Artificial patination as a selective filter: experimental insights into interpretative bias in metalwork wear analysis

Abstract

Abstract The analysis of use-wear traces on metal artefacts is increasingly applied in archaeological research, yet its interpretative reliability remains limited by post-depositional surface transformation. In particular, the impact of corrosion and patina on trace visibility has rarely been investigated under controlled experimental conditions. This study presents a controlled experiment designed to examine how an artificially induced patina affects the visibility and interpretability of known surface traces on copper alloy artefacts. A replica bronze knife with experimentally generated surface modifications was subjected to a staged patination process, followed by microscopic observation at successive stages of surface transformation and after partial mechanical cleaning. The results demonstrate that patination acts as a selective filter, systematically reducing the visibility of shallow traces while preserving only the most pronounced surface features. Advanced corrosion stages may obscure the majority of traces, whereas cleaning procedures re-exposed original features and introduced new surface modifications. These observations indicate that the observable surface of metal artefacts does not represent a direct record of past use, but rather a transformed and selectively preserved signal shaped by corrosion and conservation processes. This introduces a systematic bias in metalwork wear analysis, favouring the detection of high-intensity traces and potentially leading to underestimation or misinterpretation of artefact function. The study highlights the need to explicitly account for surface transformation processes in the analysis of archaeological metal artefacts and provides an experimental framework for evaluating the reliability of trace-based interpretations.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 0 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.