In a speech on 22 January, as he set out his plans for a national strategy on science and innovation, French president Nicolas Sarkozy lambasted the country's university system as “infantilizing” and “paralysing for creativity and innovation”. Sarkozy implied that French researchers were fainéants (layabouts) with cushy jobs, and no match for their supposedly more industrious British counterparts.

The speech was a typically melodramatic example of la méthode Sarkozy and, if it contained some home truths, it was largely a caricature. His harsh rhetoric in this case (see http://tinyurl.com/av7flg) can only reinforce the resistance he has set out to overcome. In 2000, the incumbent science minister, Claude Allègre, saw his plans for sweeping reforms dashed after scientists united against him, weary of his unnecessary provocations and sceptical of reforms imposed from on high with little consultation. Sarkozy is tempting a similar fate.

To their credit, Sarkozy and his science minister, Valérie Pécresse, have pushed through much-needed modernizations. These include putting universities on the road to independence from the centralized administration, giving them badly needed cash, and injecting a healthy dose of grants awarded on the basis of competitive proposals (see Nature 453, 133; 2008).

But a massive strike across French universities that began this week (see page 640) suggests that, applied to the research community, la méthode Sarkozy has reached its limits. Sarkozy should heed Allègre's earlier mistakes and understand that he cannot modernize France's research system unless he has scientists on board. As things stand now, even top researchers who support the broad thrust of the reforms complain that their advice is being ignored, and that many changes seem as though they are being imposed by technocrats seeking grandiose institutional rearrangements as ends in themselves.

The substance of Sarkozy's reforms is right, but to succeed he must engage more with scientists. Many researchers experience the reforms as if they were in an aircraft flying through thick cloud, buffeted by the turbulence of almost weekly changes, with little idea of where the plane is taking them. Some fears are exaggerated, but others are legitimate. To arrive at their destination, Pécresse and Sarkozy need to consult on reforms with the navigators in the research community who know this airspace best. And Sarkozy, a speedy man, may have to accept that throttling back can sometimes avoid unwelcome accidents.