The Shakespeare Conference: SHK 26.446 Tuesday, 6 October 2015
[1] From: Jonathan Hope <
Date: October 5, 2015 at 6:02:29 PM EDT
Subject: RE: EEBO-TCP
[2] From: Matthew Steggle <
Date: October 6, 2015 at 4:54:03 AM EDT
Subject: EEBO-TCP
[1]-----------------------------------------------------------------
From: Jonathan Hope <
Date: October 5, 2015 at 6:02:29 PM EDT
Subject: RE: EEBO-TCP
Pervez Rizvi is right that the EEBO-TCP transcriptions are far from perfect, but also right that we should not be too quick to moan about the outputs of this huge project, made freely available to all of us.
The transcriptions were made in the knowledge that there was a fixed amount of money available, and those managing the project tried to get as much done as possible with a reasonable level of accuracy.
As I understand it, transcriptions were dual-keyed by non-native speakers (who are actually more accurate at transcribing old-spelling than native speakers). Texts were sample-checked by TCP staff, and texts that fell below a set standard were re-keyed at the expense of the commercial sub-contactors.
Transcribers were told not to guess if they were unsure of a reading: aside from errors, there are many deliberate gaps in the transcriptions. They worked from EEBO digital images, which are (as many SHAKSPERians will know) often not easy to read.
In my experience, missing pages in the transcriptions are often indicative of gaps in the image files themselves (presumably when two pages were turned over by mistake in the photographing process) rather than errors on the part of the transcribers.
It is good to be able to report that many colleagues are mobilising to contribute to the field by helping to improve the transcriptions - perhaps most notably Martin Mueller, whose AnnoLex is a fantastic way to get students involved: http://annolex.at.northwestern.edu/about/
There will be a special plenary round-table on EEBO-TCP at the SAA next year in New Orleans where Martin will contribute, along with several other scholars who work closely with this material.
Thanks to Pervez for bringing this up - EEBO-TCP is an important resource, and it is also important that those who use it understand its limitations as well as its value (not least because they can then help to mitigate the problems)
Best
Jonathan Hope, Professor of Literary Linguistics,
School of Humanities,
Lord Hope Building
University of Strathclyde
[2]-------------------------------------------------------------
From: Matthew Steggle <
Date: October 6, 2015 at 4:54:03 AM EDT
Subject: EEBO-TCP
For Pervez Rizvi -
There’s a review of the debates around the accuracy of EEBO-TCP in my article, “The Cruces of Measure for Measure and EEBO-TCP”, Review of English Studies 65 (2014): 438-455. It’s free-access till the end of the year from http://res.oxfordjournals.org/ .
All the best,
Matthew Steggle
Professor of English
Sheffield Hallam University