LingBuzz is not a publisher and not a journal. It is an article archive and community space for linguistics, created around 2004 by the linguist Michal Starke and run by him ever since - originally hosted at the University of Tromso and still living at ling.auf.net. Inside generative grammar it functions the way arXiv functions in physics: it is where a paper appears first, where colleagues find it, and where it is cited from for the years between the first draft and the printed version.

The archive is small by repository standards - the site counter showed 10,072 papers in August 2026 - but its coverage of theoretical syntax, semantics, phonology and morphology is unusually complete. Posting is free, reading is free, and authors keep their copyright. This guide covers how uploading works, what screening does and does not do, how to cite a LingBuzz paper, and how posting interacts with journal submission.

1. Register an account. LingBuzz asks that at least one author of a submission be a registered user; accounts are open to members of the relevant linguistics disciplines.
2. Upload the PDF. Submissions are distributed as PDFs, with title, authors, abstract and keywords entered as metadata.
3. It appears immediately. There is no queue for acceptance - the paper shows up in the recent-submissions feed as soon as it is posted.
4. Fix mistakes early. Authors get roughly a one-week grace period to correct submission information; after that, editable fields narrow to metadata such as keywords.
5. Add versions freely. New versions can be added at any time. The newest version is displayed, with earlier ones kept alongside it, so a manuscript can be updated from first draft through accepted version without breaking the link.
6. Withdraw if needed. A paper can be marked withdrawn, which pulls the current version while leaving the version history intact.

LingBuzz applies no formal peer review. Material is expected to be of interest, relevance and value to the disciplines represented, and the site reserves the right to remove content that is not; the documentation has floated an arXiv-style endorsement system as a way to reduce manual moderation. In practice, quality control is social rather than editorial - the archive is used almost entirely by working theoretical linguists, and reputational cost does the filtering.

The practical consequence for readers: a LingBuzz posting is not evidence of peer review. Many entries are accepted or published papers, many are submitted manuscripts, and some are drafts that never went further. Check the paper's own note for its status before treating it as vetted.

Each entry gets a six-digit archive ID, and the convention in the field is to cite that ID directly - for example: Author, A. 2025. Title of the paper. Ms., lingbuzz/006543. Add the URL (https://ling.auf.net/lingbuzz/006543) where the style allows it, and state the version or download date, since papers on LingBuzz are revised in place.

Three rules worth following:

• Cite the journal version once it exists. If the paper has been published, cite the published article and optionally note the LingBuzz ID for open access.
• Label unpublished work as unpublished. "Ms." or "manuscript" belongs in the reference so the reader knows what they are getting.
• Do not treat the ID as a DOI. LingBuzz does not mint DOIs; the ID and URL are the persistent handles.

Posting on LingBuzz almost never blocks later journal submission in this field. Linguistics journals - including Linguistic Inquiry, Natural Language & Linguistic Theory, Glossa, Syntax and Semantics and Pragmatics - generally accept manuscripts that already circulate as preprints, and Glossa is itself a fully open-access, no-fee venue. Two cautions apply:

• Anonymous review. A widely circulated preprint under your name can undercut double-anonymized review. Some journals ask you not to post during review; check the specific journal's policy first.
• Version rights. After acceptance, what you may post is set by your publishing agreement - usually the accepted manuscript rather than the typeset PDF, sometimes after an embargo.

There is no charge to post and no charge to read. Authors retain copyright on material they submit; LingBuzz does not require a transfer and does not impose a single blanket licence, so reuse terms depend on whatever the author states in the document. If you want explicit reuse permissions, add a Creative Commons statement to the PDF yourself.

LingBuzz is deliberately minimal, and that is both its strength and its constraint. It has no DOIs, no formal metadata API on the scale of arXiv or OSF, no institutional backing to guarantee long-term preservation, and no peer review. It is not indexed as a journal in Scopus or Web of Science, though papers are routinely picked up by Google Scholar. For work that needs a DOI, versioned archival guarantees, or a discipline-neutral home, pair a LingBuzz posting with a deposit in an institutional repository or a general-purpose archive.