higher Education, Research Innovation and Science (DFHERIS).
Data Availability Statement
The interview transcripts on which this study is based cannot be shared to protect participant confidentiality and ensure compliance with informed consent agreements.
Competing Interests
The authors declare none.
Supplementary Material
The supplementary material for this article can be found at https://doi.org/10.1017/aap.2026.10184.
Supplementary Material 1. Themes and Subthemes Produced by Thematic Analysis (table).
Supplementary Material 2. Perceived Costs Associated with Open Research in Archaeology (table).
Supplementary Material 3. Respondents’ Characterizations of Open Access Publishing Costs, Article Processing Charges (APCs), or Both (table).
Open access
