Welcome to LitTEL

A Typological Database for the 'Little' Languages of the Ancient World

 

What is LitTEL?

LitTEL (List for the Typological Encoding of corpus Languages) is a typologically oriented linguistic database. Like many other databases, such as WALS or Grambank, it collects information about linguistic features in a range of different languages. LitTEL differs from other projects, however, in its particular focus on the corpus languages of ancient world (prior to 1000 CE).

Ancient corpus languages are typically ‘little’ languages in documentary terms. They may be attested in only a few texts, typically short or lacunary, and may have no living descendants. As a result, they are not always amenable to the kind of data-intense typological description that much existing work relies on. The aim of LitTEL is to provide a dataset which is tailored to the strengths of these 'little' languages.

This will allow LitTEL to capture the full linguistic diversity of the ancient world, and provide a richer insight into how the distribution of linguistic features has changed over the past five thousand years.

 

A guide to LitTEL

Fragmentary corpus languages are often messy and poorly described. Extracting useful and intercomparable information from 'little' ancient corpora requires a fine-tuned methodology. The parameters of the LitTEL database were selected and designed to avoid observation bias as much as possible, while at the same time maximising the linguistic usefulness of the often very lacunary textual corpora it focuses on.

A detailed explanation of the methodological guidelines which LitTEL follows, as well as the rationale underlying them, can be consulted under the Guidelines tab.

 

Explore LitTEL

A link to the LitTEL data (version 1.0) can be found under the Download tab, along with the pylittel module, which offers extended functionalities for querying these data (particularly in diachronic terms).

Additionally, this site aims to make the LitTEL dataset accessible as a clld web app. The dataset can be browsed by features or by language. Past typological distributions can be geographically visualised: the maps on this web app display the oldest observation for each language in the dataset.