Inferring access when establishing a copyright infringement case

Once, Silent Witness was one of the better crime dramas on television, but as some point a few years ago (after changes to the cast had already made it a shadow of its original self) it seemed to lose its connection to reality. I have not watched it for many years now, and the claim that an episode infringed copyright in an earlier screenplay made me wonder whether claiming even such an association with the show wouldn’t be detrimental to the purported author’s reputation. But there you are – that’s probably just me being a grumpy old man.

Molavi v Gilbert & Ors [2023] EWHC 646 (Ch) (23 March 2023) is an unusual case, in that it came before the court on an application by the defendants (which included the BBC and BBC Studios Production Ltd as well as Virginia Gilbert, the named writer of the accused episodes of Silent Witness) for summary judgment, on the grounds that the claimant had no real prospect of succeeding in her claim – Part 24.2 of the Civil Procedure Rules. And indeed it is hard to see, on the basis of the reported judgment, how the judge’s (Marcus Smith J’s) conclusion could have been otherwise.

The case is an important one because copyright infringement claims so often rely on inferences. Direct evidence of the defendant taking the claimant’s work is unlikely to present itself. The claimant’s job is to satisfy the court about three matters: first, that the work is their original work; second, that the defendant has engaged in copying (or some other infringing act); and third, that what has been taken is a substantial part of the work.

The second part of that list is the one that matters here. The defendants accepted that the claimant’s work was original, and that if there had been copying it was of a substantial part of it, so the judge did not have to worry about those issues. It all came down to whether there had been copying at all.

The defendants said that the claimant could not show that they had had access to the work, and without access copying would simply be impossible. The claimant could not say how her work had been copied, because of course she was not privy to the process by which the defendants’ screenplay had been written: she had to rely on the inferences that copyright law, recognising that it is unrealistic to expect the claimant to be able so show definitely what happened, allows to be drawn. There must be objective similarities between the original work and the accused copy, and the defendant must have had an opportunity to copy – if these conditions are satisfied, it is up to the defendant to explain away the similarities.

In the present case, it’s almost as if a further level of inference is brought in. The claimant had no evidence that the defendants had ever had an opportunity to copy her work, so pointed to the similarities and argued that their very presence in the defendants’ screenplay suggested that copying had taken place. The similarities could not, the claimant argued, be down to coincidence.

That reeks of desperation, though clearly there will be cases where the claimant is rightly desperate. If the alleged copying is literal, I can see that the court might be prepared to draw an inference – if the defendant has come with something that is word-for-word the same as the claimant’s work, it would be an extraordinary coincidence if there were no copying. Most cases, however, involve non-literal copying, and here it is plain from the judge’s analysis of the respective works that they weren’t very similar. The timeline also worked against the claimant (the basic premise of Silent Witness had been established long before the claimant wrote anything).

Of course in the field of non-literal copying we have to be wary of the idea-expression dichotomy, lest ideas be given copyright protection to which as a matter of policy they are not entitled. When considering an inference of copying, the court has to take great care to ensure it does not overstep this boundary, so when the allegation is of non-literal copying the claimant is going to have an uphill struggle to convince the judge to infer that copying has taken place.

It’s also very important to remember that you can’t break a copyright work down into its component parts and compare them against the elements of the alleged copy. In that great old case, Ladbroke (Football) Ltd v. William Hill (Football) Ltd [1964] 1 WLR 273 at 277, Lord Reid warned us that this approach can lead to the wrong result. UK copyright law tends to look at works as a whole (in the US, a rather different approach is taken) rather than chasing similarities in the components of the works.

Establishing that copying has taken place is never exactly easy, unless there is a smoking gun in the form of identical choices of words or something like that. We probably didn’t need this case to remind us of that, but it is always useful to be reminded of points like this that may determine the outcome of ligitigation.