Book People Archive

Re: copyright of photoreproduced text

Ben Crowell wrote:
>>>    Does anybody know the extent to which a photoreproduction
>  of an uncopyrighted text can be copyrighted?
> >   It's protected to the full, as a photo is a new artistical work.

> I would add a couple of caveats. For one thing, in the US, a
> legitimate defense against a copyright infringement lawsuit
> is that you were unaware the material was copyrighted. 

This is the "innocent infringement" defense.  You can still be
held liable for some damages even if you infringed innocently,
but they can be considerably less than for willful infringement.
Note that, for this defense to work, you have to prove that you
neither knew, nor had good reason to know, that the work was
copyrighted.  (So, to take a somewhat silly example, you can't
simply go out of your way to avoid looking at the copyright page
and then claim you "didn't know it was copyrighted" as a defense.)

> Another thing to consider is that there is some case law in
> the US that says that you can't copyright something that
> requires no originality to create, like a mere compilation of data.
> [...] So I really have to wonder whether
> merely scanning a PD document results in a copyrightable work.
> Obviously there's a big difference between that and an
> Ansel Adams photo...

Indeed.  In fact, in the US there was a decision in 1999
where a court found that Corel had *not* abridged copyright by
copying photographs of public-domain paintings, because such
photographs, intended to exactly duplicate an existing image,
were not sufficiently original to be copyrightable.  The decision,
in _Bridgeman Art Library v. Corel_,  can be found on-line at

  http://www.law.cornell.edu/copyright/cases/36_FSupp2d_191.htm

See paragraphs 19-25 for the court's reasoning about why those photos
were not copyrightable under US law.  (In paras. 26-39, they also
argue that they weren't copyrightable under UK law either, but since
they aren't a British court, UK citizens might not want to rely on it
as a source for legal advice.)

If exact copies of public-domain paintings aren't copyrightable in the US,
it would seem to follow that exact copies of public-domain book pages 
aren't copyrightable in the US either.   (Especially since, to the best
of my knowledge, under US law typesetting itself usually isn't considered
original enough to be copyrightable anyway.)  However, the rules may be
different in other countries.  (One of the quotes above, for instance,
was from a poster in Sweden.)

> I just feel that the people who copyright scans of PD documents
> are antisocial, and deserve to be treated that way. For instance
> Cambridge or Oxford (I can't remember which) has scans of a
> manuscript of the Song of Roland on the web. They claim it's
> copyrighted...

Actually, the manuscript as an *image* might be copyrighted, if the image
hasn't been published or publicly displayed until recently.  In
many countries, including the US, unpublished works can be
copyrighted for longer periods of time than published works.  
In fact, the US has traditionally kept unpublished works under
perpetual copyright as long as they remained unpublished.  (Once
they were published, normal copyright terms applied from the
point of publication.)   Recent revisions to copyright law have
changed this, but copyright expirations aren't scheduled to start
until the New Year's Day 2003.  At that point, anything that hasn't
been published prior to that date, by authors who died before 1933,
will enter the public domain in the US.   I'm hoping *that* date doesn't
get postponed as well.  (Fortunately, most of the entertainment
companies that rammed through the Bono copyright extensions don't
have much use for this material, compared to published books,
music, movies, and the like.)

Note that the *text* of the Roland manuscript is probably still
in the public domain regardless, since the text presumably was
published long ago.  But if the manuscript itself has only
surfaced in public recently, then whatever's unique to the manuscript,
and that has some originality (including things like the handwriting and
decoration on the MS), may still be eligible for copyright.

The rest can be used at will, though.   (And fair use, of course,
still applies even to the copyrightable portions.  For instance,
if a scholar wants to argue that a particular word in the poem was actually
different from the word usually assumed, they could include some
images from the manuscript, and argue, say, that a particular letter
was clearly written as "o" rather than as "a".)

John (I'm not a lawyer, etc.) Mark Ockerbloom