Published in Nexus (Newsletter of the Australian Sociological Association), Vol. 16, No. 2, June 2004, pp. 15-16
pdf of published article
Brian Martin's publications on fraud and plagiarism
Brian Martin's publications
Brian Martin's website
Several Australian universities are proposing to introduce use of plagiarism-detection services, specifically turnitin.com, for checking student essays. Having studied plagiarism issues for over 20 years,  I decided to look at educational rationales for using such services, especially (1) deterring and detecting cheating, and (2) fostering learning of proper acknowledgement practice. A wider treatment would also cover implications for workloads, intellectual property and institutional reputation.
Plagiarism involves claiming credit for ideas or creations without proper acknowledgement. In an academic context, acknowledgement is typically given in the form of citations or explicit statements of thanks. This is important for several reasons, including to give credit for ideas or words, to provide support for one's argument, and to show that one is aware of sources. To speak of proper acknowledgement is to focus on the positive side of scholarly practice; to speak of plagiarism is to focus on the negative.
In most cases, software for detecting plagiarism can detect only word-for-word plagiarism for those documents in its database. It cannot detect plagiarism of ideas or plagiarism of authorship unless they also involve detectable word-for-word plagiarism. Students who take ideas from others but express them in their own words will not be detected. Nor will students who purchase custom-written essays. Nor will those who copy from sources not on detection databases, such as many printed texts, CD-ROMs, certain subscription databases and the deep web, or who use translations of documents. 
The positive side of plagiarism-detection software is that it can be used to detect students who attempt to cheat by using online sources rather than doing their own writing. If students know that their essays might be checked this way, they may be deterred from this form of cheating.
Cheating by students is undoubtedly a major problem, as attested by various surveys. Plagiarism is one important mode of cheating, though cheating occurs in all forms of assessment.  Widespread student plagiarism predates the Internet but electronic sources have made the practice far easier.
Many academics believe that they can pick up plagiarism, but in most cases they can detect only a small proportion of what occurs. Thorough checking for plagiarism is incredibly labour-intensive. One article on the topic, pre-Internet, recommended reading student essays four times each in order to detect plagiarism.  Plagiarism-detection software automates much of the process.
Plagiarism-detection software has a number of shortcomings. Most obviously, not all sources are included in databases. There is no check for plagiarism of ideas and no conceivable check for false authorship, as when students submit essays specially written for them by someone else. In these circumstances, a software check may give a false certificate of probity.  Students may even be stimulated to use other innovative methods of cheating.
Plagiarism-detection software should be compared to alternative methods of preventing cheating.  One is to design assignments so that plagiarism is difficult, for example by requiring students to link their topic to current events or to activities in the classroom, for which no Internet or other sources are available.  Another way to reduce cheating is by fostering adherence to an honour code in which students pledge not to give or receive assistance, and to report violations by others. Using plagiarism-detection software, with its presumption that cheating is tackled by screening essays, may discourage initiatives along these lines.
Quoting, paraphrasing and citing sources appropriately is something that has to be learned: it is neither obvious nor automatic for people new to writing. Scholarly acknowledgement practice can be likened to etiquette: doing the proper thing according to standards suitable for the occasion. This way of thinking about the matter focusses on learning.
There are various ways to foster learning of any social convention. One is the punitive approach, with severe penalties for transgressions. Research in learning shows that this approach is usually far less effective than encouragement of good practice, through modelling appropriate behaviour, regular practice and rewarding successful performance.
Much if not most plagiarism in student essays is due to ignorance, sloppiness or panic rather than an attempt to cheat.  Most students treat proper acknowledgement practice seriously  ; some are mortified when informed that they have done things inappropriately.
In line with this way of thinking, some teachers treat acknowledgement practice as something to be learned like other scholarly skills such as giving seminars or carrying out experiments. Others, though, treat plagiarism as a serious transgression, akin to a sin, deserving of the most severe penalties.
Plagiarism-detection software can play a role in fostering proper acknowledgement practice by alerting teachers and students to passages that are incorrectly quoted or insufficiently acknowledged. It can also frighten students about being caught plagiarising and hence stimulate them to learn proper practice.
Plagiarism-detection software also can have a negative effect on learning. If used on a blanket basis, the presumption is that every student is a potential cheat. This can discourage an openness to learning and instead foster an attitude that whatever gets through the system, such as plagiarism of ideas, is okay.
If students trust their teachers - to help them learn, and not to penalise them unfairly - they are much more likely to put energy into their studies. Universal plagiarism-checking implies a lack of trust in students that will be reciprocated by some of them, with negative consequences for learning. 
Some teachers, believing the punitive approach to be pedagogically unsound, may decide not to follow formal procedures for reporting plagiarism, especially if the procedures are cumbersome.  Some may choose not to take notice of suspected plagiarism.
When students are asked to satisfy high standards of acknowledgement practice, it is reasonable that they expect similarly high standards of university staff. But there are many instances of "institutionalised plagiarism" - plagiarism that is accepted, often as part of the institutional hierarchy - that reveal a double standard.  There are many stories of lecturers who "borrow" material for their subject notes from colleagues and who present material in lectures drawn from unacknowledged sources. Memos are regularly circulated by university officials under their own names, even though the text was written by someone else. Many university documents do not specify authorship accurately. Students may well ask why they are expected to adhere to standards not followed by those who teach them and administer their education.
 . This is an abbreviated version of a longer article circulated at the University of Wollongong. I thank Robert Briggs, Stewart Russell and especially John Royce for valuable comments on a draft of this paper.
. Full text of most of these is available.
. John Croucher, Exam Scams: Best Cheating Stories and Excuses from around the World (Sydney: Allen and Unwin, 1996); Harold J. Noah and Max A. Eckstein, Fraud and Education: The Worm in the Apple (Lanham, MA: Rowman & Littlefield, 2001).
. John Royce, "Has turnitin.com got it all wrapped up? (Trust or trussed?)," Teacher Librarian, Vol. 30, No. 4, April 2003, pp. 26-30, surveys four investigations of turnitin.com and says "The bottom line is that innocent students may be falsely accused of plagiarism, and that many plagiarists may go undetected."
. This is recommended by a number of authors, for example Royce, "Has turnitin.com got it all wrapped up? (Trust or trussed?)"; Robin Satterwhite and Marla Gerein, "Downloading detectives: searching for on-line plagiarism," http://www2.coloradocollege.edu/Library/Course/downloading_detectives_paper.htm, 2002 (accessed 10 June 2004), state "As with many of the sources we consulted in our literature review, we recommend instead spending time and energy on proactively avoiding plagiarism in the first place, rather than trying to detect it after the fact."
. Robert Briggs, "Shameless! Reconceiving the problem of plagiarism," Australian Universities' Review, Vol. 46, No. 1, 2003, pp. 19-23, argues that a moralistic attitude towards plagiarism can be counterproductive for learning and even inhibit deterrence and detection of plagiarism.
Some editorial writers have highlighted trust as a key issue, for example "Catching the copycats: fighting plagiarism must not spoil the university experience," Ottawa Citizen, 20 October 2003, p. A14: "Plagiarism is a scourge that must be confronted. But in doing so we must be careful not to poison the student-teacher relationship and sour the university experience."
. Brian Martin, "Plagiarism: a misplaced emphasis," Journal of Information Ethics, Vol. 3, No. 2, Fall 1994, pp. 36-47.