Welcome to __Xposition__, a semantic database of adpositions (prepositions and postpositions) in the world's languages!

The database is being constructed as part of a research project by computational linguists at Georgetown University and collaborators around the world. It is currently very much a work in progress. Following the **SNACS** annotation scheme ([Schneider et al., 2018](/bib/schneider_et_al_2018/); [Schneider et al., 2022](/bib/schneider_et_al_2022/)), the database contains:

* __[supersenses](/supersenses)__: a universal set of 52 coarse-grained classes of semantic relations expressed by adpositions, organized in a taxonomy
* __[construals](/construals)__: *role*-*function* pairings, where the role and function are both supersenses; see [Hwang et al. 2017](http://people.cs.georgetown.edu/nschneid/p/prepconstrual2.pdf) for explanation
* __languages__ that we have begun to document; for each language we document
    - __adpositions__: morphological entries for prepositions and postpositions in the language, each with
        * __usages__: "sense" entries, i.e., the adposition paired with a semantic construal, as well as case information for languages where different adpositions license different complement cases

Use the toolbar at the top of the page to browse the database.

Supplementary pages: [What counts as an adposition?](/what-counts-as-an-adposition) and [Constraints on role and function combinations](/constraints-on-role-and-function-combinations)

Languages
---------

The languages listed in the menu at the top of the page have at least some lexicographic documentation (lists of adpositions and their possible supersenses).

SNACS Corpora
-------------

A number of corpora in several languages have been annotated with SNACS: [here is a list](https://github.com/carmls/datasets). Thus far, Xposition contains the data from:

* [3 English corpora](/en/):
    - [STREUSLE](https://github.com/nert-gu/streusle/), consisting of web reviews
    - [PASTRIE](https://github.com/nert-nlp/pastrie), consisting of Reddit posts in English from writers of several presumed native language backgrounds
    - [The Little Prince](https://github.com/nert-nlp/English-Little-Prince-SNACS), English translation of the _Le Petit Prince_ story.
* [1 Chinese corpus](/zh/): [The Little Prince](https://github.com/nert-nlp/English-Little-Prince-SNACS)

Attestations from these corpora appear at the bottom of adposition and supersense pages, and can be browsed in a new window. Work on importing other SNACS corpora into the site is underway.


Other Preposition Resources
---------------------------

For a finer-grained database of English prepositions, see [The Preposition Project](http://www.clres.com/).


About
-----

This website was developed at Georgetown University, adapted from [django-wiki](https://github.com/django-wiki/): [Xposition codebase](https://github.com/nert-gu/Xposition).
The research on the crosslinguistic SNACS framework is led by [Nathan Schneider](http://nathan.cl), [Vivek Srikumar](https://svivek.com/), and [Jena Hwang](https://jenahwang.github.io/) in cooperation with many other contributors.

### Recommended Citations

For the SNACS annotation scheme in general:

- Nathan Schneider, Jena D. Hwang, Vivek Srikumar, Jakob Prange, Austin Blodgett, Sarah R. Moeller, Aviram Stern, Adi Bitan, and Omri Abend. [Comprehensive supersense disambiguation of English prepositions and possessives](http://aclweb.org/anthology/P18-1018). In *Proc. of ACL*, pages 185–196, Melbourne, Australia, July 2018.

For descriptions of the latest SNACS supersense categories with respect to English:

- Nathan Schneider, Jena D. Hwang, Vivek Srikumar, Archna Bhatia, Na-Rae Han, Tim O'Gorman, Sarah R. Moeller, Omri Abend, Adi Shalev, Austin Blodgett, and Jakob Prange. [Adposition and Case Supersenses v2.6: Guidelines for English](https://arxiv.org/abs/1704.02134v8). Technical Report arXiv:1704.02134, version 8, July 7, 2022.

For this website/database:

- Luke Gessler, Nathan Schneider, Joseph C. Ledford, and Austin Blodgett. [Xposition: An online multilingual database of adpositional semantics](http://www.lrec-conf.org/proceedings/lrec2022/pdf/2022.lrec-1.194.pdf). In *Proc. of LREC*, pages 1824–1830, Marseille, France, June 2022.

For annotated corpora and language-specific annotation guidelines: see [this page](https://github.com/carmls/datasets).