Much has been written in recent days about the CJEU's decision in the so-called Safe Harbour case C362/14 involving personal data. To most IP fans, 'safe harbor' (especially the American spelling of it) means something completely different, namely the degree of immunity from litigation afforded to internet service providers of various types by section 512 to the US Copyright Act 1976 and Articles 12 to 15 of the EU eCommerce Directive 2000/31/EC. Neither of these pieces of legislation actually uses the term safe harbour (the EU Directive prefers the near simile 'mere conduit') yet we, laymen and lawyers alike, instantly recognise what it refers to. It is a shorthand way of referring to a set of circumstances and conditions which are intended to lead to a particular legal outcome. I was tempted to write 'certainty' but that would have been foolhardy.
We rely on such shorthand phrases to make day to day discourse easier, but this facility is easily undermined when the same phrase gains currency in an entirely different sphere, as with the data protection example already quoted. There are a number of other examples where this duality can cause confusion, such as 'public domain' and 'derivative work' to name but two of them. These words can lack any legal definition, and indeed are often used in place of legal concepts too complex to be defined in a single sentence, hence their value as shorthand tags.
Public domain is probably one of the most pernicious phrases of this type. Its more general meaning of something which is accessible by the public can lead to the mistaken belief that such information or work is therefore also freely available to be copied. The classic type of example is "I found it using Google, so it's OK to use this image as it's in the public domain". This situation is not helped by the fact that, in an IP context, the more legally correct 'not copyright protected' is something that anyone can say or write in connection with a work without any legal consequence even if it is untrue. The same situation is also not helped when legislation has had the retroactive effect of restoring copyright to works which were previously out of copyright, often in a less than transparent way. Hence the concept of the public domain in the context of copyright lacks certainty. Given that primary copyright infringement is a matter of strict liability, it is strange that making a false claim to copyright or a false claim that copyright no longer exists in a work, are not sanctioned per se, while falsely attributing a work to an author (section 84 CDPA) is one of the moral rights that copyright law does see fit to protect. Arguably false statements like the former might be actionable under the tort of deceit but proving intent will rarely be easy, and in any case, the existence of a false statement to the effect that copyright no longer exists (or is waived to some degree by a false Creative Commons licence) will not be a defence against a claim of primary infringement.
Derivative as a means of describing a work is equally problematic, but for different reasons. In this case there is a legal definition under US Copyright law (section 101) to cover a very wide range of subsidiary works derived from an underlying work:
A “derivative work” is a work based upon one or more preexisting works, such as a translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, art reproduction, abridgment, condensation, or any other form in which a work may be recast, transformed, or adapted. A work consisting of editorial revisions, annotations, elaborations, or other modifications which, as a whole, represent an original work of authorship, is a “derivative work”.
but in most other jurisdictions there is no such corresponding legal meaning yet the term is still used, often to mean much the same thing as in the USA, by those in the UK or EU, because it is readily understood that the word means something based on a pre-existing protected work. But nowhere in UK or EU copyright law will you find the term 'derivative'. It is true that in the 2010 Systrans case (T-19/07) the General Court of the CJEU, perhaps unwisely, used the word (10 times) with this meaning, albeit in the context of software (the Software Directive 2009/24 merely refers to "translation, adaptation, arrangement or any other alteration ..") thus giving the phrase slightly more legal weight than it had had previously. The UK's Copyright Designs and Patents Act provides nothing so sweeping as the US definition; the nearest equivalent being 'adaptation' (section 21) which only applies to a very limited list of specific examples, and only then for literary, musical and dramatic works. The lack of a statutory application to artistic works means that the courts have to decide this themselves, with varying results. Article 3 of the German Urheberrecht Gesetz also uses the word adaptation (Bearbeitungen) but in a wider context than UK law. The Berne Convention (Article 1) uses the term derivative works as a sub-heading but has a fairly ambiguous description ('other alterations') of what it encompasses. The EU Directives largely avoid any direct definition of copyright works, let alone derivatives or adaptations, leaving this to national legal frameworks.
Why does this lack of a universally-accepted definition of 'derivative work' matter? The main reason is that the worlds of derivative works, adaptations, fair use (especially the transformative use part) and fair dealing (parody/caricature/pastiche) all intersect. What's more there are different approaches to whether the relationship between the underlying work and derivative work, such as whether permission is required, that in turn can affect the question of originality in the derivative work. There is now a wide variation between how the courts in the USA, UK and other EU member states, along with the CJEU, interpret and apply the law in this area. As far as harmonised copyright law is concerned, the EU generally is still in the foothills of settled law on matters surrounding derivative works, whereas in the USA there exists a large body of caselaw around transformative use, albeit often with conflicting outcomes, such that universally speaking, the term 'derivative' work is far from precise and is heavily dependent on the jurisdictional context.
As the internet becomes ever more central to many infringement cases, this lack of uniformity between the key players is, to put it mildly, unhelpful. Of course this does not undermine the basic value of such shorthand terms such as public domain, derivative work, or safe harbo(u)r in everyday usage, but I suggest litigants-in-person, advocates and the courts themselves should eschew their use in formal settings where this leads to imprecision.