In May 2026 the CNC launched a fund of 700,000 euros a year for innovation in film and image education, and it compares the fund itself to a research and development unit. Yet to answer a call for projects, you have to describe in advance what research only finds out by trying, and it is the organisations that pay for the trials. I put forward a few ideas so that public money can also pay for that time. I was a member of this commission.
The CNC’s new fund and “research and development”
On 11 May 2026 the CNC announced the creation of a grant for innovative schemes in film and image education, with 700,000 euros a year. It implements measure 14 of the plan to rebuild film and image education that the French ministries of Culture and of Education presented in November 2025, which provides for encouraging innovative projects in order to “support the rapid changes in young people’s cultural practices”. Applications were open from 1 to 30 June, and the CNC plans a call for projects every year, subject to its budget.
I agree with much of what the CNC wrote to present this fund. At the top of the educational objectives in the specifications, the CNC put taking young people’s practices into account, “so as to include them in the construction of the projects of which they are the target, in a participatory and collective approach”. In the FAQ it states that formats of works will not be considered innovative in themselves, and that innovation will come rather from a project’s ability to start from young people’s practices. I have often enough regretted that innovation gets confused with buying new equipment to appreciate seeing this said so clearly. In the same document, the CNC describes its fund with a comparison that I would like to take at its word: “It could be likened to a ’Research and development’ unit in the field of film and image education.”
Writing the project before trying it
According to the specifications, the application includes a description of the methodology with “precise indicators for evaluating the impact of the project”, where relevant a report on experiments already carried out, a timetable running up to the deployment of the project “at a higher level”, and two provisional budgets, one for the initial scale and one for deployment. The lead organisation must bring in at least one other organisation and secure the commitment of five partner venues that receive the public. Finally, the CNC states that particular attention will be paid to projects part of whose funding is already secured.
I understand each of these requirements, since public money is involved. Yet a project that meets them well is a project whose course is already known, which has often been tried, which has found other funders and which is ready to grow, in other words a project that has left the research phase. The CNC also accepts new projects that need support in order to be tested, but it asks them for the same indicators, the same deployment timetable and the same deployment budget. To fill in these sections, you therefore have to have already carried out trials, or to write as if they had already produced their results. In the first case, the trials were very often paid for by the organisation itself, out of its own funds or with its team’s unpaid time. Writing such an application in a few weeks, with the commitments of five partners, is already a piece of work that only some organisations can take on.
I find a way to think about this difficulty in a text that the sociologists Madeleine Akrich, Michel Callon and Bruno Latour published in two parts in 1988, “À quoi tient le succès des innovations ?” (Gérer et comprendre, nos. 11 and 12), translated into English in 2002 as “The Key to Success in Innovation”. They contrast two ways of explaining why an innovation succeeds. In the diffusion model, an innovation spreads thanks to its own qualities, which could therefore be assessed before it is launched. In the model they defend, that of interessement, an innovation succeeds because it is transformed through contact with the people who take it up, and each adoption is at the same time an adaptation, through successive prototypes and compromises between technology and people. A film and image education scheme will work, or not, depending on the adaptations made with young people, teachers and mediators, and these adaptations cannot appear in an application since they have not yet taken place.
With the first objective of the specifications, the difficulty becomes clearer. If we take seriously the idea of including young people in the construction of the project, which for me is how cultural rights are put into practice, then at the time of applying we do not know what we will build with them. An honest application ought to say so, and I do not see how anyone could, with selection criteria, prefer that sentence to an application that describes what will happen session by session. I tried to show in The sidestep as a craft that what is new arrives without the criteria that would allow it to be judged, since our criteria were formed in contact with what we already knew. This holds for any selection body, however attentive the people on it may be.
Who does the research today
The sociologist Norbert Alter showed in L’innovation ordinaire (PUF, 2000) that innovation differs from invention by its process. It is carried out by ordinary actors, in their daily work, often at odds with established rules, and the organisation only recognises it and gives it an institutional form afterwards. It seems to me that film and image education has often renewed itself in this way. Mediators, teachers, youth workers and artists invent ways of working in the field, without having been funded to do so, and some of these inventions become, years later, recognised schemes.
I experienced this with mobile phones. In 2005, when we founded the Pocket Films festival with the Forum des images, as cameras were appearing in phones, there was no criterion for judging these films, and we learned what they could be by showing them, from one edition to the next, until 2010. Later, from 2018, I led in the Île-de-France region the action research project « Image Education 2.1 », co-produced in particular by Arcadi and the Forum des images, whose stated aim was to produce methodological resources from field experiments on teenagers and cinemas. Arcadi was closed by the Region that same year, and I recounted in Why Are Cultural Actors So Fragile ? how quickly the traces of its work were erased.
When funding goes through calls for projects, however well they are written, it favours the organisations that have the means to write applications. In my reading of The White Paper on Cultural Decentralisation, published in March 2026 by the Observatoire des politiques culturelles from the contributions of some forty national networks, I had noted that almost all the organisations were asking for a gradual move away from calls for projects, because they impose a disproportionate administrative cost on organisations and favour the areas best equipped with project engineering. People who invent a great deal will never feel entitled to apply, or have no one to draw up a deployment budget.
I saw another way of doing things this year, at the Erasme innovation lab. Greater Lyon, which has since become the Lyon Metropolis, created it in 2002, and employs public servants there whose job is to prototype, to try things out and sometimes to fail. In a digital culture class in 2025-2026, three classes of lower secondary pupils made six short films in virtual reality there, something no pupil of that age in the metropolis had done before them, and I described it in Making virtual reality a ground for student autonomy. In October 2025, at the start of the project, we did not know what would come of it. The Metropolis was not betting on that particular project; it funds a place and a programme, the digital culture classes, which have existed for some twenty years and where this kind of project can be attempted.
A few ideas for funding the trial
The CNC wrote in the FAQ that the specifications could be adjusted for the following years, in the light of this first edition. So I put forward a few ideas, as hypotheses for discussion.
The first would be to fund short research phases, with small sums, for many organisations. It would be accepted in advance that some of them come to nothing, since that is the condition for something new to appear. The Youth Experimentation Fund (Fonds d’expérimentation pour la jeunesse), created by the French State in 2008, supported 959 experiments in ten years, also through calls for projects, but each with an evaluation protocol; it seems to me that in this logic, learning that an action does not produce the expected effect counts as a result.
The first deliverable should also be a public account, one that also tells what did not work. I worked with the team of an adolescent centre that had experienced, before I arrived, the failure of a workshop on young people’s video practices. The project had been decided from the top, entrusted to a practitioner whose expertise was individual consultation, and the local branches were supposed to send groups of young people to the main site. Almost nothing had happened. I drew a lesson of method from it, which I described in Making images in an adolescent centre: a mediation workshop is not stuck onto an institution, it is built with it.
For evaluation, I believe more in exchanges between peers than in self-evaluation in front of the funder, where one is tempted to write that everything went well. From September 2025 to March 2026, I supported for ALCA, the regional agency for books, film and audiovisual media, a collaborative assessment of film and image education across the region, which I described in Collaborative assessment of screen literacy education in Nouvelle-Aquitaine. Some twenty professionals took part, in four two-hour video sessions, and many of them did not know each other, or only by name. Each session began with a quarter of an hour of discussion, which I noted as a mind map projected on the screen, then everyone wrote alone for about twenty minutes on a shared platform, so that the contribution of a small rural association carried as much weight there as that of a large festival. Teams supported by the fund in the same year could work in this way, or visit each other in pairs to analyse together what happened at each of them.
The same work showed me that, in a region of twelve departments, people doing similar work do not know each other. Part of the money could pay for this kind of support in the regions, to find those who invent without thinking of themselves as innovators, and to help them put into words what they do, even before they submit an application.
Support for deploying projects that have proved themselves remains necessary, and the fund already makes it possible. What should be added is the means to pay teams for a few months of research, with a modest sum and a commitment to publish an account of it.
A first year
The CNC opened this fund on an experimental basis, and I think it right that a national institution should take on the mission of supporting innovation in film and media literacy while defining it from young people’s practices. At Erasme, the Lyon Metropolis has funded a team and a place since 2002. In film and image education, research work is done today by the organisations themselves, on their own time and with their own funds, and it seems to me that it often leaves no written trace. For my part, I continue to publish on this site the accounts of my projects, including failures such as the one at the adolescent centre, and I am in discussion with the CNC teams who are preparing the next edition.
