Skip to main content

About

What this site is, what it isn't, and how to read it honestly.

What is this?

Sculpture in Data is an interactive data explorer for sculpture history — people, geography, movements, institutions, works, and recorded lineages since 1800. It is meant for exploration and question-forming, not as a complete canon or a substitute for art-historical research.

The project began when a friend at the National Sculpture Society asked for a graph of artists and their lifespans. That small request became an experiment in using AI-assisted development to make art data visual, playful, and inspectable.

Scope & what you're seeing

The published dataset includes 3,543 sculptors born since 1800, selected from an analytically eligible candidate frame of 6,710. The source query returned 6,711 records; evidence-backed record exclusions are documented on Transparency. A sculptor appears if they satisfy at least one signal: a recorded art movement, a documented mentor/student edge, curation onto the focus list, multiple citizenships, or meaningful non-English Wikipedia reach.

The lineage graph also includes 681non-sculptor teachers and influences — painters, composers, architects, and others connected to sculptors but not classified as sculptors in Wikidata. They appear as diamond-shaped nodes so cross-media relationships aren't silently dropped.

A smaller curated focus list (48 sculptors), assembled with input from the National Sculpture Society, powers the Timeline view and drives detail-page highlights. It emphasizes the American figurative tradition and is not a comprehensive survey of global sculpture.

This is not an art-historical argument — it's a lens on structured public data. Where the data is incomplete, we say so, and the Transparency page spells out exactly who we include, why, and what biases the rule introduces.

Explorer, laboratory, and future scope

The public explorer favors interpretable views whose data lineage, uncertainty, and limitations can be explained. The repository also acts as a laboratory for prototypes — including temporal networks, institutional paths, places, and coordinated views — that should graduate into the public product only after they answer a clear reader question and pass evidence, accessibility, and performance gates.

Sculpture is the evidence-building domain, not a permanent technical ceiling. The model may expand additively to other artists, places, institutions, objects, and periods when source quality supports honest comparison; expansion should not flatten every art form into one universal ranking or undocumented influence graph.

Data sources & licenses

Current:

Contains information from the Union List of Artist Names (ULAN)®, made available under the Open Data Commons Attribution License. The combined export has mixed provenance; source-specific terms and object/file attribution still apply.

Methodology

Most displayed fields are derived from public databases, and the inclusion rule is versioned rather than scored subjectively. A small set of focus-list choices and evidence-backed corrections or exclusions is maintained explicitly so curation is visible instead of being disguised as raw source data. See the Transparency page for exact counts, inclusion signals, and demographic breakdowns.

Gender, citizenship, nationality/culture, movement, and relationship values are attributed source assertions. The project does not infer missing identities, treat citizenship as culture, or treat a recorded relationship or graph path as proof of causal influence.

Read the maintained dataset datasheet for composition, licenses, processing, missingness, intended and non-intended uses, risks, citation, and maintenance. The public claim register records the evidence and language boundary for each analytical view.

Credits

Built by Asher Zafar

Inspired by Fabio J. Fernández and the National Sculpture Society.

Data pipeline in Python, web app in Next.js with D3.js. View source on GitHub.