The ComparaLex Team

Keith Snider – Director
Larry Hayashi – Assistant Director
Doug Rintoul – Site Administrator
Wayne Dirks – Website Developer

Why ComparaLex?

Given that comparative word lists are organized differently, comparing data from one list with data from another list is not always easy. For example, while the word for "child" is number 203 in the Comparative African Wordlist that is used for many Kwa languages spoken in Ghana, the word for "child" is number 66 in the Benue-Congo Comparative Wordlist that is used for many Benue-Congo languages spoken in Nigeria. This, of course, makes it difficult for anyone who wishes to use data from both sources in order to carry out comparative reconstruction in Niger-Congo languages.

ComparaLex includes an online database that cross-references comparable items from different standard word lists. All standard list items are indexed against a master conglomerate list whose items have been categorized using a standard semantic domain hierarchy. This makes it easy for linguists to compare language data that has been collected against different standard word lists. They can both upload their lexical data to the database as well as download comparative data sets.

Principles of Operation

Making the data upon which scientific conclusions are based available to other researchers is an essential part of the research process. Results cannot be verified or improved unless we provide our colleagues with this information. The ComparaLex team, submitters and all users agree to the following principles with respect to data submitted to ComparaLex:

• Submitters provide data in electronic format for use by others (to the extent reasonably possible).
• Users acknowledge any use made of the data obtained from ComparaLex according to the normal conventions of scholarship.
• Users make no commercial use of data obtained from ComparaLex, except by prior and specific written agreement with the submitters of the data.
• The ComparaLex team will post data to its database server and will present to database users all acknowledgements and copyright restrictions provided by the submitter.
• The ComparaLex team is held harmless for any errors or misuse of data that might occur.

ComparaLex will post data to its database server and will present to database users acknowledgements and copyright restrictions provided by the submitter.

Adapted from the CBOLD Bantuists' Manifesto.

History

ComparaLex had its early beginnings in a presentation by Keith Snider at an SIL Africa Area Linguistic Consultants Seminar, held in Ruiru, Kenya in 1995. At the time, there was a perceived need for SIL field teams to format their lexical data in a manner that better lent itself to comparative analysis. Ideas spawned from discussions at that meeting led to the creation of the 1700 item SIL Comparative African Word List (SILCAWL) by Snider and Jim Roberts. This list would provide linguists with a basis for easy comparison across a broad range of word items. SILCAWL was developed and tested by SIL teams over a number of years and published in its current form in 2004 (Snider, Keith and James Roberts 2004. SIL comparative African word list (SILCAWL). The Journal of West African Languages 31.2, 73-122. (Also published in 2006 on sil.org)).

It was also the desire of SIL Africa Area to create a data repository to facilitate linguistic comparison and to make SIL’s work maximally available and maximally useful to the broader linguistic community. Under the direction of Keith Snider, Dan Duke of SIL Cameroon began cross-referencing many of the existing elicitation lists that were being used by SIL and other linguists in Africa to elicit language data.

The development of the comparative data repository, however, did not actually begin until 2005 when Keith Snider and Larry Hayashi began working on a database at the Canada Institute of Linguistics, on the campus of Trinity Western University. This database would incorporate the cross-referencing previously done in Cameroon and would allow linguists to submit data that been collected against any of the standard word lists included in the database. As part of coursework for Dr. Alma Barranco, a group of TWU Computer Science students (Antonios Angelatos, Philip Miguel, Tae Yeon David Kim, and Joel Hendrickson) did preliminary requirements analysis and began to build a web-based system for the repository. At about this time, Christina Page cross-referenced additional standard word lists. An alpha version of the database was presented at two linguistics conferences in 2006 (Snider, Keith and Larry Hayashi 2006. "The Comparative African Word List Project" presented at the 37th Annual Conference on African Linguistics, University of Oregon, Eugene and at the 5th World Congress of African Linguistics, University of Addis Ababa, Ethiopia (Aug. 7-11, 2006).

Doug Rintoul of the CanIL IT department, transplanted the database from the TWU student project site to a dedicated subdomain on the CanIL site. In late 2006, Wayne Dirks, who was working with SIL Africa Area in Nairobi at the time, took over the website development and has brought the implementation to its current state. In 2008 the word list database and website were christened “ComparaLex.” In the final stages of getting ComparaLex ready for public release, Cam and Valerie Hamm from SIL Cameroon edited errors found in standard word lists, tested the interface and provided feedback to streamline the interface.

Acknowledgements

The ComparaLex team wishes to thank the many people, some of whom are mentioned above, who have contributed significantly to the development of ComparaLex. In addition, we wish to acknowledge the following:

• Larry Hyman and the CBOLD project for providing the Guthrie, Meussen, and Tanzanian Language Survey word lists in electronic format. Hyman also provided his Proto-Bantu Grassfields Glosses, which are also being used as a standard word list.
• The late Lionel Bender for providing us with his Comparative Nilo-Saharan Word List in electronic format.
• The Dallas SIL library for providing us with Kraft’s Comparative Chadic Word List in electronic format.
• Gudrun Miehe (University of Bayreuth) for providing us with the Beyreuth Grundwortliste and Bayeuth Kulturewortliste in electronic formats.
• All past and future contributors of data. Without their data, this endeavor would be quite meaningless

ComparaLex continues to be a work in progress. As stated above, the original vision was to facilitate research and to make SIL Africa Area work maximally available and useful. We hope that other linguists will also contribute their data and find the site useful for research. We welcome your comments and suggestions.

Links

In developing this site we have made use of a number of high-quality, open-source and freeware software packages including:

• Apache HTTP Server httpd.apache.org
• PHP script language php.net
• MySQL database mysql.com
• jQuery jquery.com
• jQuery UI jqueryui.com
• jQuery autocomplete github.com/dyve/jquery-autocomplete
• XSPF Web Music Player musicplayer.sourceforge.net
• Nuvola Icons icon-king.com
• iCandy Icons iconaholic.com
• Javascript HMAC code pajhome.org.uk
• Programmer's Notepad pnotepad.org
• DejaVu fonts dejavu.sourceforge.net
• Countless bits of code from the helpful people at php.net/manual

Canada Institute of Linguistics

CanIL provides training for the linguistics departments of Trinity Western University and Tyndale University College. CanIL is a partner organization of SIL International. CanIL offers the following degree programs: BA Linguistics, MA in Linguistics and Translation and Masters of Applied Linguistics and Exegesis in addition to a number of specialized training tracks, workshops and a summer program.

Address
7600 Glover Road
Langley, B.C. V2Y 1Y1
Canada

tel: 604.513.2129
fax: 604.513.2128
email: Use contact form: LINK.
web: www.canil.ca