So it happened like this. I noticed that the Gutenberg Project version of Mary Shelley’s Frankenstein didn’t indicate which edition – 1818 or 1831. The two are rather different, as the author rewrote whole passages. Example: I just read a piece by Brian Aldiss in which he asserts, in passing, that H.G. Wells must have been misremembering when he wrote that “Frankenstein, even, used some jiggery-pokery magic to animate his artificial monster.” But: this would only be clearly wrong concerning the 1818 version, which is actually more ‘scientific’. An appendix to a recent edition of the 1818 edition notes: “the 1831 reader is allowed to think that the faculty at Ingolstadt [where young Victor gets his schooling] in the 1790s, even the previously sympathetic Waldman … were indeed teaching arcane magic under the name of natural science.” That’s sort of debatable, as a reading of the 1831 edition, seems to me. But it probably explains Wells’ impression.

Anyhoo. The Gutenberg version is definitely 1831. But, since Gutenberg editions are – Frankenstein-fashion – cobbled from the corpses of works passed into the public domain – it seemed possible that it was maybe a bit of both. Also, there are lots of typos (you get what you pay for.) Mostly just misplaced commas, colons and semicolons, but hundreds of those. (I’m convinced that the main export of Victorian England was the overused semicolon. Seriously, Shelley’s punctuation is bizarre. What’s with all the colons followed by dashes? Is that really necessary? Oh well.) So for the last few weeks I’ve been working through it, a chapter a night, with a public domain (1912 Everyman) edition of the 1831 edition in hand. I’m up to chapter 18. When I’m done I’m going to make a nice public domain edition. (Maybe do a book event.)

So here’s my question. When I’m done, I’d like to double-check it against an appendix to a still-in-copyright 1818 edition, which gives all the differences between the 1818 and 1831 editions. On the one hand, I really ought to be working from public domain material. On the other hand, I’m not exactly copying this appendix; merely verifying the correctness of text I’ve independently produced. Suppose I end up adding, subtracting or shifting 100 characters worth of punctuation, thanks to consultation of this appendix? Would that be a violation of copyright? Seems a bit weird if it is. 100 character total ought to be fair use, right? There must be copyright traditions concerning editorial questions like this, yes? Am I allowed to treat the appendix as containing information I am allowed to use freely?

UPDATE: Ben Wolfson has related, deep thoughts: “Punctuation marks can be very expressive, especially em dashes (my favorites!—maybe tied with semicola), so why ought one restrict their use to single isolated occurences? Surely in combination they can achieve heretofore undreamt-of degrees of subtlety in expression. (My gloss on the comash was that it implies a degree of reticence or hesitancy, and then:—suddenly elsewhere, or the dam is burst.)”