South Korea has become the latest country to offer scientists cash prizes for publications in top-level international journals (see page 792). Other nations, including China and Pakistan, already have such programmes in place. The thousands, or even tens of thousands, of dollars on offer can be a fat prize for researchers in countries with lean science budgets.

Many researchers will turn up their noses at this practice. Scientists, after all, are supposed to be motivated by curiosity, by a devotion to finding the truth, by a desire to solve various philosophical or social problems — not by money. And funds should find their way to self-motivated scientists with a project deemed important. This assessment should be made by taking account of the project's feasibility, originality and scientific significance.

But this is no easy task in any country. In countries with inexperienced or understaffed scientific evaluation committees, it is almost impossible. Without the proper mechanism in place, funding can be diluted by equal distribution to all, or hijacked by projects approved on the basis of personal connections rather than sound science. Where there is little faith in review committees, giving money directly to those who have proven themselves might seem a beneficial, albeit imperfect, way of encouraging scientists.

Proponents can point to other potential advantages too. Scaling up bonuses for high-impact papers, as these programmes often do, might stem the urge to churn out quick papers in order to beef up a publication list. (Graduate students in China often need to have several publications for higher degrees.) And they also encourage scientists in countries with traditions of local-language publishing to think more internationally.

A researcher measuring science in terms of dollars might be more tempted to plagiarize or fabricate data.

But there are some powerful arguments against the widespread adoption of the practice. Cash bonuses tied to specific publications are likely to exacerbate corrupting tendencies in the scientific community. Debates over who should be included on author lists, and who should be the first author and the corresponding author, will surely get even more vicious when a chunk of money is on the line. A scientist struggling to meet a mortgage payment might be more willing to forgo a potentially fantastic result for a quick cash-earner. And a researcher measuring science in terms of dollars might even be more tempted to plagiarize or fabricate data.

In countries recently damaged by high-profile cases of scientific corruption, where it is all the more essential to develop a culture of integrity, the award of large sums of money for high-impact publications is even less desirable. The scientific world already places too much importance on high-impact journals in assessing individuals, and on the crude ‘impact factor’ in particular (see Nature 435, 1003–1004; 200510.1038/4351003a). Papers published in journals with high impact factors do indeed tend to be significant papers, but a literal formula highly geared in their favour cannot do justice to the way science works.

Money does matter, of course. Scientifically developing countries need to compete for excellent scientists in an increasingly global marketplace. And scientists everywhere use their reputations, based on their latest and greatest papers, to negotiate raises, promotions or new jobs. So countries struggling to find ways of motivating their scientists need to reward outstanding researchers for good publications, perhaps by paying a bonus based on a peer review of achievements over at least a year.

But nations and agencies should avoid resorting to crude cash-per-paper incentives. They should instead attempt to be less formulaic and more considered in the ways they reward their scientists. And they should redouble their efforts to ensure that, in the midst of concerns about rewards, young scientists are committed above all to the ethical pursuit of scientific truths.