Which Shakespeare Would You Trust?
Literature scholars spend much time and money trying to make sure that the texts they are studying represent the best, most accurate, nearest to the author's intentions version of the work they can find. Once scholars agree upon that "best" edition in any era, they also agree to use it when quoting the author of the work in articles and books. Sometimes huge changes come about in the scholarly edition that comes into use, like the discovery that the "First Folio" edition of Shakespeare's King Lear is missing very good and important passages found only in the earlier, "bad quarto" pirated edition published much earlier and without Shakespeare's permission by a freebooting printer. Now both editions are studied, and the Norton Anthology of English Literature prints a composite text based on the best of both, assuming that neither accurately captured what Shakespeare wanted performed. All of this, and much more Shakespeare scholarship has taken place in the last few decades of the twentieth century. Editions published before that are inaccurate and largely useless as a source of scholarly text, nevertheless, students using "Internet" editions of Shakespeare are unconsciously using those ancient, outdated editions of Shakespeare every day because they're free, easily available and can be searched quickly. Just how far out of date are those editions? The only way to know is to ask them what their source is, and many "seemingly reliable" editions, like those posted by MIT and by Project Gutenberg, turn out to be founded on ancient, inaccurate texts.
The MIT and Project Gutenberg Shakespeare texts both are based on Jeremy Hylton’s “Moby” edition, which is no longer available, itself, because the MIT server it was stored on crashed. The Moby edition is not around any more to tell us upon which paper text it was based, but, according to Terry A. Gray’s “Mr. William Shakespeare on the Internet” site (the only sort of "history" currently available for this kind of source), it appears to be based on the 1866 Globe edition, an even less reliable edition than the 1914 Oxford. (See the comment on the 1866 edition at http://shakespeare.palomar.edu/works.htm and Harry Connors’ answer to a query about access to the Moby edition on Shakespeare.com at http://shakespeare.com/queries/display.php?id=4395&replies=2.)
This is not to say you can’t find scholarly editions of WS online. If writers fish around a little more carefully—doing something as simple as a Google search on “Shakespeare scholarly edition,” for instance—they will find this site: http://ise.uvic.ca/ If you click on the “Foyer” icon and skim a bit of what they say about their editorial policies, you will see what the process that produces accuracy looks like.
Then look at “The Library” and at the bottom of the home page and you will see something important—how many scholarly editions have they actually produced and long has it been since they’ve added to this site, which is far from complete? Why are they moving so slowly? As Herman Melville famously exclaimed in the midst of Moby Dick, or The Whale, “Oh Time, Strength, Cash, and Patience!” (Or did he? That’s the way I remember if from my most recent re-reading about a decade ago so don’t quote me!) Someday, this will replace the Riverside Shakespeare, but don’t hold your breath.
Are old editions necessarily worse than new, but non-scholarly editions? The answer, surprisingly, is no. Sometimes an older edition can be better than a newer one like the nearly modern electronic "Moby," for instance Hardin Craig's 1914 Oxford University Press edition, now in public domain. Older does not necessarily mean “more primitive.” Victorian and early 20th-century editing practices set the standard to which all modern editions now conform, electronic or paper. More importantly, because they were dealing with so much less information, they had the time to do it right. Individual scholars labored for years correcting “proofs” or preliminary pages of type-setting from printers, and reviewers checked their work after publication, down to spelling and punctuation and line breaks in poems. Mistakes caught after publication are indicated by “corrigenda” slips tipped into the printed books listing pages and lines where corrections should be made by the owner of the book. That’s how careful they were. Modern electronic editions, even those checked carefully like those at U. Va.’s etext library, tend to be produced in such haste that they are not given similar care in production. Things are getting worse, not better, in the arena of textual accuracy, and this will affect any source of information you use on the job, any job.
So if you can’t find the Shakespeare text you need on the web at the University of Victoria (Canada) site, you have to use the Riverside Shakespeare to check the accuracy of what you’re hanging your reputation on. Use the ‘net for what it’s good for, cheaply and quickly locating information of doubtful quality, and then use paper sources for what they’re good for, applying quality control to what you’re using in public.