Blog post by Stephen Wyber
Copyright reform is again on the horizon in Europe. Not only is the 2019 Directive on Copyright in the Digital Single Market under evaluation, but the European Commission has also called for evidence about a series of potential reforms and, just yesterday, 29 September, launched a targeted consultation.
One idea on the table is a possible reform of the Union’s copyright exception for research, to facilitate researchers’ work when engaging with copyrighted materials. There will be plenty of discussion about the precise wording of this in the months and years to come.
In this context, we will see the contours of a broader question: what are the aspects of copyright law that need to be avoided if we want to uphold the interests of users, the institutions they rely on, and society more broadly?
These legislative ‘poison pills’ sharply reduce the scope and beneficial impact of otherwise potentially beneficial legislative provisions. When engaging in copyright reform discussions, it is therefore valuable to look out for these and highlight the threat they pose early on.
Here are 7 such poison pills:
💊 Pill 1 | Making Exceptions Optional: A key flaw in the 2001 Information Society Directive, still the most extensive legislation on EU copyright law, is that almost all of the exceptions provided for are purely optional. This means that Member States often simply do not apply them.
While in the short term, the ones who lose out are researchers and libraries within that country, this can also have a wider effect, complicating research collaborations with institutions in other countries. Conversely, there is a network effect to harmonised copyright exceptions, with each new country implementing them also increasing the size of the pool of researchers with whom everyone else can collaborate more easily. A mandatory exception is the way forward.
💊 Pill 2 | Limiting Exceptions to Non-Commercial Uses: A further problem comes with the limitation of the InfoSoc research exception to non-commercial purposes. This directly contradicts the EU’s broader emphasis on promoting knowledge valorisation, not least through collaborations between research and industry. It is also a relatively recent invention, only introduced in 2001, which forced some countries to narrow down existing exceptions that simply allowed for “research” irrespective of the purpose.
The impact of restricting it to non-commercial uses is that when a library knows that a researcher is involved in a commercial project, they must refuse to serve them, or navigate further licences to do so. Neither option is a win – Europe needs to remove this distinction.
💊 Pill 3 | Allowing Private Contracts to Override Legislation: While in the physical world, a simple exception may be enough, the challenge with digital is that access is provided under licences. While these can be permissive, going beyond the minimum that the law allows, more frequently they are much more restrictive.
A recent report commissioned by KR21 shows quite how often this happens, either through explicit override or by leaving out key research uses from an exhaustive list of permitted acts. Not only does this bring the immediate frustration of not being able to use licensed works to fulfil typical library missions (such as inter-library document supply), but the sheer volume of different potential contract terms represents a significant administrative burden which draws resources away from serving users and supporting research. More countries need to follow the examples of Ireland, UK, Belgium, Portugal and Bulgaria, among others, and make clear that contracts cannot override exceptions.
💊 Pill 4 | Offering No Clear Means of Removing Digital Locks: EU copyright law has two weaknesses when looking at technological protection measures (TPMs). First, the rule that allows people to get around a digital lock to make a lawful use does not apply to licensed works. Second, even when people do have that right, the EU has no effective way for them to exercise it in a timely manner.
This has a major impact on the behaviour of researchers and libraries, obliging them to change their plans or take the time to look for TPM-free alternatives. They can also impose significant costs in order to find means of circumvention, with little legal certainty (see, for example, the report from CREATe, and work by LACA in the UK. Clear rules and tools for circumvention are needed.
💊 Pill 5 | Making Compliance Next to Impossible: Looking outside of the pure research space, a strong lesson in how to undermine the effectiveness of legislation comes from the 2012 Orphan Works Directive. This came with the promise of opening access to the large number of works that sit in library collections for which it is not possible to seek the necessary permissions from rightholders because they are unknown or uncontactable.
However, the legislation ultimately included a list of sources to be consulted before a work could be declared orphan, including some available only in person or on a very restrictive basis. This was driven by claims by the rightholder sector that it was important to avoid ‘false negatives’ (i.e. works with contactable rightholders accidentally being declared orphan). The problem was that it condemned the Directive to irrelevance, with few institutions bothering to go through the process.
💊 Pill 6 | Imposing High Costs for Making Mistakes: An effective way to dissuade people from using the rights that they have is by giving the impression that there could be significant penalties around any potential errors. From the political pressure put on the national library of Sweden for allowing legitimate use of collections for AI training, to broader efforts to imply that use of copyright exceptions facilitates piracy, there is plenty to push already cautious researchers and librarians to be more cautious still and not use the rights they have in law.
Key routes around this include clear provisions on limitations on liability for mistakes made reasonably and in good faith. There may also be grounds for taking action against actors who repeatedly seek to undermine the impact of the law by creating needless concern among researchers and libraries.
💊 Pill 7 | Leaving Key Decisions up to Actors With Little Interest in Access: a final problem arises when the ultimate decision about whether the law applies is left to players with no real incentive to facilitate access. This has been the case with the out-of-commerce works provisions in the 2019 Directive on Copyright in the Digital Single Market, where a collecting society deemed ‘representative’ has full freedom to decide whether or not to offer licences for relevant out-of-commerce works. As we see in Sweden, while universities want to put their collections online, the collecting society refuses to offer a licence.
It is too easy for them to decide that the effort required is too great with little financial return, or that they may face complaints from creators, without bearing in mind the benefits that can come from greater access through libraries. A helpful step is therefore to make clear what can happen when there is a refusal to license.
There are many ways in which the positive potential of a law can be undermined through the use of poison pills such as those set out above. It is our job to watch for these and call them out, reminding legislators that if their laws are to be taken seriously, they must address these critical points of failure. We may well need to do so as copyright reform comes to the table in the coming months.
30 September 2026
