2022年经济学人 结束科学停滞期（2）

（单词翻译）

This suggests that breaking bad funding habits should make a difference.

这表明，打破不良的资助惯例应该会有所帮助。

The DARPA model, which has more in common with venture capital than traditional funding structures, is an attempt to do just that.

相比传统资助体系，与风险投资有更多共同点的DARPA模式就是一种尝试。

It empowers programme directors to finance high-risk, high-reward projects with a bent¹ towards real-world use.

在这种模式下，授权项目主管有权资助高风险、高回报且更为实用的项目。

But though it has proved successful in the defence industry—funding groundbreaking technologies from the early internet to GPS—it may not be quite as successful elsewhere.

尽管在国防工业，从早期的互联网到全球定位系统(GPS)，这种资助突破性技术的模式很成功，但在其他领域可能效果欠佳。

A book chapter published in February by Ms Li and Pierre Azoulay at MIT notes that the DARPA model does best when its programme directors have a clear understanding of the sort of breakthroughs that are needed.

李和MIT的皮埃尔·阿祖雷在2月份出版的一本书中指出，当DARPA的项目主管熟知需要哪种突破时，DARPA的模式表现最好。

This is often the case at DARPA itself, where both the funder and user of new tech is the defence department.

DARPA本身往往就是这种情况，新技术的资助者和使用者都是国防部。

In areas like energy or health care, things are rather less straightforward³.

在能源或医疗保健等领域，事情就没那么简单了。

The end users are many and dispersed⁴ rather than a single government department.

这些领域的终端用户众多且分散，并非单一的政府部门。

Indeed, other work by Mr Azoulay and colleagues notes that although ARPA-E, an energy-focused outfit⁵ launched in 2009, is still in its relatively⁶ early days, it is yet to produce advances on a par² with its defence-orientated predecessor⁷.

事实上，阿祖雷和其同事们的其他研究指出，尽管2009年成立的专注于能源的美国能源部先进能源研究计划署（ARPA-E）仍处于相对早期的阶段，但它尚未取得与其重视国防的前身同等水平的进展。

Another approach in vogue⁸ is to fund “people not projects”.

另一种流行的方法是资助“人而不是项目”。

Most conventional grants fund specific projects for a specific amount of time, usually a few years, which researchers worry prevents them from pivoting⁹ to new ideas when old ones do not work out and fails to allot¹⁰ enough time for risky¹¹ ones to come to fruition.

大多数传统的赠款都是在特定的时间内资助特定的项目，时间通常是几年，研究人员因此担心，当旧的想法不起作用，也没有分配足够的时间来实现高风险想法时，他们会来不及转向新想法。

A study in 2011 compared researchers at the Howard Hughes Medical Institute, where they are granted considerable flexibility¹² over their research agendas and lots of time to carry out investigations¹³, with similarly accomplished¹⁴ ones funded by a standard NIH programme.

2011年的一项研究将两组水平相当的研究人员进行了对比，第一组由霍华德·休斯医学院在研究议程上给予了很大灵活性，并提供了大量时间开展调查，而第二组由NIH标准项目资助。

The study found that researchers at the institute took more risks.

研究发现，第一组承担了更多风险。

As a result, they produced nearly twice as much highly cited work, as well as a third more “flops” (articles with fewer citations¹⁵ than their previously¹⁶ least-cited work).

最终，他们写出了近两倍的高被引论文，而只多出了三分之一的“惨败文章”(被引用次数创新低的文章)。

These results may be hard to replicate¹⁷ elsewhere.

其它领域可能很难出现同样的结果。

Researchers at the Howard Hughes institute are selected for attributes that suggest they will thrive in such a flexible environment.

霍华德·休斯研究所的研究人员被选中，是因为他们具备在灵活环境中从容应付困难的特质。

But the gap is big enough to indicate that others may also benefit from more freedom.

但两组对比的差距之大足以表明，其他人也可能从更多的自由中受益。

Despite the uncertainty¹⁸ about exactly how best to fund scientific research, economists²⁰ are confident of two things.

尽管资助科学研究的最佳方式尚未确定，但经济学家们对两件事深信不疑。

The first is that a one-size-fits-all approach is not the right answer, says Heidi Williams of Stanford University.

第一，一刀切的方法并非正解，斯坦福大学的海蒂·威廉姆斯如是说。

DARPA models, the Howard Hughes Medical Institute’s curiosity-driven method, and even handing out grants by lottery²¹, as the New Zealand Health Research Council has tried, all have their uses.

DARPA的模式，霍华德·休斯医学研究所的好奇心驱动法，乃至像新西兰健康研究理事会所尝试的那样，通过抽签发放赠款，都有它们的用途。

Evaluation²² of them can then build knowledge of what works, says Matt Clancy, an economist¹⁹ who curates a continuously updated online literature survey on innovation, itself an experiment in how to improve science.

经济学家马特·克兰西称，对它们进行评估可以让人了解其效用。克兰西策划了一项关于创新的持续更新的线上文献调查，这本身就是一项测试如何促进科学发展的实验。

The second is that this burst of experimentation²³ must continue.

第二，这一系列实验必须继续下去。

The boss of the NSF, Sethuraman Panchanathan, agrees.

NSF的老板塞图拉曼·潘查纳坦对此表示赞同。

He is looking to reassess projects whose reviews are highly variable—a possible indication of unorthodoxy.

他希望重新评估那些评价变化很大的项目--这可能是出现非正统想法的迹象。

He is also interested in a Willy Wonka-style funding mechanism²⁴ called the “Golden Ticket”, which would allow a single reviewer to champion a project even if his or her peers do not agree.

他还对威利·旺卡式的资助机制“黄金票据”感兴趣，这种机制允许单一评审者在其他人都否决的情况下也能支持一个项目。

Mr Clancy notes that many venture-capital partnerships²⁵ employ similar policies, because they prioritise the upside of long-shot projects rather than seeking to minimise failure.

克兰西指出，许多风险投资合作伙伴关系都采用了类似的政策，因为他们会优先考虑长期项目的好处，而不是想方设法将失败降至最低。

Thus far, there is little quantitative²⁶ evidence on whether Golden Tickets produce Golden Outcomes.

到目前为止，几乎没有量化证据能够表明黄金票据是否会产生黄金结果。

All the more reason to try them.

那就更应该去尝试了。