1. Introduction

In osteoarchaeological research, the demand for integrated, cross-disciplinary datasets has grown significantly. The inclusion of bioanthropological information within archaeological analyses enables a more nuanced reconstruction of past lived experiences than archaeology alone can provide (Richards / Richards 2022; Richards 2024). However, anthropological documentation is still characterised by heterogeneous descriptive practices, inconsistent terminology, and varying levels of detail. Such variability complicates data comparison across sites, research teams, and regions, and it hampers the integration of bioanthropological information into digital research infrastructures. In the context of DH, where interoperability, machine-readability, and long-term reusability are essential, the absence of standardized vocabularies poses a considerable challenge.

This paper presents a newly developed controlled vocabulary for human bones, tailored specifically for osteoarchaeological and bioanthropological contexts. Its primary aim is to establish a consistent semantic structure for describing skeletal elements, thereby enabling researchers to document anthropological observations in an interoperable and standardised manner. The paper further discusses the integration of the vocabulary into the database system OpenAtlas (https://openatlas.eu/) and THANADOS (https://thanados.net/), a web application to present archaeological and anthropological datasets. Through the combination of semantic modelling and technical implementation, the vocabulary enhances data quality, transparency, and analytical potential within digital osteoarchaeology.

2. Background

2.1 Anthropological Data in Digital Archaeology

Anthropological datasets are typically recorded according to discipline-specific conventions that vary in terminology, structure, and descriptive scope. Descriptions of skeletal elements, preservation, or pathologies rely on free-text entries or locally developed schemas. While such approaches may suffice for isolated analyses, they hinder data integration and comparability. In addition to inconsistent naming conventions, the scarcity of formalised relationships between anatomical entities limits computational processing and machine-actionable analysis. This fragmentation restricts the potential of digital tools, from database-driven querying to semantic exploration and automated pattern detection.

2.2 Role of Ontologies and Controlled Vocabularies

Controlled vocabularies offer an effective solution to these challenges by providing stable terminology, hierarchical structures, and persistent identifiers (Hedden 2008; Harpring 2010). Within the FAIR framework (https://www.go-fair.org/fair-principles/), they are essential for ensuring semantic interoperability and long-term reusability (Ďurčo / Illmayer 2020). Their use enables the creation of structured, machine-readable datasets that can be efficiently linked, queried, and reused across disciplinary boundaries.

Several ontological resources—such as the Foundational Model of Anatomy (FMA; http://si.washington.edu/projects/fma/), the Anthropological Notation Ontology (ANNO; https://ols.imise.uni-leipzig.de/index), or the Terminologia Anatomica (TA98; Whitmore 1998) - address anatomical structures. However, their scope does not always align with the needs of osteoarchaeological practice, where fragmented, burnt, or otherwise taphonomically altered remains must be systematically categorised. The vocabulary proposed here addresses this gap by providing an application-oriented, domain-specific solution.

3. A New Vocabulary for Human Remains

3.1 Motivation and Development Goals

The controlled structured vocabulary was developed to close semantic gaps in the description of skeletal elements within datasets. Key objectives included:

(1) Establishing a hierarchical structure for skeletal entities (2) Defining consistent terminology for human remains (3) Supporting machine-readable formats (4) Enabling interoperability with existing ontologies and data models

Rather than replacing established anatomical ontologies, the vocabulary complements them by emphasising granularity, archaeological applicability, and ease of implementation.

3.2 Structure and Organisation

The vocabulary is organised hierarchically, beginning with a distinction between adult and subadult anatomical structures, followed by major body regions represented as SKOS collections. It then progresses to individual bones and bone segments. Each concept is enriched with multilingual synonyms (English, Latin, German) and semantically meaningful relationships. This structure ensures both human-readable documentation and machine-actionable semantic representation. Metadata are provided for the vocabulary as a whole and for each concept individually.

Where possible, concepts are aligned with established standards such as TA98, ANNO, and Wikidata (https://www.wikidata.org/). The latter enables indirect connections to additional vocabularies, including FMA, ICD-11 (https://icd.who.int/), or Uberon (https://obofoundry.org/ontology/uberon.html), enhancing the vocabulary’s interoperability and linking potential.

3.3 Methodology

The development process was iterative, combining literature review, analysis of existing ontologies, and expert consultation. Classical anthropological handbooks and osteological guidelines (e.g. Mann et al. 2016; Scheuer et al. 2008; White / Folkens 2005; Whitmore 1998) provided the foundational structure. Draft versions were reviewed by osteologists and archaeologists, whose feedback informed refinements in terminology and conceptual modelling.

For digital implementation, the vocabulary was encoded in SKOS (Miles / Bechhofer 2009) and serialised as RDF/XML and Turtle (TTL) (Carloni / Trognitz 2025). These formats support interoperability with CIDOC CRM (https://cidoc-crm.org/) and related ontologies widely used in archaeological data modelling. Following final revisions, the vocabulary will be published on ACDH Vocabs (https://vocabs.dariah.eu/de/), a Skosmos-based (https://skosmos.org/) browsing environment, offering hierarchical visualisation, REST API access, and persistent identifiers for all concepts (Ďurčo / Illmayer 2020; Andorfer / Illmayer 2023).

4. Integration into Digital Research Environments: OpenAtlas and THANADOS

The vocabulary enables precise, consistent, and CIDOC CRM–compliant documentation of anthropological data in OpenAtlas. It standardises bone identification, supports structured querying, and facilitates coherent links between anthropological, spatial, and archaeological information. In THANADOS, the vocabulary replaces an earlier terminology and serves as a second major use case, enabling evaluation within real-world archaeological publication workflows.

5. Discussion

The vocabulary provides substantial benefits by enhancing documentation consistency, improving data comparability, and supporting FAIR-aligned practices. Nonetheless, challenges remain. The level of granularity requires some training for users unfamiliar with semantic structures. Furthermore, sustained community engagement and tool support will be essential for long-term adoption.

6. Conclusion and Outlook

The vocabulary constitutes a significant step toward the standardisation and semantic enrichment of anthropological data in digital osteoarchaeology. By adhering to semantic standards, it facilitates the integration of anthropological information into broader knowledge-graph environments, enabling complex analyses such as regional comparisons of burial practices or the integration of skeletal data with isotopic or palaeopathological results.

Its implementation in OpenAtlas and THANADOS underscores the value of controlled vocabularies within Digital-Humanities infrastructures. Future work will focus on open publication, community feedback, and expanding content—particularly for subadult bone structures. Collaborative development will be crucial for ensuring long-term relevance and sustainability.