Traffic law must be completely overhauled for self-driving cars

Dutch traffic regulations are not equipped for self-driving cars. That is the conclusion of UG researcher Nynke Vellinga. In this episode of the JTS Scholars, she talks about her research into the legal framework of autonomous vehicles. ‘If we allow self-driving cars, we need to adapt the entire system.’

Text: Jelle Posthuma

Vellinga sketches an attractive vision of the future: imagine getting into a self-driving car on a Friday afternoon after work, falling asleep in the evening, and waking up the next morning on the French Côte d'Azur. That does not sound bad at all. But there is still a long way to go before this scenario becomes reality, especially in terms of traffic legislation in the Netherlands, Vellinga says. ‘Elsewhere in the world they are much further ahead.’

Vellinga specializes in technology law, with a focus on autonomous transport and medical innovations. In her research, she concentrates on the legal framework of technological innovations. Autonomous systems offer great advantages, she explains. People inevitably make mistakes: they fall asleep or get distracted. For self-driving cars, this is not the case. At the same time, these vehicles are not without risks either. A computer may crash, and a sensor may fail. Yet Vellinga emphasizes that there is the potential for a substantial safety gain with autonomous vehicles.

How is it arranged elsewhere?

According to the researcher, important international steps have already been taken to make self-driving cars possible. In the American state of California, taxi services with autonomous vehicles are now permitted. Closer to home, there is also increasing room for this technology. In Germany, cars with an automation level of SAE Level 4 have been admitted to the Autobahn, which is known for its long, straight stretches. The SAE scale runs from 0 (no automation) to 5 (full automation) and indicates to what extent a vehicle can drive independently.

At the European level too, there are possibilities. EU legislation concerns the safety requirements of vehicles. By now, self-driving cars may, in specific cases, be approved for European roads. In the Netherlands, the RDW (Netherlands Vehicle Authority) is responsible for granting approval. But, Vellinga states, Dutch traffic rules are lagging far behind. The Road Traffic Act, which dates from 1994, is not equipped for autonomous transport. Therefore, self-driving cars cannot simply take to the roads in the Netherlands.

Traffic law and liability

Vellinga explains: ‘The law assumes that a car is driven by a human being.’ A human driver must follow the rules and is responsible in the event of a violation. The very idea of self-driving cars is that there is no human driver. As a result, there is a lack of what is known as a ‘norm-addressee’. This is a (legal) person to whom a rule, in this case the Road Traffic Act, is addressed and who is responsible for complying with it.

Besides these criminal law provisions, there is civil liability. In other words, who pays for damages? Dutch liability rules are very broadly designed, Vellinga explains. It essentially revolves around blameworthy conduct. Should you have been more careful, or: did someone behave wrongly? In Dutch traffic, there is moreover a very strong protection of vulnerable road users, such as pedestrians or cyclists.

With self-driving vehicles, the question is who acts blameworthily: the owner of the car or the manufacturer? European rules now state that the manufacturer is liable if the software fails, but other situations may arise. Think of an owner who neglects to install a software update or forgets to check the sensors around the car. In that case, it is the owner of the self-driving car who has acted blameworthily.

Political considerations and maritime rules

There is still a long way to go before the Dutch Road Traffic Act can be adjusted, Vellinga argues. ‘If we allow self-driving cars, we need to adapt the entire system, because the main character of the law – the driver – disappears.’ The Dutch Ministry of Infrastructure and Water Management is, however, working hard on an exploration to adjust the legislation. By the end of 2025, a ‘roadmap’ is expected for the revision of the legislation. ‘An ultimate solution is difficult. In the end, it is also a political consideration of how we adapt the Road Traffic Act.’

As a researcher, Vellinga naturally focuses on the academic side. Still, she has a few suggestions for a revision of the legislation. ‘One possibility is a system based on the maritime world. There, the captain remains ultimately responsible, even if the helmsman is steering the ship. You could translate something like that to self-driving cars.’

The researcher also points to the duty of care of the owner of an autonomous vehicle. The owner must ensure software updates and regularly check the sensors. The manufacturer is ultimately responsible for the fail-safe: if something goes wrong, the system must be designed in such a way that it does not pose any danger. Think of level crossing barriers, which close during a power failure.

Different perspectives

In her research into the legal framework of technological innovations, Vellinga seeks to work in an interdisciplinary manner. According to her, such interdisciplinary collaboration is ‘of great importance in advancing research on all fronts’. ‘Self-driving cars and medical innovations come with numerous technical aspects, yet I am not an engineer. That is why I work with colleagues who can explain these technical aspects in layman’s terms. At the same time, I can explain to them whether something is legally feasible.’

As a JTS Scholar, Vellinga wants to create as many moments as possible where different disciplines come together. ‘You can come up with all sorts of rules, but they also need to fit within society. That is why it is important, for example, to take ethical and psychological issues into account.’ The first event is already planned: a conference on autonomous transport and legal issues. ‘Here too, I hope to welcome researchers from different disciplines. It helps enormously to exchange different perspectives.’