A year ago, it seemed as if Italy's stodgy National Research Council (CNR) was taking its reform programme seriously. It had drawn up plans to merge its 330 institutes and centres to 100 or so larger units of 'critical mass'. And, with great fanfare, it announced in December its intentions of recruiting directors to the new institutes from the international community — a radical change from its traditional tendency to recruit from its own ranks. Moreover, advertisements referred to the need for 'continual, original and sound' research experience. The CNR claimed that the new research directors were key to ensuring that the reforms delivered in terms of high-quality science.

Disappointingly, the first round of appointments, for the largest of the new institutes, suggests business as usual. Of the first 22 appointments selected last month, all are Italian scientists who have served as CNR directors. And there are examples of clear injustices, where scientists with significant scientific and research-management experience have been passed over in favour of those of lesser merit. In one case, a leading scientist was passed over in favour of another with one-tenth of his publication output in the past decade. Moreover, the average citation rate of the rejected candidate was three times that of the winner. There is no mention of what factors led to the choice.

What went wrong with the plan for the CNR to break out of its mould? One key issue is that it was not given a budget for its reform — essential for making the positions attractive to top researchers from other countries. Not surprisingly, few highly qualified foreigners applied. But why there were no appointments of Italian scientists from outside the CNR deserves explanation, as do those aberrant choices.

Another problem is the way in which the selection process took place. The procedures were out of line with international norms. CNR president Lucio Bianco is responsible for the final decision, but he is advised by (and heads) his Consiglio Direttivo, a committee of eight academics from all disciplines, from medicine to law. These committee members vote on candidates for each new institute directorship.

The Consiglio Direttivo did turn to experts for help. It appointed special commissions for each directorship, comprising three experts in the relevant field. But very few outsiders were brought in to ensure truly independent advice. The first 21 commissions whose names are publicly available included only four foreigners.

Moreover, the consiglio asked the commissions to provide unranked short-lists for its less technically expert members to choose from. And non-unanimous cases were combined to be voted on as a group rather than case by case. This system leaves room for suspicion of vote exchange — 'vote for my candidate and I'll owe you a favour' — the bad habit that the reforms should have forced the CNR to lose.

If the CNR appointments continue along the lines of these first ones, then it needs to come up publicly with a plausible explanation for why it cannot attract, or appoint, to key positions scientists from beyond its own ranks. Under attack from the new centre-right government of Silvio Berlusconi for inefficiency, the CNR needs more than ever to prove that it stands for research excellence and can strive to achieve it by seeing beyond its own highly politicized horizons.