A bibliometric analysis of cultural heritage research in the humanities: The Web of Science as a tool of knowledge management

The paper investigates how cultural heritage research has expanded and diversified across disciplines, with roots in the humanities and strong connections to nationalism, memory, and governance. It contextualizes growth in publication volume within broader academic trends emphasizing productivity and bibliometric assessment. The authors note a critical turn in heritage studies since the 2000s, expanding attention to discursive, participatory, affective, social, political, economic, and sustainability dimensions of heritage. Despite this expansion, there is a lack of comprehensive analysis linking publication patterns to structural forces, notably shifts in the international cultural heritage regime since 2003 (e.g., UNESCO and ICOMOS conventions). The study aims to map output volume, patterns of collaboration, and intellectual structures in WOS-indexed humanities research on cultural heritage, and to critically assess WOS as a knowledge management tool that shapes reputations and field perceptions.

The authors review the rise of bibliometric evaluation in academia and its application to humanities, acknowledging critiques such as disciplinary differences in indexation and the assumption that all citations equally indicate quality. Prior bibliometric studies in heritage-related domains include analyses of heritage tourism (Kumar et al., 2020; Bhowmik, 2021; Zhang et al., 2022), historical wall paintings (Zhu et al., 2022), and intangible cultural heritage (Chen et al., 2020; Su et al., 2019). They also situate cultural heritage scholarship within longer intellectual traditions (antiquarianism, archaeology, nationalism) and critical developments (e.g., Smith, 2006; Waterton and Smith, 2009; Logan, 2012; Harrison, 2013a,b). The literature highlights the influence of international governance (UNESCO/ICOMOS) and the critical turn addressing power, participation, and contested heritage, while noting the paucity of comprehensive bibliometrics for the humanities-focused cultural heritage corpus in WOS.

The study conducts a bibliometric analysis combining performance analysis and science mapping. Data source: Web of Science (WOS) Core Collection. Search strategy (run Aug 5, 2022): Topic field TS="cultural heritage" (with quotation marks), to capture occurrences in titles, abstracts, or keywords. Initial results: 27,205 documents. Refinement: WOS Category filter to "Humanities Multidisciplinary" yielding 2,410 documents (~8.86% of initial). Document type filter to "article" reduced the set to 1,845. Timeframe restriction to 2003–2022 (aligned with the post-2003 shift in heritage regime) produced a final dataset of 1,843 journal articles. No language restriction, but WOS requires English titles/abstracts/keywords for non-English articles. Data export: full record + cited references as tab-delimited; saved marked list. Tools: WOS Analyze Results and Citation Reports for descriptive statistics; VOSviewer (v1.6.16) for science mapping. Analyses: - Co-authorship networks at levels of authors (minimum 3 papers per author; n=78 authors met threshold; 46 connected into 12 clusters), institutions (minimum 8 papers per institution; 36 met threshold; 25 connected into 7 clusters), and countries (minimum 10 papers per country; 40 met threshold; 8 clusters). - Keyword co-occurrence analysis: author keywords; thesaurus used to merge variants (e.g., plural/singular, spelling); minimum occurrence threshold=6; association strength normalization. Data cleaning included thesaurus files for author/institution disambiguation. Statistical tests included correlations between link strength and citations/publications at author and institution levels.

- Corpus and growth: 1,843 WOS-indexed humanities articles (2003–2022) on cultural heritage. Strong growth since 2017; 1,509 articles (≈82%) published 2017–2022. Peaks in 2019 and 2021 (>320 articles/year). - Languages (of 1,843): English 71.19% (1,312), Spanish 6.67% (123), Russian 6.29% (116), Italian 5.26% (97), French 3.64% (67), Portuguese 1.79% (33), Turkish 1.25% (23), Slovak 0.98% (18), Slovenian 0.87% (16), German 0.54% (10). - Research areas (top within Humanities Multidisciplinary collaborations; % of 1,843): Social Sciences Other Topics 12.86% (237); Science Technology Other Topics 9.12% (168); Other Topics 8.03% (148); Computer Science 5.81% (107); Chemistry 5.81% (107); Materials Science 5.81% (107); Spectroscopy 1.41% (26); Linguistics 0.65% (12); Information Science & Library Science 0.43% (8). - Top publishing countries (share of 1,843): Italy 15.79% (291); England 9.28% (171); Spain 8.84% (163); USA 7.98% (147); Russia 6.84% (126); France 5.64% (104); Australia 4.39% (81); Greece 3.53% (65); Germany 3.47% (64); China 2.82% (52). Together these account for 69% of publications. - Most productive affiliations (examples): University of London (49), UDICE French Research Universities (43), National Research Council CNR (39), University College London (37), Russian Academy of Sciences (23), CNRS (20), National Research Tomsk State University (20), University of Bologna (19), University of Seville (18), Sapienza University Rome (17), University of Ljubljana (17), Polytechnic University of Milan (16), University of Florence (16), University of Macerata (15). - Journals (record counts among 1,843): International Journal of Heritage Studies (221; 11.99%); Heritage (168; 9.12%); ACM Journal on Computing and Cultural Heritage (135; 7.32%); Heritage Science (107; 5.81%); Il capitale culturale (78; 4.23%); International Journal of Intangible Heritage (74; 4.02%); Tomsk State University Journal of Cultural Studies and Art History (58; 3.15%); Historic Environment: Policy & Practice (48; 2.60%). The top three journals (n=524 articles) garnered 2,977 citations, ≈49% of total citations; average citations per article in these journals (approx. 8–11–6) exceed overall mean 3.34. - Most-cited articles: Bekele et al. 2018 (AR/VR/MR survey) 202 citations; Chirikure et al. 2010 (community participation in Africa) 103; Harrison 2013 97; Winter 2014 66; Holtorf 2015 65; Roberts & Cohen 2014 65; Logan 2012 62; Leissner et al. 2015 58; Rubino et al. 2015 57; Mydland & Grahn 2012 57. - Authorship patterns: ~48% single-author; 21% two authors; 31% three or more. Most prolific author: Massimo Montella (12 articles). Melissa Terras authored seven joint articles (2017–2021; 53 citations noted). - Co-authorship (authors): 78 authors with ≥3 papers; 46 interconnected into 12 clusters. Notable clusters include Greek researchers at FORTH (strong links; e.g., Partarakis, Zabulis), a UK-Norway cluster around UCL/Brighton/Durham (e.g., Echavarria, Samaroudi, Weyrich), and an Italian cluster at Marche Polytechnic University (e.g., Pierdicca). Weak overall inter-author connectivity; collaborations often within the same institution. Correlation between prolific authors’ total link strength and citations r≈0.19 (p<0.1). - Co-authorship (institutions): 36 institutions with ≥8 papers; 25 interconnected into 7 clusters. CNR (Italy) most connected, collaborating with Italian, Spanish, Portuguese, and UK institutions. Correlations: link strength vs citations r≈0.45 (p<0.01); link strength vs number of articles r≈0.72 (p<0.01), indicating institutional collaboration associates with higher impact and productivity. - Co-authorship (countries): 40 countries with ≥10 papers; Italy (291; link strength 97) and England (171; 95) most connected. Some highly productive countries (e.g., Russia, 126; link strength 3) show weak international ties; some with moderate outputs (e.g., Germany, Switzerland, Austria, Netherlands, Ireland) show high collaboration intensity. Eight cross-national clusters identified. - Keyword co-occurrence (n=108 keywords with ≥6 occurrences): 10 thematic clusters. Major clusters include digital heritage technologies (VR/AR, 3D, visualization, storytelling, serious games, museums); authenticity/identity/sustainability and industrial/urban/landscape heritage; protection/intangible heritage/community/UNESCO/World Heritage; digitization, social media, archives, archaeology, design, heritage education; governance/politics/legislation/EU/cultural tourism/management; tangible/built heritage, restoration/conservation/values/education (incl. Turkey); art/culture/language/history/living heritage/globalization; memory/migration/digital cultural heritage/music/cultural landscape/COVID-19/climate change; risk management/protection/reconstruction of architectural heritage. - Knowledge-oriented subfield: 37 articles explicitly referencing knowledge terms; 119 keywords forming 12 clusters (e.g., knowledge representation/semantic web/ontologies/open data; digital heritage/archives/copyright/provenance/repatriation; professions/conservation/education/values; crafts/women/empowerment/inequality; innovation/sustainable development; libraries/society/memory; knowledge management/heritage impact assessment/ethnography; traditional knowledge/indigenous data sovereignty/legislation; semantics/narratives/storytelling/AR; identity/funerary traditions/Caribbean; citizen science/vernacular architecture/web & mobile apps). Earliest (2009); peak in 2021 (nine articles). - Funding context: EU programs (Seventh Framework, Horizon 2020, Horizon Europe) and UKRI prominently acknowledged, aligning with Eurocentric output and collaboration patterns.

The findings demonstrate that WOS-indexed humanities research on cultural heritage has grown rapidly since 2017, with output concentrated in a relatively small set of European countries and institutions and limited broad-based collaborative ties. Thematically, the field aligns with societal and sectoral shifts of the last two decades: digitization of GLAM institutions, adoption of VR/AR and gamification, increasing community participation, and growing attention to sustainability and climate impacts. Keyword clusters mirror priorities in international heritage governance (UNESCO/ICOMOS conventions on intangible, digital, urban/rural landscapes) and the critical turn that interrogates power, inclusion, and contested narratives. Collaboration correlates with higher impact, especially at the institutional level, but co-authorship remains frequently intra-institutional. The Eurocentric profile of the WOS-indexed corpus appears linked to European funding frameworks prioritizing multidisciplinary, collaborative, and open-access outputs, reinforcing visibility for European institutions and journals. As a knowledge management tool, WOS structures and ranks the field through categorization, indexation, and citation metrics, thereby shaping perceptions of what constitutes esteemed cultural heritage research and potentially perpetuating Western-centric biases.

The study provides a comprehensive bibliometric mapping of WOS-indexed humanities research on cultural heritage (2003–2022), revealing exponential growth, concentration of outputs in European countries and institutions, influential journals with “heritage” in their titles, and dominant themes centered on digitization, participation, governance, and sustainability. It extends methodology in heritage scholarship through combined performance analysis and science mapping, uncovering interdependencies among authorship, institutions, countries, citations, and funding. Key recommendations include: (1) fostering international and inter-institutional co-authorship to enhance impact; (2) expanding cross-continental collaborations to counter Eurocentrism and diversify perspectives; (3) applying bibliometric approaches to specific subtopics for finer-grained insight into production dynamics; and (4) advancing knowledge-oriented approaches (e.g., knowledge representation, indigenous data sovereignty, professions, and big data-driven knowledge maps) as promising research avenues.

- Corpus scope: Focus limited to peer-reviewed journal articles; monographs/edited volumes (highly valued in the humanities) were excluded. - Language: Dependence on English titles/abstracts/keywords biases inclusion and visibility toward English-language or English-abstracted works. - Indexing and disambiguation: Variability in institutional naming and author name spellings required manual thesaurus corrections, with residual inconsistencies possible. - Disciplinary/indexing biases: WOS coverage and categorization can underrepresent certain humanities outputs and non-Western scholarship; bibliometric assumptions (e.g., citations as quality proxies) have known limitations. - Collaboration measures: Co-authorship networks may not capture informal or non-published collaborations; thresholds (e.g., ≥3 papers per author; ≥8 per institution; ≥10 per country) may omit relevant but less prolific actors.