The third edition of MLA Style Manual and Guide to Scholarly Publishing has made several significant adjustments to keep pace with the ever-expanding realm of web scholarship.
In a major shift, the new MLA manual – considered the authority for most of the humanities – no longer considers print to be the default medium of publication.
“Given the ubiquity of electronic sources in contemporary research, it is now clear that we can no longer assume a common, default medium for the sources of scholarship,” writes David G. Nicholls, director of book publications for the Modern Language Association, in the preface. The medium of publication – whether web, print, television, radio or other – must now be included for every entry in a works-cited list.
Many observers agree that this is a good step. “I think that’s actually very forward thinking by the MLA,” says Aimée Morrison, a professor of English language and literature at the University of Waterloo. She says the move places print and web on equal footing, and she approves of the shift away from a print bias.
Geoffrey Rockwell, professor of philosophy and humanities computing at the University of Alberta, says that with the scales now slowly tipping toward the web as default source, putting guidelines in place for clearly identifying the medium is good policy.
Web sources must be searchable
Perhaps the most controversial change in the third edition, released in May, is that in most cases the MLA no longer recommends the inclusion of URLs when citing web sources. In this, the MLA parts ways with many other scholarly style manuals, and with its own previous edition.
Arguing that URLs provide “information of dubious usefulness,” the manual observes that they can be specific to a subscriber or session, often change over time, and can be so long, complex and cumbersome that they are commonly transcribed erroneously.
Instead, the manual guides scholars to give all the traditional publication information, which should allow anyone with the wherewithal to plug those details into Google to find the document or page. It allows for the inclusion of URLs “only when the reader probably cannot locate the source without it or when your publisher requires it,” and gives instruction on how to do this.
Dr. Morrison from Waterloo says that this change makes sense. “I think they’re absolutely right that URLs aren’t persistent. These new guidelines acknowledge that it is probably easier in most cases to Google some combination of the author name and the title to find electronic resources than it is to type in what can be more than 150 characters of gibberish as a URL.”
Dr. Rockwell agrees. “If the point is to help people easily recapitulate and connect with the research that you found interesting enough to quote, then use the tools you have at hand, and … give enough information that someone could double-check that they have found the right thing.”
One of the complications with URLs, he notes, is that porn sites often snap up domain names that aren’t renewed.
“It’s a little bit like, if I was referencing Hamlet and giving it a location, saying ‘Look for the book on the third shelf of the bookstore on White Avenue between 102nd and 103rd.’ And that bookstore closes down and an adult entertainment store opens up. All of a sudden the thing on the third shelf, ten feet in, is something else.”
However, some observers harbour serious concerns.
At the University of Prince Edward Island, university librarian Mark Leggott and emerging technologies and metadata librarian Melissa Belvadi say the new MLA approach could lead to problems in the area of authenticity. While a search engine may take you to the authoritative, genuine article, it could just as easily deliver something else – perhaps a fraudulent or poorly reproduced page, or a less-than-definitive version of the research.
Although she feels that the MLA’s issues with URLs are legitimate, Ms. Belvadi argues that “throwing away the URLs and leaving the researcher to the vagaries of the search engines I don’t think is yet the right solution.”
And indeed, it doesn’t appear that URLs will disappear from the humanities just yet.
While Holly Keller, manager of production and editorial services for the University of British Columbia Press, says the MLA is moving in the right direction, she’s still not ready to drop the URLs. The more an author can provide, she says, the better.
“The key thing is getting researchers to cite more than just the URL, to put as much information as possible about the source. We’re always nagging, saying, ‘more details, more details please.’”
And that, she says, will continue to be the practice “until everything settles down and there are some conventions set in this area.”
Hmm, this issue of persistence is indeed an interesting one. Research and academic opinion needs a solid foundation. References need to be verifiable both as to existence and content.
In the old days, ten years ago, content was encoded in the website itself. Modern web techniques use an object-oriented approach. They have a place on the site which points to a location where information is stored. They are designed to rapidly swap out content. So, what if the web site now points to another file and the old file is gone?
Carrying on the analogy of the bookstore, what if the play Hamlet was not to be? That is, what if it were no longer published? A Google search would not find it, even though it once existed.
If information can be so ephemeral, then it can’t form the basis for reasonable discourse. One possible solution would be the web equivalent of the Library of Congress where one copy of every webpage could be stored. This would never work as a centralized function – it is contrary to the nature of the Web. But there is no reason why the LoC couldn’t provide the disk space and bandwidth and browser designers couldn’t provide a tool to register a page with the LoC. That way, the onus is on the researcher to click the “Register with LoC” button. Failure to do so leaves the reference into doubt.
That still leaves the questions of authenticity and duplication. It strikes me that having a unique identifier for content similar to the ISBN number on a book would be what is needed. This may be an issue for the International Standards Organization to resolve. An authenticity certificate could be stored in the metadata of the website but then that would be up to the publisher to obtain. How would a researcher go about checking authenticity or requesting that a site be certified? My solution doesn’t seem complete.
The problem of duplication is that popular sites may be logged into the LoC hundreds of times which is wasteful of disk space and search resources. A subset is what if the duplication was only 99%? What if an article is posted in one place saying: “This proves aliens do not exist” and in another place “This proves aliens do exist”? Again some ideas on solutions but nothing comprehensive.
Perhaps the MLA could host a seminar on possible solutions. If the web is to be a primary source of research, the MLA should influence the designers of the web to consider referencing needs when they write their programs.