Next time you’re walking out of a German supermarket with 302 items hidden in your coat that you didn’t pay for and you are faced with some uncomfortable questions by a security guard try the following: say that the suspicion of theft is absurd. Then, after your loot has been tallied by the security guard, admit that there may have been a few mistakes in your shopping procedure, but that you were incredibly busy thinking of other stuff, and stick to your original claim that the suspicion of deliberate theft remains an absurd one.
At least this seems to be working for German defence minister Karl Theodor Maria Nikolaus Johann Jacob Philipp Franz Joseph Sylvester Freiherr von und zu Guttenberg, who, even after 302 instances of copied or directly translated material have been spotted in his dissertation by a crowdsourcing wiki still denies having committed deliberate plagiarism. According to the wiki, over 20% of all lines of text in the dissertation have been taken from other publications without giving proper credit. The university has taken away his doctorate title yesterday, but stopped short of alleging deliberate plagiarism. And Guttenberg is adamant that there was no intent of fraud.
The above graph visualizes the findings of the wiki’s ‘intermediate report.’ On the x-axis are the pages. The table of contents and appendices are colored in blue. Black marks pages on which suspected plagiarized material has been found, red marks pages on which multiple instances of plagiarism have been found. Yellow is used for a substantial part which was probably ripped from a report prepared by employees of the parliament. White, finally, is used for pages on which there have so far not been any suspicious findings. The percentage of pages that are not white (excluding the blue pages in the total) is close to 70% (although the percentage of more than 20% of the lines seems more meaningful–not that the difference in percentages really matters).
Update March 3:
Guttenberg has resigned and admitted that his dissertation was ‘fehlerhaft’. The word means something like ‘it contained mistakes’, and clearly implies that there was no intent. This is a nice euphemism for what anyone who has looked at the findings on the wiki would call deliberate plagiarism using various tricks to conceal the original sources (either by Guttenberg or a ghostwriter). The resignation speech was emotionally charged, and did not stop short of enlisting fallen German soldiers in an attempt to raise anger against the naughty media who wouldn’t let go of this story, but was carefully crafted to only admit what even the strongest supporters of Guttenberg had to come to accept was true. The University of Bayreuth is still deliberating whether they can safely conclude that this was intentional plagiarism. The successor of Guttenberg’s advisor at Bayreuth has in the meantime called a spade spade, or rather, a con man a con man.



crowdsourcing workshop
Talking about crowdsourcing, this looks like it’s going to be an interesting workshop at this years LSA institute:
Full Title: Crowdsourcing Technologies for Language and Cognition
Date: 27-Jul-2011 – 27-Jul-2011
Location: Boulder, Colorado, USA
Contact Person: Robert Munro
Web Site: http://www.crowdscientist.com/workshop/
Linguistic Field(s): Computational Linguistics; Text/Corpus Linguistics
Call Deadline: 01-Mar-2011
Meeting Description:
This workshop will bring together linguists who are utilizing crowdsourcing technologies and those who want to know more about them. It combines a half-day ‘how-to’ session where participants will learn to conduct experiments using crowdsourcing platforms and a half-day workshop where researchers come together to share results, ideas, and strategies.
It is being held in conjunction with the 2011 LSA Summer Institute.
Call for Papers:
We are eliciting abstract submissions for afternoon presentations. They should present either:
a) Novel empirical results in linguistics or the language sciences that have been enabled by internet-based crowdsourcing technologies, or b) novel approaches to data collection and evaluation, especially when there are no ‘correct’ responses to a given stimuli. Crowdsourcing technologies could include technologies like Amazon’s Mechanical Turk, CrowdFlower, online games, collaborative platforms like instant messaging and Wikipedia, or custom software built for collecting language and cognition data. Talk slots will be 10-15 minutes, depending on scheduling.
Please send 250-word abstracts to crowdscientistgmail.com by March 1st, 2011. Do not include author names or affiliations in the abstracts as reviewing will be blind (but list the authors and affiliations in the email itself). Abstracts can be in plain text or PDF. You may optionally include figures, tables and/or references. They will not count towards the 250 words, but please limit the entire submission to one page.