Several French surveys or barometers1 assess trust in science in France. Which figures should we focus on?
As a pollster, I’m very wary of surveys! Their results are actually highly dependent on how the questions are phrased: they are therefore more informative when repeated over long periods. In France, we are fortunate to have a source of this kind: a series of nine surveys on social perceptions of science, which began in 19722. Although the questions have often been reworded from one wave to the next, it is of genuine scientific quality, thanks in particular to the work of Daniel Boy at CEVIPOF.
What trends can be identified?
The proportion of French people stating that they do not trust science is low (around 10%) and has remained stable over time. But this consistency masks a shift in the identity and motivations of these ‘sceptics. Fifty years ago, science’s main rival was religion, against a backdrop of criticism levelled at science itself, accused of wrongfully claiming a monopoly on the truth. This has not been the case for the past thirty years or so – believers no longer really differ from non-believers. Moreover, scepticism has shifted towards scientists, who are suspected of having hidden agendas or of serving private interests.
It has long been said that the most highly educated people are the most confident in science. Is this still true?
It is a more nuanced picture. In the 1970s, having completed higher education – which applied to less than 10% of the population – was virtually a guarantee of trust in science. Today, with over 50% of a cohort going on to higher education3, anti-science attitudes can be found across all educational levels. Pursuing a long course of study does indeed enable a better understanding of the scientific method, but also allows for the development of a more sophisticated critique of it… Trust remains, however, more firmly established among the most highly qualified. In 2025, it applied to 96% of those with a degree at bachelor’s level or higher and ‘only’ 80% of those with a qualification below A‑level.
While the level of trust in science has remained stable since 1972, what about trust in its benefits?
In the first wave, in 1972, more than one in two French people believed that science brought humanity “more harm than good”, nearly 40% that it brought “more harm than good”, and around 5% that it brought “as much good as harm”. For 40 years, this breakdown remained more or less the same, although the proportion of critics increased slightly at the expense of the optimists.
Since 2011, a new balance seemed to be emerging. The proportion of those who believe that science does as much good as harm was increasing, at the expense of the two other, more polarised attitudes; we were witnessing the normalisation of a certain indifference or mild suspicion towards the benefits of science. But the optimistic attitude seems to have enjoyed a resurgence since the Covid pandemic, as confirmed by the latest wave, where it has returned to its 1972 level. The proportion of critics, on the other hand, has continued to decline and has fallen to around 5%.
In France, the level of trust in science4, although high, is below the global average, and levels of trust in the government, political parties or other citizens are below the OECD average5. How can this gap be explained?
There is traditionally a high level of conflict in France, and a great deal of mistrust towards anything coming ‘from above’. For some within the MEDEF and the political class, this is explained by the fact that the French are ‘Gauls resistant to change’. For my part, I believe that responsibility is in fact shared: the population may be distrustful, but the elites always end up growing impatient with its supposed conservatism and adopting paternalistic attitudes. Unfortunately, we know all too well how to destroy trust, but far less about how to build it. The processes involved are sometimes centuries old… Research carried out in Sweden, notably that of Bo Rothstein and Jan Teorell, has shown that the origins of strong social trust date back to the 19th Century, with the establishment of state education and the modernisation of the civil service.
Has science been overtaken by mistrust of the elites?
This is a theory that is beginning to be explored, from the perspective: is there such a thing as scientific populism, and what is its connection to political populism? Populism consists of a set of ideas pitting the people, presumed to be virtuous, against an elite—whether political or academic—presumed to be malicious. A recent European study6 examined the reciprocal influence of these two forms of populism. It found that whilst distinct, they are linked: mistrust of science cannot be explained solely by political populism, but the anti-elitism of our times does not spare scientists.
Politicians themselves are no strangers to this contagion. It is very visible in the United States, but this attitude also exists in Europe: challenged by their voters, they are only too happy to find other elites to blame… All the more so as scientists are easy targets, unaccustomed to defending themselves. Politics actually has an ambiguous relationship with science: it claims to defend it, yet at the same time mobilises it in a selective, even opportunistic manner. This stems from a lack of political ethics and comes at a cost to trust in science in the long term.
How, in this context, can we strengthen the public image of scientists?
Firstly, there is a need for educational efforts regarding the scientific method. Our fellow citizens often have a distorted view of science, imagining it to be deterministic and infallible. This fuels mistrust, because from this perspective, it is difficult to accept that one might revisit a hypothesis or a result without this being seen as a sign of failure… It would also be helpful if politicians supported scientists rather than attacking them.
How can we ensure that politicians support scientists?
We must try to depoliticise science. In late 2006, Nicolas Hulot’s ecological pact, signed by ten of the twelve candidates in the 2007 presidential election, listed ten objectives and five concrete proposals relating to sustainable development. The effect did not last, but during the election campaign, it made environmental issues visible and cross-party. Why not imagine that science journalists and communicators draft a ‘scientific pact’ and send it to the candidates for the next presidential election?
