About This Reference

This site is an open-access reference for the reconstruction of Proto-Indo-European (PIE), the hypothetical ancestor of the Indo-European language family. It aims to present reconstructed roots, sound laws, and comparative data in a form that is useful to students, researchers, and anyone with a serious interest in historical linguistics.

Methodology

Reconstructions follow the standard comparative method as established in the Neogrammarian tradition and refined over the past century and a half. The principle is straightforward: regular sound correspondences between attested daughter languages allow the inference of ancestral forms. Where Latin has pater, Greek patēr, and Sanskrit pitá, we reconstruct PIE ph₂tēr.

We adopt the laryngeal theory in its standard three-laryngeal form (h₁, h₂, h₃), following the consensus that has prevailed since the mid-twentieth century. Where scholars disagree on specific reconstructions, we note the dispute and cite the relevant authorities.

Data Sources

The reconstructions and descendant forms presented here draw on the following major reference works, among others:

• Pokorny, Julius. Indogermanisches etymologisches Wörterbuch (1959). The classical PIE dictionary. Outdated in many specifics but still an essential starting point.
• Rix, Helmut, et al. Lexikon der indogermanischen Verben (LIV, 2nd ed. 2001). The standard reference for PIE verbal roots.
• Beekes, Robert S. P. Etymological Dictionary of Greek (2010). Comprehensive treatment of Greek etymologies with PIE reconstructions.
• Mallory, J. P. & Adams, Douglas Q. The Oxford Introduction to Proto-Indo-European and the Proto-Indo-European World (2006). Accessible overview of PIE culture and language.
• de Vaan, Michiel. Etymological Dictionary of Latin and the other Italic Languages (2008).
• Kroonen, Guus. Etymological Dictionary of Proto-Germanic (2013).
• Derksen, Rick. Etymological Dictionary of the Slavic Inherited Lexicon (2008).

Reconstruction Confidence

Each root page carries a confidence assessment reflecting the scholarly consensus:

Content Generation

Root pages, sound law descriptions, and discussion sections are drafted by large language models (Opus-class) operating under strict scholarly prompts. All generated content is reviewed against the reference literature before publication. Every page records the generating model, generation date, and revision number.

This is not on-demand generation. Content is produced in offline batches, reviewed, and published only when it meets the project’s accuracy standards. If a root page does not yet exist, the site returns a 404 rather than generating speculative content.

Notation

For a complete guide to the notation used on this site—the asterisk convention, laryngeal notation, ablaut grades, and IPA transcription—see the Notation page.

Limitations and Disclaimers

PIE is a reconstruction. No PIE text survives, and the reconstructed forms represent hypotheses about a language spoken millennia ago. The comparative method is powerful but not infallible: chance resemblances, borrowings, and areal features can mislead. Where the evidence is ambiguous, we aim to say so.

This site does not claim to replace the primary literature. It is a reference and gateway, not a substitute for reading Beekes, Rix, Pokorny, or the specialist journals. We encourage readers to consult the original sources cited on each page.

Sponsorship

This project is sponsored by etymologist.ai, which provides the underlying word data and infrastructure. Sponsorship does not influence content. There are no display advertisements, affiliate links, or paywalls on this site.

Contact

For corrections, suggestions, or scholarly inquiries, contact us through etymologist.ai.