This is one of the concrete consequences of the Nuclear Policy Council which was held on 3 February. The Nuclear Safety Authority (ASN) and the Institute for Radiation Protection and Nuclear Safety (IRSN) could soon merge, leading to the disappearance of the latter. This measure will, according to an official statement, “strengthen independence”, “consolidate skills” and “increase synergies” within the remaining organization. But according to Thierry Charles, former Deputy Director General of IRSN, this decision could, on the contrary, induce risks from the point of view of safety.
L’Express: What was your first reaction when you learned of this decision to unite the two organisations?
Thierry Charles: I found the press release very brutal. IRSN is a known and recognized entity. It works within the framework of a control system reputed throughout the world. And there, suddenly, we learn that it is suppressed and integrated into the ASN. Traditionally, this kind of decision means for the organization concerned that it has failed. But this is clearly not the case with IRSN. Obviously, the decision was not discussed at length. She came out suddenly. It must be difficult for the teams to take.
I am not defending the existence of the IRSN because I worked there. Any system must evolve and the one we have in France is cumbersome because it is based on several poles. However, the risk is to destabilize a sensitive organism. Let me explain: if you put an airplane pilot in permanent stress, his piloting will surely suffer. We have a similar problem here. Incidentally, this question of rapprochement with the ASN is relatively old. Even before the year 2000, the Le Déaut report mentioned it. The Court of Auditors also gave its opinion in a report submitted in 2014. However, these two documents converge in their conclusions: they recommend not to proceed with the merger.
What are your main concerns?
The first risk relates to IRSN’s expertise. How to maintain it at the best level? The planned development is to separate research from IRSN and therefore basically cut the direct link between the two. This poses a real problem. IRSN conducts research to consolidate its own expertise. So if you say that part is no longer done internally, everything becomes more complicated.
There is a real risk of impoverishment, in terms of the quality of expertise, of skills development, but also of the ability, for the expert to formalize a research need that can be quickly appropriated by the researcher, and for the researcher to produce results quickly appropriated by the expert. Another concern: the current system makes several documents (IRSN opinion, ASN decision, operator safety file) available to the general public. What will it be tomorrow? We will undoubtedly have less transparency.
This merger of organizations already exists in the United States. But the American model, no doubt more pragmatic than ours, seems to work fine. Isn’t this the way to go?
The clear separation between expertise and decision-making is precisely one of the strengths of our current system. Until now, IRSN has given an opinion on a dossier based on technical and scientific knowledge. A decision is then taken on the basis of IRSN’s opinion but also taking into account all the imperatives that may exist elsewhere. Until now, as the IRSN was independent, it could take the time necessary to carry out its work, if it considered for example that there was a need to dig into a subject.
Tomorrow, in a different context, I don’t know what will happen and that may pose a risk. I give you an example. In the 2000s, a leak was observed on a steam generator tube of an American reactor. The operator has asked to continue operation for some time. A first authorization was granted, then a second. But during this second extension, a tube rupture occurred. This did not lead to serious consequences but it was still a notable incident. The investigation later showed that the organization responsible for carrying out the expertise was, at the same time, in discussions with its supervisory authority, which threatened to reduce its budget if it did not go a little faster.
Without there being a written document, the system of expertise had received “a kind of moral pressure”. The expert who had worked on the subject had asked a series of questions, but he had not received all the answers. He still continued his work based on the information he had. So you have to be careful. Once again, I am not here to defend at all costs the fact that IRSN must remain as it is. However, I don’t want us to mess up a working system. We know what we lose, but not necessarily what we gain.