VoxModify

2026-01-22 02:55:00

2026年1月21日，美联储主席丽莎·库克（Lisa Cook）和律师阿贝·洛威尔（Abbe Lowell）离开最高法院。| 凯文·迪茨赫/盖蒂图片社

通常，最高法院的共和党多数派认为，总统唐纳德·特朗普可以随意解雇联邦机构的工作人员。例如，去年7月，他们允许特朗普政府解雇教育部近一半的雇员。然而，今年5月，法院也暗示美联储是一个特殊机构。在特朗普诉威尔克斯案（Trump v. Wilcox, 2025）中，法院指出特朗普不能解雇美联储领导人，因为该机构是一个“独特结构的准私营实体，遵循第一和第二美国银行的历史传统”。这句话的意思并不明确，但在周三早上的特朗普诉库克案（Trump v. Cook）口头辩论中，大多数大法官都表示将遵循威尔克斯案中确立的观点。六位大法官——三位民主党人、首席大法官罗伯茨（Roberts）、大法官卡瓦诺（Kavanaugh）和巴雷特（Barrett）——似乎很可能拒绝特朗普试图掌控美联储的企图。

与此同时，即使是通常被视为共和党死忠派的大法官萨姆·阿利托（Samuel Alito）也对特朗普律师提出了质疑。美联储负责制定利率等技术性决策，如果利率过高，企业借贷成本将上升，投资和雇佣将停滞；如果利率过低，短期内经济可能快速增长，但长期将面临严重的通货膨胀。换句话说，美联储有权力像给经济注射可卡因一样，带来短期的繁荣，但代价是更大的经济痛苦。因此，国会通过立法保护美联储主席不受总统控制，仅允许总统在“有正当理由”时解雇他们，以防止总统在选举年通过施压美联储来降低利率，从而获得短暂的经济提振。

库克案似乎涉及特朗普试图绕过这一法律，通过编造虚假理由来解雇一位美联储主席。如果特朗普在库克案中胜诉，他的政府已经表明将对美联储主席杰罗姆·鲍威尔（Jerome Powell）提出类似不实指控。

解释特朗普试图削弱美联储独立性的行为

去年8月，特朗普试图解雇由拜登任命的美联储董事会成员丽莎·库克，理由是她声称在一份抵押贷款申请中，“密歇根州和佐治亚州的房产同时作为她的主要住所”。然而，特朗普尚未提供任何实质证据支持这一指控，也没有给予库克解释的机会。据路透社9月份的报道，这些指控似乎完全是捏造的。库克确实签署了一份文件，表明她将使用亚特兰大的房产作为主要住所，但该文件也指出，银行可以书面同意该房产用于其他用途。此外，库克在另一份文件中告诉贷款人，该房产将作为“度假屋”使用。

库克的律师保罗·克里门特（Paul Clement）在周三的辩论中告诉大法官们，这些文件中的任何不一致之处最多只是无意的错误。从一开始，几位关键的大法官就对特朗普以抵押文件中的小错误为由解雇库克的决定表示怀疑。正如任何获得抵押贷款的人都会知道，申请过程中需要签署大量文件，许多文件由政府起草，几乎没有时间审查或修改，即使可以修改。因此，首席大法官约翰·罗伯茨（John Roberts）在辩论初期就告诉司法部长约翰·索厄尔（John Sauer），“我们可以在一堆房地产文件中讨论，其中一份文件包含不准确的信息，是否足以成为解雇库克的理由。”罗伯茨还质疑索厄尔所说的库克“欺诈”一词是否适用于“一个被其他文件所反驳的无意错误”。

大法官卡瓦诺（Kavanaugh）则警告索厄尔，如果总统可以因为发现一个最多只是小的文书错误就解雇美联储主席，那么这将摧毁美联储的独立性。在这种情况下，总统将进行一场“搜索与摧毁”行动，寻找借口解雇所有不认同的美联储主席。

大法官艾米·科尼·巴雷特（Amy Coney Barrett）则提出了一个更基本的问题：“你为什么害怕听证？”如果特朗普如此确信库克犯了错误，为什么不在试图解雇她之前通知她并给予她辩护的机会？

然而，只有戈萨奇（Gorsuch）似乎给出了一个合理的理由支持特朗普的立场，即库克只能通过寻求一种很少使用的法院命令——“强制令”（writ of mandamus）来挑战特朗普的解雇决定。这种案件几乎不可能胜诉，而索厄尔则声称强制令从未被用于对抗总统。因此，戈萨奇的立场很可能会成为特朗普的全面胜利。

大法官们似乎对如何解决本案中的几个复杂问题感到不确定

尽管大多数大法官对特朗普的论点持怀疑态度，但他们都对本案中存在的一些复杂问题感到困扰。例如，如果库克有权获得听证，那么听证应该是什么样的？“因正当理由”解雇某人到底意味着什么？法院是否应参考其他联邦法律、立法历史或英美普通法，或者三者结合来寻找答案？

这些问题是如此复杂，以至于一些大法官，包括巴雷特和金斯伯格（Ketanji Brown Jackson），建议法院可能在下级法院有更多时间考虑此案后再做决定。虽然最高法院决定举行口头辩论，但此案实际上是在所谓的“影子法庭”（shadow docket）上提出的，这是一个包含紧急动议和其他通常不举行口头辩论甚至不解释决定的案件集合。

当下级法院裁定特朗普不能解雇库克后，特朗普向最高法院申请了临时紧急命令，试图解除她的职务。但根据2009年的尼克恩诉霍尔德案（Nken v. Holder），申请此类命令的一方必须证明他们不仅有可能胜诉，还必须证明如果不立即干预将遭受“不可挽回的损害”，并且立即下令不会损害“公共利益”。

尽管最高法院的共和党人似乎在过去曾豁免特朗普遵守尼克恩案的规则，但巴雷特，一位由特朗普任命的共和党大法官，却询问索厄尔，经济学家提交的简报中指出，如果特朗普解雇库克，将引发经济衰退。巴雷特暗示，经济衰退显然不符合公共利益。她还要求克里门特提供最佳论据，证明如果库克继续留在美联储董事会，特朗普不会遭受不可挽回的损害。

换句话说，巴雷特似乎在寻找一种方式，可以在不解决本案中更复杂问题的情况下驳回特朗普的紧急请求。然而，并非所有大法官都认同这种狭窄的裁决方式。罗伯茨尤其怀疑是否有必要再给库克一次听证，因为本案中并没有太多事实争议。库克在听证中只会声称自己最多犯了一个无意的错误。因此，罗伯茨似乎认为，法院应直接决定，这种错误是否足以成为解雇她的理由。

其他大法官也对如此快速地决定此案并缺乏足够证据感到沮丧。例如，阿利托抱怨此案以如此“仓促的方式”提交给最高法院，以至于库克的抵押贷款申请文件甚至尚未成为他所在法院的正式记录。

总之，本案的裁决结果尚不明确，甚至可能只是表示“我们会在未来某个时候决定此案”。但即使如此，对库克有利的下级法院裁决仍有效。如果最高法院裁定反对特朗普，也可能阻止特朗普继续针对鲍威尔采取行动。而且，从目前来看，大多数大法官似乎并不认为特朗普应该有权命令美联储主席降低利率，否则就予以解雇。

The Supreme Court’s Republican majority ordinarily believe that President Donald Trump is allowed to fire virtually anyone who works for a federal agency. Last July, for example, they permitted the Trump administration to fire nearly half of the Department of Education’s employees.

In May, however, the Court also signaled that the Federal Reserve is special. In Trump v. Wilcox (2025), the Court indicated that Trump may not fire the Fed’s leaders because that agency is a “uniquely structured, quasi-private entity that follows in the distinct historical tradition of the First and Second Banks of the United States.”

It is not at all clear what this cryptic sentence means, but at Wednesday morning’s oral argument in Trump v. Cook, most of the justices signaled that they will adhere to the view that they laid out in Wilcox. Six justices — the three Democrats plus Chief Justice Roberts, Justice Brett Kavanaugh, and Justice Amy Coney Barrett — appeared very likely to reject Trump’s attempt to seize control of the Federal Reserve. Meanwhile, even Justice Samuel Alito, who is ordinarily a kneejerk Republican partisan, asked some skeptical questions of Trump’s lawyer.

The Federal Reserve is supposed to make technocratic decisions about where to set interest rates. If they set those rates too high, it will be too expensive for businesses to borrow money and investment and hiring will stagnate. At the same time, if they set rates too low, the economy will take off in the short term, but will experience much more damaging inflation in the long term.

The Fed, in other words, has the power to inject cocaine into the economy — giving it a temporary high at the price of much greater economic pain down the road.

For this reason, Congress shields the Fed’s governors from presidential control, only permitting the president to fire them “for cause.” This is to prevent the president from pressuring them to lower interest rates in an election year, when the president’s party would benefit from a temporary economic high.

The Cook case, meanwhile, appears to involve Trump’s attempt to bypass this law by making up a fake reason to fire a Fed governor. And, if Trump prevails in Cook, his administration has already signaled that it will bring similarly dubious allegations against Fed chair Jerome Powell.

Trump’s attempt to neutralize the Fed’s independence, explained

Last August, Trump attempted to fire Lisa Cook, a Biden appointee to the Fed’s Board of Governors, claiming that she falsely claimed on a mortgage application that “both a property in Michigan and a property in Georgia would simultaneously serve as her principal residence.” But Trump has yet to provide any meaningful evidence that supports this allegation, and he never gave Cook a hearing where she could explain this alleged falsehood.

According to a Reuters report from last September, moreover, these allegations appear to be fabricated. While Cook does appear to have signed a document indicating that she would use the Atlanta property as a primary residence, that document states that the bank may agree in writing that the property may be used for something else. And, in a separate document, Cook told the lender that the property would be used as a “vacation home.”

Cook’s lawyer, Paul Clement, told the justices on Wednesday that, “at most,” any discrepancies in Cook’s mortgage documents are inadvertent.

Right out of the gate, several key justices appeared skeptical that a minor discrepancy on mortgage documents could justify Trump’s decision to fire Cook. As anyone who has ever obtained a mortgage can testify, the process requires the borrower to sign a huge pile of documents, many of which are drafted by the government, with little time to review them or to ask for them to be changed — even if such a change can be made. 

Thus Chief Justice John Roberts told solicitor general John Sauer early in the argument that “we can debate…how significant it is that, in a stack of papers you have to fill out when you’re buying real estate,” one of those papers contains an inaccurate representation. Roberts also expressed skepticism of Sauer’s claim that Cook engaged in “deceit,” asking whether it is really fair to apply this word to “an inadvertent mistake contradicted by other documents in the record.”

Justice Brett Kavanaugh, meanwhile, warned Sauer that, if the president can fire a Fed governor because they discover what is at most a minor paperwork error, then that will destroy the Federal Reserve’s independence. In such a world, presidents will engage in a “search and destroy” operation, combing through the records of every Fed governor they disagree with in order to find a pretext to justify firing them.

Meanwhile, Justice Amy Coney Barrett asked an even more basic question: “Why are you afraid of a hearing?” If Trump is so confident that Cook actually did something wrong, why wouldn’t he give her notice of the allegations against her and an opportunity to defend herself before he tries to fire her? 

Of all the justices, only Gorsuch seemed to offer a coherent reason to rule in favor of Trump — Gorsuch suggested that Cook may only challenge Trump’s decision to fire her by seeking a rarely used court order known as a “writ of mandamus.” It is nearly impossible to win a mandamus case, and Sauer argued that mandamus may never be used against the president. So Gorsuch’s approach would likely amount to a total victory for Trump.

The justices did appear uncertain about how to resolve several tough issues lingering in this case

Though most of the justices appeared skeptical of Trump’s arguments in Cook, many of them also were frustrated by several difficult issues lurking within this case. If Cook is entitled to a hearing, what should that hearing look like? What exactly does it mean to fire someone “for cause?” And should courts look to other federal statutes, to legislative history, to the English common law, or to some combination of the three to find the answers to these questions?

Indeed, these questions are so vexing that some of the justices, including Barrett and Justice Ketanji Brown Jackson, suggested that the Court might avoid deciding them until the lower courts have more time to consider this case.

Although the Court decided to hold an oral argument in Cook, the case technically arose on the Court’s “shadow docket,” a mix of emergency motions and other matters that the justices typically decide without oral argument or even without explaining their decision. After a lower court ruled that Trump may not fire Cook, Trump asked the Supreme Court to issue a temporary emergency order removing her from office.

But when a party seeks such an order, the Supreme Court said in Nken v. Holder (2009), they must do more than show that they are likely to prevail in the case. Among other things, they must also show that they will be “irreparably injured” if the Supreme Court does not intervene right away. They also must show that an immediate order would not harm the “public interest.”

Although the Court’s Republicans appear to have exempted Trump from having to comply with Nken in the past, Barrett, a Republican appointed by Trump, asked Sauer about an amicus brief filed by economists who argued that allowing Trump to fire Cook would trigger a recession. A recession, as Barrett suggested, would not be in the public interest. She also asked Clement to provide his best argument that Trump will not experience irreparable harm if Cook gets to remain on the Fed’s board while this case is fully litigated in lower courts.

Barrett, in other words, appeared to be looking for a way to reject Trump’s emergency request without having to decide any of the harder questions this case presents.

Not all of the justices, however, appeared to believe that such a narrow decision would be desirable. Roberts, in particular, expressed doubt that there would be any point to giving Cook an additional hearing, because there aren’t really many factual disagreements to be explored here. Cook will simply claim in such a hearing that she, at most, made an inadvertent mistake.

So Roberts seemed to believe that the Court should just decide, as a legal matter, whether such a mistake is sufficient grounds to fire her.

Other justices appeared exasperated that they were being asked to decide this case so quickly and with so little evidence in the record. Alito, for example, complained that this case came up to the Supreme Court in such a “hurried manner” that Cook’s mortgage applications aren’t even part of the official record before his Court.

All of which is a long way of saying that it is far from clear what the Cook opinion will say, or even if it will say anything more than “we’ll decide this case at some point in the future.”

But even such a narrow order would be good news for Cook in the short term, because a lower court order keeping her in office remains in effect. A Supreme Court ruling against Trump might also dissuade Trump from continuing to target Powell. And it does appear that, whenever they get around to deciding Cook, most of the justices do not think that Trump should be able to order Fed governors to lower interest rates or else they will be fired.

2026-01-21 21:30:00

美国食品药品监督管理局（FDA）于2026年1月发布了初稿指导文件，鼓励在临床试验中使用贝叶斯统计方法，这一转变可能对罕见病患者（如肌萎缩侧索硬化症（ALS）患者）带来益处。目前，我们测试新药的方式有些奇怪：每项临床试验都必须假装之前从未进行过类似研究。即使临床医生已经测试过类似药物多年，或者几十年的研究都指向某个方向，每项试验仍需独立证明药物有效，仅依据该研究内部的结果。以往的研究成果不被考虑。

这种“白纸一张”的方法在过去60多年里一直是FDA的标准做法，而且有其合理性。如果允许以往研究正式影响药物是否有效的判断，制药公司可能会挑选那些结果更有利的研究。因此，这样的规则引发了学术界关于过往研究是否应影响最终药物评估的争论。然而，对患者而言，每次从头开始的代价很高。对于罕见病患者来说，全球可能只有几百人患病，进行传统试验几乎不可能，因为患者数量不足。对于儿童来说，这意味着需要重新证明药物在成人中已知有效的结果。而对于所有人来说，这都意味着更慢、更昂贵的试验，并且浪费了有用的信息。

现在，FDA告诉制药公司和研究人员，他们不必每次都从头开始。上周，该机构发布了新的指导文件，鼓励企业使用一种通常按个案处理的统计方法——贝叶斯方法。这意味着，企业可以正式利用之前的研究、相关药物或真实世界证据，来帮助判断药物是否有效。尽管该指导文件仍处于草案阶段，细节可能在未来几个月内发生变化，但政策信号已经很明确。

“听起来用之前的数据来指导下一步研究是如此直观，”麻省总医院神经科医生Merit Cudkowicz表示，“而不是像现在这样存在一种‘健忘’。”对于药物获得FDA批准，需要经过三个阶段的临床试验。但“证明有效”意味着不同的事情，取决于如何处理不确定性。传统方法——称为频率学派统计——提出一个狭窄的问题：如果这种药物实际上无效，那么我们看到的这些结果有多大可能是偶然的？如果这个概率非常低（通常低于5%），药物就会通过测试。这种方法强调客观性，试验数据本身说话，而进入试验前的信念不会正式纳入计算。

而贝叶斯统计方法则翻转了问题。它询问：基于我们已有的所有知识，这种药物有多大可能有效？然后，随着新试验数据的出现，它会不断更新这一估计。结果不是简单的通过或失败，而是一个概率，比如药物有效的可能性为94%。这并不意味着一切都可以随意进行，FDA仍然需要在试验前设定一个明确的界限。

贝叶斯方法的实际效果是，允许研究人员正式“借用”其他来源的信息。如果已经测试过某种药物在成人中的效果，就可以在评估儿童使用时利用这些数据。如果在一项试验中测试多种药物，一个试验组的数据也可以用于其他组。这种灵活性在患者难以招募的情况下尤为重要。

“我们之所以在儿科研究中看到贝叶斯方法的广泛应用，是因为我们通常都有成人研究的数据，”FDA药物审查部门的统计学家James Travis表示。

然而，引入外部信息也带来了明显的担忧：研究人员是否会挑选那些能证明其药物有效的研究？传统试验有一个硬性门槛——“p值”，衡量结果是否可能由偶然因素导致。你要么达到统计显著性，要么不达到。而贝叶斯方法则要求研究人员选择“先验”，即基于现有证据对药物效果的预期。但这种批评假设传统试验是完全客观的，而实际上并非如此；它们只是更好地隐藏了假设。

每项临床试验都涉及选择：哪些患者参与、测量哪些结果、进行哪些比较。p值可能让人觉得是数学在决定结果，但实际上主观判断贯穿始终。贝叶斯方法的支持者认为，这种方法迫使研究人员公开这些假设，必须在试验开始前明确说明并加以证明，这样FDA审查人员和其他人都能清楚了解这些假设是否合理。

这一切听起来可能像是一场学术统计争论，但对严重疾病患者及其家属来说，这关系到生死。以ALS为例，这是一种神经退行性疾病，大多数患者在确诊后两到五年内死亡。根据美国疾病控制与预防中心（CDC）的国家ALS登记，每年约有5000名美国人被诊断为ALS。然而，尽管已有数十年的研究，药物试验仍然屡屡失败。每次单独测试一种药物，从头开始，对这种没有太多等待时间的疾病来说，进展非常缓慢。

2019年，FDA批准了一项不寻常的贝叶斯试验，以寻找新的ALS药物。在HEALEY ALS平台试验中，麻省总医院的研究人员能够同时测试多种ALS药物，这对于患者来说至关重要，因为他们没有太多时间等待。试验中患者的数据（包括接受安慰剂的患者）可以用于指导其他药物的测试。这意味着试验可以在不重新开始的情况下，淘汰无效药物并加入有潜力的药物。

在该试验运行的四年中，已有七种药物被测试，而传统方法可能只能测试两种。Cudkowicz表示，新的FDA统计指导文件应为其他试验提供类似的模式铺平道路。

“之所以能迅速招募到患者，是因为ALS患者觉得这是一个以患者为中心的试验，”Cudkowicz说。其中两种药物显示出足够的潜力，现已进入最终阶段试验。

“贝叶斯方法的目标是充分利用参与者提供的所有数据，”Berry Consultants的统计学家Melanie Quintana表示，她参与了HEALEY试验的设计。

然而，更多的灵活性也意味着更大的风险。2018年，哈佛大学的Aaron Kesselheim教授与他人共同审查了超过100项适应性试验（一种允许试验过程中调整的类似方法，通常使用贝叶斯方法）。他们发现，只有三分之一的试验使用了独立委员会来监控数据，仅6%的试验在分析中期数据时对统计学家进行了盲法处理。如果没有这些保障措施，偏见更容易渗透，早期结果也可能误导。

FDA官员表示，贝叶斯试验的保障措施仍将存在。每个提案都将由FDA统计学家审查，公司必须在试验开始前确定其方法。

“你不能在看到数据之后再选择先验，”FDA生物统计主管John Scott表示，“对此有非常严格的规定。”

但问题是，这些方法是否会被企业实际采用？目前，该指导文件尚未最终确定，公众意见征集将持续到3月13日，最终版本预计在大约18个月后发布。此外，FDA面临领导层更替和政治不确定性，企业可能更加谨慎地尝试新方法。

“制药公司讨厌不确定性，”前FDA官员Adam Kroetsch表示，他撰文探讨了FDA的发展。“他们可能会认为风险太大，而选择使用传统方法，因为有FDA的先例。”

不过，FDA并不是唯一在推动这一转变的机构。欧洲药品管理局（EMA）也在探索扩大贝叶斯方法在药物开发中的应用。对于罕见病患者，或等待已有成人适用药物的儿童来说，这种统计方法的改变可能意味着生死攸关。HEALEY试验已经展示了这种可能性，而FDA也已经打开了大门。现在，更多公司需要迈出这一步。

Here’s something strange about how we test new drugs: Every clinical trial has to pretend that nothing like it has ever come before.

Even if clinicians have tested similar drugs for years, or if decades of research point in a certain direction, each trial must prove — independently — that the drug works based solely on what happens inside that specific study. Prior knowledge doesn’t count. 

For more than 60 years, this blank slate approach has been the Food and Drug Administration’s gold standard — and for good reason. If you let prior research formally count towards proving a drug works, drug companies might easily cherry-pick the studies that flatter their results. 

Naturally, such rules have led to academic circle jerks over whether past research should factor into the final verdict on a drug. But for patients, the cost of starting from scratch every time can be high. 

For people with rare diseases, where only a few hundred individuals worldwide might have a condition, running a traditional trial can be nearly impossible, because there simply aren’t enough patients to enroll. For children, it has meant re-proving what we already learned in adults. And for everyone, it has meant slower, more expensive trials that throw away useful information.

Now, the FDA is telling drug companies and researchers they don’t have to start from scratch anymore.

Last week, the agency released new guidance encouraging companies to use a statistical approach, that would usually be used on a case-by-case basis, called Bayesian methods. (We’ll get more into that later.) 

What that means is that, for the first time, companies can formally incorporate what they already know — from earlier studies, from related drugs, from real-world evidence — to help answer the central question of whether a drug works. The FDA’s guidance is still a draft, and details may shift over the coming months, but the policy signal is clear. 

“It sounds so intuitive to just use the data that you have before to inform the next thing that you do,” said Merit Cudkowicz, a neurologist at Massachusetts General Hospital who runs a major ALS clinical trial, “instead of just having this sort of amnesia.”

Two ways of looking at the world

For a drug to get FDA approval, it has to prove it works in three phases of clinical trials. But “proving it works” can mean different things, depending on how you handle uncertainty. 

The traditional approach — called frequentist statistics — asks a narrow question: If this drug doesn’t actually work, how likely is it that we’d see results this strong just by chance? If that probability is very low (typically below 5 percent), the drug passes the test. The appeal is objectivity; the trial data speaks for itself, and what you believed going in doesn’t formally enter the math.

Bayesian statistics, the new rule of the land, flips the question. It asks: Based on everything we already know, how likely is it that this drug works? Then, it updates that estimate as new trial data comes in. The result isn’t a binary pass/fail, but a probability — say, a 94 percent chance the drug is effective. That doesn’t mean anything goes, and the FDA still has to draw a line in the sand that’s pre-agreed before the trial runs. 

The practical upshot is that Bayesian methods let you formally “borrow” information from other places. If you’ve already tested a drug in adults, you can use that data when evaluating it in children. If you’re running a trial with multiple drugs, data from one arm of the study can inform another. This flexibility matters most in situations where patients are hard to come by. 

“The availability of prior information is why we see such use in pediatric,” said James Travis, a statistician in the FDA’s drug review division. “We pretty much always have adult information, so it’s very easy to do things like that in the pediatric space.”

But being able to bring in outside information raises one obvious concern: What is stopping researchers from cherry-picking the studies that make their drug look good?

Traditional trials have a hard threshold — the “p-value”, a measure of whether results are likely due to chance — that seems to remove human judgment out of the equation. You either hit statistical significance, or you don’t. Bayesian methods, by contrast, require researchers to choose “priors,” or assumptions about what they expect to find based on existing evidence. 

But this critique assumes that traditional trials are capital-O objective, and that’s not necessarily the case; they just hide their assumptions better.

Every clinical trial involves choices: which patients to enroll, what outcomes to measure, what comparisons to make. A p-value can make it seem like the math is deciding, when, in fact, subjective judgments are baked in throughout.

Bayesian methods, proponents argue, force those assumptions into the open. You have to state your priors upfront, and justify them. And then everyone — including FDA reviewers — can see exactly what you assumed and evaluate whether it was reasonable.

Why patients care about statistics

All of this might sound like an academic statistical debate. But for people with serious diseases and their loved ones, the stakes are stark. 

Consider Amyotrophic Lateral Sclerosis (ALS), a neurodegenerative disease that kills most patients within two to five years of diagnosis. Around 5,000 Americans are diagnosed each year, according to CDC’s National ALS registry.

But despite decades of research, drug trials kept failing. Testing one drug at a time, starting essentially from scratch each time, was painfully slow for a disease that doesn’t have much wait time. 

In 2019, the FDA greenlit an unusually Bayesian trial to hunt for new ALS drugs. In the HEALEY ALS Platform Trial, researchers at Massachusetts General Hospital were able to test multiple ALS drugs at once, fast enough to matter for patients who didn’t have time to wait. Data from patients in one part of the trial — including those receiving placebos — can be used to inform drugs in other parts of the large-scale trial. This means the trial can drop drugs that aren’t working and add promising ones without starting over each time.

In the four years the trial has been running, seven drugs have been tested so far. A traditional approach might have managed just two. The new FDA statistical guidance, Cudkowicz said, should clear the path for other trials to follow this sort of model.

“The patients enrolled so fast because the patients with ALS felt that this was a patient-centered trial,” said Merit Cudkowicz, the neurologist who leads the study. Two of those drugs showed enough promise that they’re now advancing to final-stage trials.

“The Bayesian approach is just trying to take all of that data that participants give – and they give a lot of themselves – and use it in the most effective way,” said Melanie Quintana, a statistician at Berry Consultants, who helped design the HEALEY trials.

The catch

More flexibility also means more room for things to go wrong. 

A 2018 review, co-authored by Aaron Kesselheim, a Harvard professor who studies FDA policy, examined more than 100 adaptive trials, a related approach that also allows mid-trial adjustments and often uses Bayesian methods. They found that only a third of trials used independent committees to monitor the data, and just 6 percent kept statisticians blinded when analyzing mid-trial. Without these safeguards, there’s more room for bias to creep in or for early results to mislead.

FDA officials say the safeguards for Bayesian trials will remain. Every proposal will be reviewed by agency statisticians, and companies must lock in their methods before the trial starts. 

“It’s not like you get to pick the prior after you’ve seen the data,” John Scott, who oversees biostatistics at the FDA. “There’s really strict rules about that.” 

But whether individual companies actually start using these methods is another question. The guidance is not yet set in stone. The proposal is open for public comment until March 13, with a final version expected in about 18 months. And with FDA facing leadership turnover and political uncertainty, companies may be even more cautious about trying something new. 

“Drug companies hate uncertainty,” said Adam Kroetsch, a former FDA official who has written about the agency’s evolution. “They might decide it’s not worth the risk and just go with the traditional approach where they know there’s FDA precedent.”

But the FDA isn’t alone in this shift – the European Medicines Agency has also been exploring expanded use of Bayesian methods in drug development.

For patients with rare diseases, or for children waiting on treatments that already work in adults, the stakes of this statistical change are potentially life or death. The HEALEY trial has already shown what’s possible, and the FDA has opened the door. Now, more companies have to walk through it.

2026-01-21 19:00:00

澳大利亚的珊瑚礁修复与适应计划（RRAP）的研究技术人员萨拉·戈迪内斯-埃斯皮诺萨（Sara Godinez-Espinosa）在近库林斯的国家海模拟器（National Sea Simulator）中将一个名为“Acropora kenti”的分枝珊瑚成年群落放入一个容器中。| 哈里特·斯派克（Harriet Spark）为Vox报道

在澳大利亚昆士兰北部海岸约25英里处，一个阴云密布的12月夜晚，大约晚上10点，海洋科学家彼得·哈里森（Peter Harrison）俯身船边，用手电筒照向黑暗的海水，说道：“我闻到了，真的闻到了。”他和一群科学家、旅游业经营者以及原住民一起，在过去几天里一直在大堡礁上寻找珊瑚产卵。据哈里森称，嗅觉是唯一能发现珊瑚产卵的方法。珊瑚产卵是地球上最大的繁殖事件之一，也被称为“世界上最大的性高潮”。在某些区域，珊瑚产卵如此密集，以至于在海面上形成大片带有纹理的滑腻物质，就像发生了化学泄漏一样。一些人说珊瑚产卵闻起来像西瓜或新鲜牛奶，而对哈里森来说，它只是略带鱼腥味。

“我们准备好了，”另一位科学家马克·吉布斯（Mark Gibbs）说道，他是澳大利亚海洋科学研究所（AIMS）的工程师。他和船员们开始用改良的泳池网打捞产卵，并将它们倒入一个大塑料桶中。那天晚上，团队收集了数以十万计的珊瑚卵，作为一项艰巨努力的一部分，试图让大堡礁继续生存下去。

全球变暖和其他挑战威胁着这个标志性生态系统。大堡礁是澳大利亚的瑰宝，也是世界遗产地，是该国大规模旅游业的重要支柱。面对这些生存威胁，澳大利亚政府启动了名为“大堡礁修复与适应计划”（RRAP）的项目。该项目的目标是帮助世界上最大的珊瑚礁生态系统应对气候变化。该项目汇聚了近3亿美元的资金和数百名科学家、工程师和其他专家，是全球规模最大的珊瑚礁保护项目之一。

RRAP与原住民合作，许多原住民依赖大堡礁生存，并愿意帮助其恢复。大堡礁的珊瑚并非单一的结构，而是由约3000个独立珊瑚礁组成的群岛。尽管如此，大堡礁仍然壮观。在昆士兰的港口杜鲁和库林斯，我潜水时看到珊瑚的形状和颜色令人惊叹。一些珊瑚的蓝色深得仿佛被涂上了颜料。

大堡礁的珊瑚并不是单一的动物，而是由许多小动物——珊瑚虫组成的群体。珊瑚虫依赖特定的藻类生存，当海水温度过高时，它们会失去这些藻类，导致珊瑚变白并逐渐死亡。全球珊瑚礁的前景令人担忧。自1950年代以来，世界已经失去了大约一半的珊瑚覆盖面积，而过去二十年的损失更为严重。如果富裕国家继续燃烧化石燃料，全球气温上升超过2摄氏度，珊瑚礁可能完全消失。

一项发表在《自然通讯》（Nature Communications）上的研究预测，大堡礁的珊瑚覆盖面积将在未来15年内平均下降超过50%，即使是最乐观的排放情景也是如此。只有在立即、极端的碳排放削减情况下，大堡礁才有可能恢复到现在的状态。

为了应对这一问题，RRAP正在尝试通过“辅助繁殖”来提高珊瑚的繁殖率。科学家们在海上收集珊瑚产卵，并将其放入受保护的浮池中，这些浮池连接着珊瑚礁。浮池中悬挂着数千个用于珊瑚幼虫附着的陶瓷结构，类似于植物苗圃中的空花盆。几周后，科学家们会将这些结构用于重新种植受损的珊瑚礁。

在实验室中，科学家们也在培育珊瑚，以恢复大堡礁的生态。国家海模拟器（SeaSim）位于库林斯以南几小时的汤斯维尔郊区，是世界上最大的珊瑚繁殖实验室之一。通过观察，我看到珊瑚在模拟器中自然产卵，产卵后，科学家们将胚胎转移到更大的室内水箱中，再转移到其他水箱中，让它们附着在混凝土小块上。然后，这些混凝土块被插入陶瓷结构中，用于重新种植珊瑚礁。

尽管RRAP的规模巨大，但其仍无法解决根本问题：温室气体排放的增加。虽然修复工作可能在短期内维持珊瑚礁的存在，但如果全球不立即减少碳排放，这些成果将是短暂的。哈里森表示：“这一切都建立在世界会立即减少排放的前提之上。如果我们不这样做，那一切努力都是徒劳，因为这是一场无法逆转的列车。”

然而，RRAP的执行主任塞德里克·罗比洛（Cedric Robillot）认为，即使世界停止燃烧化石燃料，这些生态系统仍会继续衰退，仍然需要我们的支持和帮助。目前，全球正在跨越1.5摄氏度的临界点，预计大部分珊瑚礁将死亡。罗比洛说：“即使今天停止排放，珊瑚礁仍会遭受损失。而我们不可能今天就停止排放。”

尽管大堡礁的保护工作令人感到荒谬，但这些努力是科学家、原住民和有远见的潜水者所能采取的措施之一。Divers for Climate的创始人和CEO约兰达·沃特（Yolanda Waters）告诉我：“在大堡礁上，有如此多的工作正在进行。所有这些科学家和旅游业经营者都在尽他们所能。如果澳大利亚政府能说‘我们也可以为大堡礁做些什么’，并加强气候行动，那就更好了。”

CAIRNS, Australia — “I just got a whiff,” said Peter Harrison, a marine scientist, as he leaned over the edge of the boat and pointed his flashlight into the dark water. “It’s really coming through now.” 

It was shortly after 10 pm on a cloudy December night, and Harrison, a coral researcher at Australia’s Southern Cross University, was about 25 miles off the coast of northern Queensland. He was with a group of scientists, tourism operators, and Indigenous Australians who had spent the last few nights above the Great Barrier Reef — the largest living structure on the planet — looking for coral spawn.  

And apparently, it has a smell. 

Over a few nights in the Australian summer, shortly after the full moon, millions of corals across the Great Barrier Reef start bubbling out pearly bundles of sperm and eggs, known as spawn. It’s as if the reef is snowing upside down. Those bundles float to the surface and break apart. If all goes to plan, the eggs of one coral will encounter the sperm of another and grow into free-swimming coral larvae. Those larvae make their way to the reef, where they find a spot to “settle,” like a seed taking root, and then morph into what we know of as coral. 

Spawning on the Great Barrier Reef has been called the largest reproductive event on Earth, and, in more colorful terms, “the world’s largest orgasm.” Coral spawn can be so abundant in some areas above the reef that it forms large, veiny slicks — as if there had been a chemical spill.

This was what the team was looking for out on the reef, and sniffing is one of the only ways to find it, said Harrison, who was among a small group of scientists who first documented the phenomenon of mass coral spawning in the 1980s. Some people say coral spawn smells like watermelon or fresh cow’s milk. To me it was just vaguely fishy. 

“Here we go,” said Mark Gibbs, another scientist onboard and an engineer at the Australian Institute of Marine Science (AIMS), a government agency. All of a sudden the water around us was full of little orbs, as if hundreds of Beanie Babies had been ripped open. “Nets in the water!” Gibbs said to the crew. A few people onboard began skimming the water’s surface with modified pool nets for spawn and then dumping the contents into a large plastic bin.

That night, the team collected hundreds of thousands of coral eggs as part of a Herculean effort to try to keep the Great Barrier Reef alive. Rising global temperatures, together with a raft of other challenges, threaten to destroy this iconic ecosystem — the gem of Australia, a World Heritage site, and one of the main engines of the country’s massive tourism industry. In response to these existential threats, the government launched a project called the Reef Restoration and Adaptation Program (RRAP). The goal is nothing less than to help the world’s greatest coral reef survive climate change. And with nearly $300 million in funding and hundreds of people involved, RRAP is the largest collective effort on Earth ever mounted to protect a reef.

The project involves robots, one of the world’s largest research aquariums, and droves of world-renowned scientists. The scale is unlike anything I’ve ever seen.

But even then, will it be enough?  

The first thing to know about the Great Barrier Reef is that it’s utterly enormous. It covers about 133,000 square miles, making it significantly larger than the entire country of Italy. And despite the name, it’s not really one reef but a collection of 3,000 or so individual ones that form a reef archipelago. 

Another important detail is that the reef is still spectacular.

Over three days in December, I scuba dived offshore from Port Douglas and Cairns, coastal cities in Queensland that largely run on reef tourism, a whopping $5.3 billion annual industry. Descending onto the reef was like sinking into an alien city. Coral colonies twice my height rose from the seafloor, forming shapes mostly foreign to the terrestrial world. Life burst from every surface.

What really struck me was the color. Two decades of scuba diving had led me to believe that you can only find vivid blues, reds, oranges, and pinks in an artist’s imaginings of coral reefs, like in the scenes of Finding Nemo. But coral colonies on the reefs I saw here were just as vibrant. Some of the colonies of the antler-like staghorn coral were so blue it was as if they had been dipped in paint.

It’s easy to see how the reef — built from the bodies of some 450 species of hard coral — provides a foundation for life in the ocean. While cruising around large colonies of branching coral, I would see groups of young fish hiding out among their nubby calciferous fingers. The Great Barrier Reef is home to more than 1,600 fish species, many of which are a source of food for Indigenous Australians and part of a $200 million commercial fishing industry. 

“The reef is part of our life,” said Cindel Keyes, an Indigenous Australian of the Gunggandji peoples, near Cairns, who was part of the crew collecting coral spawn with Harrison. RRAP partners with First Nations peoples, many of whom have relied on the reef for thousands of years and are eager to help sustain it. “It’s there to provide for us, too,” Keyes, who comes from a family of fishers, told me.

The Great Barrier Reef is not dead, as many visitors assume from headlines. But in a matter of decades — by the time the children of today grow old — it very well could be. 

The world’s coral reefs face all kinds of problems, from big storms to runoff from commercial farmland, but only one is proving truly existential: marine heat. Each piece of coral is not one animal but a colony of animals, known as polyps, and polyps are sensitive to heat. They get most of their food from a specific type of algae that lives within their tiny bodies. But when ocean temperatures climb too high, polyps eject or otherwise lose those algae, turn bleach-white, and begin to starve. If a coral colony is “bleached” for too long, it will die.

The global prognosis is bleak. The world has already lost about half of its coverage of coral reefs since the 1950s, not including steep losses over the last two decades. And should wealthy countries continue burning fossil fuels — pushing global temperatures more than 2 degrees Celsius above the pre-industrial baseline — it will likely lose the rest of it. 

Projections for the Great Barrier Reef are just as grim. A recent study published in the prestigious journal Nature Communications projected that coral cover across the reef would decline, on average, by more than 50 percent over the next 15 years, under all emissions scenarios — including the most optimistic. The reef would only later recover to anything close to what it looks like today, the authors wrote, if there are immediate, near-impossibly steep emissions cuts. (The study was funded by RRAP.)

The reef has already had a taste of this future: In the last decade alone, there have been six mass bleaching events. One of the worst years was 2016, when coral cover across the entire reef declined by an estimated 30 percent. Yet recent years have also been alarming. Surveys by AIMS found that bleaching last year affected a greater portion of the reef than any other year on record, contributing to record annual declines of hard coral in the northern and southern stretches of the reef.

“I’ve been suffering,” said Harrison, who’s been diving on the Great Barrier Reef for more than 40 years. “I’ve got chronic ecological grief. Sometimes it’s overwhelming, like when you see another mass bleaching. It can be quite crushing.” 

The problem isn’t just bleaching but that these events are becoming so frequent that coral doesn’t have time to recover, said Mia Hoogenboom, a coral reef ecologist at Australia’s James Cook University, who’s also involved in RRAP.

“The hopeful part is if we can take action now to help the system adapt to the changing environment, then we’ve got a good chance of keeping the resilience in the system,” Hoogenboom said. “But the longer we wait, the less chance we have to maintain the Great Barrier Reef as a functioning ecosystem.”

That night in December, after filling two large plastic bins onboard with coral spawn, the crew motored to a nearby spot on the reef where several inflatable pools were floating on the ocean’s surface. The boat slowly approached one of the pools — which looked a bit like a life raft — and two guys onboard dumped spawn into it. 

The government established RRAP in 2018 with an ambitious goal: to identify tools that might help the reef cope with warming, refine them through research and testing, and then scale them up so they can help the reef at large. It is a massive undertaking. RRAP involves more than 300 scientists, engineers, and other experts across 20-plus institutions, including AIMS, which operates one of the world’s largest research aquariums called the National Sea Simulator. And it has a lot of money. The government committed roughly $135 million to the project, and it has another $154 million from private sources, including companies and foundations. It’s operating on the scale of decades, not years, said Cedric Robillot, RRAP’s executive director.

Scientists at RRAP have now honed in on several approaches that they think will work, and a key one is assisted reproduction — essentially, helping corals on the reef have babies. That’s what scientists were doing on the water after dark in December.  

Normally, when corals spawn, only a fraction of their eggs get fertilized and grow into baby corals. They might get eaten by fish, for example, or swept out to sea, away from the reef, where the larvae can’t settle. That’s simply nature at work in normal conditions. But as the reef loses more and more of its coral, the eggs of one individual have a harder time meeting the sperm of another, leading to a fertility crisis. 

RRAP is trying to improve those odds through what some have called coral IVF.

At sea, scientists skim spawn from the surface and then load them into those protected pools, which are anchored to the reef. Suspended inside the pools are thousands of palm-sized ceramic structures for the larval coral to settle on, like empty pots in a plant nursery. After a week or so, scientists will use those structures — which at that point should be growing baby corals — to reseed damaged parts of the reef.

With this approach, scientists can collect spawn from regions that appear more tolerant to warming and reseed areas where the corals have been killed off by heat. Heat tolerance is, to an extent, rooted in a coral’s DNA and passed down from parent to offspring. So those babies may be less likely to bleach and die. While baby corals are growing in those pools, scientists can also introduce specific kinds of algae — the ones that live symbiotically within polyps — that are more adapted to heat. That may make the coral itself more resistant to warming.

But what’s even more impressive is that scientists are also breeding corals on land, at the National Sea Simulator, to repopulate the reef. SeaSim, located a few hours south of Cairns on the outskirts of Townsville, is essentially a baby factory for coral. 

I drove to SeaSim one evening in December with Robillot, a technophile with silver hair and a French accent. He first walked me through a warehouse-like room filled with several deep, rectangular tanks lit by blue light. The light caused bits of coral growing inside them to fluoresce. Other than the sound of running water, it was quiet.

The main event — one of the year’s biggest, for coral nerds anyway — was just outside. 

SeaSim has several open-air tanks designed to breed corals with little human intervention. Those tanks, known as autospawners, mimic the conditions on the wild reef, including water temperature and light. So when scientists put adult corals inside them, the colonies will spawn naturally, as they would in the wild. The tanks collect their spawn automatically and mix it together in another container that creates the optimal density of coral sperm for fertilization. 

Observing spawning isn’t easy. It typically happens just once a year for each species, and the timing can be unpredictable. But I got lucky: Colonies of a kind of branching coral known as Acropora kenti were set to spawn later that evening. Through glass panels on the side of the autospawners, I saw their orangish branches, bunched together like the base of a broom. They were covered in pink, acne-like bumps — the bundles of spawn they were getting ready to release — which was a clear sign it would happen soon.  

As it grew dark, the dozen or so people around the tanks flipped on red headlamps to take a closer look. (White light can disrupt spawning.) Around 7:30 pm, the show started. One colony after another popped out cream-colored balls. They hung for a moment just above the coral branches before floating to the surface and getting sucked into a pipe. It was a reminder that corals, which usually look as inert as rocks, really are alive. “It’s such a beautiful little phenomenon,” Robillot said, as we watched together. “It’s a sign that we still have vitality in the system.”

After spawning at SeaSim, scientists move the embryos into larger, indoor tanks, where they develop into larvae. Those larvae then get transferred to yet other tanks, settling on small tabs of concrete. Scientists then insert those tabs into slots on small ceramic structures — those same structures as the ones suspended in the floating pools at sea — which they’ll use to reseed the reef. One clear advantage of spawning corals in a lab is that scientists can breed individual corals that appear, through testing, to be more resistant to heat. Ideally, their babies will then be a bit more resistant, too.

During spawning late last year, SeaSim produced roughly 19 million coral embryos across three species.

“People often don’t understand the scale that we’re talking about,” said Carly Randall, a biologist at AIMS who works with RRAP. “We have massive numbers of autospawning systems lined up. We have automated image analysis to track survival and growth. It is like an industrial production facility.”

Including the spawn collection at sea, RRAP produced more than 35 million coral embryos last year that are now growing across tens of thousands of ceramic structures that will be dropped onto the reef. The goal RRAP is working toward, Robillot says, is to be able to stock the reef with 100 million corals every year that survive until they’re at least 1 year old. (Under the right conditions, each ceramic structure can produce one coral that lives until 1 year old in the ocean, Robillot told me. That means RRAP would need to release at least a million of those structures on the reef every year.)

On that scale, the project could help maintain at least some coral cover across the reef, even in the face of more than 2 degrees C of warming, Robillot said, citing unpublished research. One study, published in 2021 and partially funded by RRAP, suggests that a combination of interventions, including adding heat-tolerant corals, can delay the reef’s decline by several years.

“We are not replacing reefs,” Robillot said. “It’s just too big. We’re talking about starting to change the makeup of the population by adapting them to warmer temperatures and helping their recovery. If you systematically introduce corals that are more heat-tolerant over a period of 10 to 20 to 30 years, then over a hundred years, you significantly change the outlook for your population.”

The obvious deficiency of RRAP, and many other reef conservation projects, is that it doesn’t tackle the root problem: rising greenhouse gas emissions. While restoration might help maintain some version of coral reefs in the near term, those gains will only be temporary if the world doesn’t immediately rein in carbon emissions. “It all relies on the premise that the world will get its act together on emissions reductions,” Robillot said. “If we don’t do that, then there’s no point, because it’s a runaway train.”  

Many groups involved in reef conservation have failed to reckon with this reality, even though they’re often on the front lines of climate change. During my trip, I would be on dive boats listening to biologists talk about restoration, while we burned diesel fuel and were served red meat — one of the most emissions-intensive foods. A lot of tour operators, some of whom work with RRAP, don’t talk about climate change much at all. Two of the guides who took me out on the reef even downplayed the threat of climate change to me.

Yolanda Waters, founder and CEO of Divers for Climate, a nonprofit network of scuba divers who care about climate change, said this isn’t surprising. “At the industry level, climate change is still very hush-hush,” said Waters, who previously worked in the reef tourism industry. “In most of those boats, climate messaging is just nonexistent.”

This makes some sense. Tourism companies don’t want people to think the reef is dying. “When international headlines describe the Reef as ‘dying’ or ‘lost,’ it can create the impression that the visitor experience is no longer worthwhile, even though large parts of the Reef remain vibrant, actively managed, and accessible,” Gareth Phillips, CEO of the Association of Marine Park Tourism Operators, a trade group, told me by email. (I asked around, but no one could point me to data that clearly linked negative media stories to a drop in visitors to the Great Barrier Reef.)

Yet by failing to talk about the urgent threat of climate change, the tourism industry — a powerful force in Australia, that influences people from all over the world — is squandering an opportunity to educate the public about what is ultimately the only way to save the reef, said Tanya Murphy, a campaigner at the Australian Marine Conservation Society, a nonprofit advocacy group. Tourists are ending their vacation with the memory of, say, a shark or manta ray, not a new urge to fight against climate change, Waters said. So the status quo persists: People don’t connect reducing emissions with saving the reef, even though that’s “the only reef conservation action that can really be taken from anywhere,” she added. 

(Not everyone in the tourism industry is so quiet. Eric Fisher, who works for a large Australian tourism company called Experience Co Limited, says he tells tourists that climate change is the biggest threat to the Great Barrier Reef. “It’s what we tell people every day,” Fisher told me. “So as they fall in love with it, they’re more likely to leave with an understanding of that connection.”)

Keeping mum on climate change, while speaking loudly about restoration and other conservation efforts, including RRAP, can also take pressure off big polluters to address their carbon footprints, Waters and Murphy said. Polluters who fund reef conservation, including the government and energy companies, are given social license to operate without stricter emissions cuts, because the public thinks they’re doing enough, they said.

In reality, the Australian government continues to permit fossil fuel projects. Last year, for example, the Albanese administration, which is politically left of center, approved an extension of a gas project in Western Australia that Murphy and other advocates call “a big carbon bomb.” The extension of the project, known as the North West Shelf, will produce carbon emissions equivalent to about 20 percent of Australia’s current yearly carbon footprint, according to The Guardian. 

A spokesperson for the Albanese government acknowledged in a statement to Vox that climate change is the biggest threat to coral reefs globally. “It underlines the need for Australia and the world to take urgent action, including reaching net zero emissions,” the statement, sent by Sarah Anderson, said. “The Albanese Government remains committed to action on climate change and our net zero targets.”

Anderson highlighted a government policy called the Safeguard Mechanism, which sets emissions limits for the country’s largest polluters, including the North West Shelf Facility. Yet the policy only applies to Scope 1 emissions. That means it doesn’t limit emissions tied to gas that the North West Shelf project exports — the bulk of the project’s carbon footprint.

Although Australia has far fewer emissions compared to large economies like the US and China, the country is among the dirtiest on a per-capita basis. If any country can reduce its emissions, it should be Australia, Waters said. “We’re such a wealthy, privileged country,” Waters said. “We’ve got the biggest reef in the world. If we can do better, why wouldn’t we?”

On a stormy morning, near the end of my trip, we returned to the reef — this time, visiting another set of floating pools, offshore from Port Douglas. They had been filled with spawn several days earlier. Small corals were now growing on the ceramic structures, and they were ready to be deployed on the reef. 

After a nauseating two-hour ride out to sea, a group of scientists and tourism operators jumped into small tenders and collected the structures from inside the pools. Then they motored around an area of the reef that had previously been damaged by a cyclone and started dropping coral babies off the side of the boat, one by one. 

As it started to pour, and I noticed water flooding into the front of the tender, I couldn’t help but think about how absurd all of this was. Custom-made pools and ceramics. Hours and hours on the reef, floating in small boats in a vast ocean. Sniffing out spawn.

“You sort of think about the level of effort, that we’re going to try and rescue something that’s been on our planet for so many millions of years,” Harrison told me on the boat a few nights earlier. “It seems a bit ironic that humans now have to intervene to try and rescue corals.”

RRAP is making this process far more efficient, Robillot says — machines, not people, will eventually be dropping the ceramic structures off the boats, for example. But still, why not invest the money instead in climate advocacy or clean energy? Isn’t that an easier, perhaps better, way to help?

It can’t be either or, Robillot said. And it’s not, he contends. Many donors who fund the Great Barrier Reef Foundation, a core RRAP partner and Robillot’s employer, are putting more of their money into climate action relative to reef conservation, he said. The government of Australia, meanwhile, says it’s spending billions on clean energy and green-lit a record number of renewable energy projects in 2025. Plus, while the scale of resources behind RRAP is certainly huge for coral reefs, it’s tiny compared to the cost of fixing the climate crisis. “We need trillions,” Robillot said. 

Investing that roughly $300 million into fighting climate change could have a small impact on reefs decades from now. Putting it into projects like RRAP helps reefs today. It’s only a waste of money — worse than a waste of money — if that investment undermines climate action. And Robillot doesn’t think it does. 

The Great Barrier Reef Foundation has been criticized for its ties to mining and energy companies, including Peabody Energy and BHP. The Reef Foundation currently receives money from mining giant Rio Tinto and BHP Foundation (which is funded by BHP) for projects unrelated to RRAP, the organization told Vox. “It is a bit concerning,” Murphy told me. “It’s really important that we get polluters to pay for the damage they’re causing. But that should be done as an obligatory tax and they should not be getting any marketing benefits from that.”

Robillot argues that these companies have not influenced RRAP’s work, or restricted what its staff can say about climate change. “If we can still scream that climate change is the main driver of loss of coral reefs, I don’t have an issue,” he said. “I don’t think it’s realistic to only take money from people who do not have any impact on climate change. I don’t know anyone.”

Yet if there’s one argument that I find most convincing for RRAP — for any project trying to help wildlife suffering from climate change — it’s that even if the world stops burning fossil fuels, these ecosystems will still decline. They will still need our support, our help to recover. The planet is currently crossing the 1.5-degree threshold, at which point the majority of coral reefs worldwide are expected to die off. “If you stop emissions today, they will still suffer,” Robillot said of reefs. “And we’re not going to stop emissions today.” 

So much of reef conservation is absurd. We shouldn’t need to collect coral spunk from the open ocean in the middle of the night or breed these animals in tanks on land. Then again, these sorts of efforts are what scientists, Indigenous Australians, and the most thoughtful divers can do — what they are doing — to help the reef today.

“There’s so much work happening on the ground,” Waters, of Divers for Climate, told me. “All of those scientists, all of those [tourism] operators, are genuinely doing everything they can. It would be great for the Australian government to go, ‘Well, this is what we can do for reefs, too,’ pick up their game on climate, and show that we’re actually in it together.”

2026-01-21 07:10:00

2026年1月16日，唐纳德·特朗普在白宫南草坪向记者发表讲话后，乘坐海军陆战队一号直升机离开。本文出自《The Logoff》——一份帮助你了解特朗普政府动态，同时避免政治新闻占据你生活的每日简报。订阅请前往此处。

欢迎来到《The Logoff》：一年前重返白宫后，特朗普变得更加不受约束，也更加不受欢迎。发生了什么？周二，特朗普花费近两个小时发表了一场冗长且不诚实的记者会，吹嘘自己过去一年的政绩，并随意谈论从汽油价格到他在皇后区的童年等话题。今晚，他将前往瑞士达沃斯，与那些世界秩序已陷入混乱的各国领导人进行交流。

为什么这重要？大约一个月前，特朗普曾发表过一场同样奇怪的晚间演讲，但语气不同（大声喊叫，而不是含糊不清）。当时，我的同事扎克·比厄姆普指出，这表明特朗普正在与政治现实对抗。这一说法至今仍然成立——特朗普的民意支持率极低，但不确定的是，他是否意识到自己正在失去公众的支持（或者是否在意）。而一个陷入幻觉的特朗普，可能比一个正在挣扎的特朗普更加危险。

背景是什么？周末时，特朗普再次向挪威首相发出关于格陵兰的威胁，称诺贝尔和平奖是其理由之一。他还威胁对反对他格陵兰行动的欧洲国家征收关税。此外，我们还了解到特朗普提议的“和平委员会”，这个机构将类似于联合国，而特朗普本人将担任主席。周二，加拿大总理马克·卡尼警告称，美国主导的世界秩序可能会出现“破裂”。

大局如何？短期内，特朗普显然认为他可以为所欲为。这在格陵兰、对委内瑞拉的攻击以及对明尼苏达居民的强硬行动中都得到了体现。但他的日益膨胀的权力主张可能会掩盖这样一个事实：特朗普无法改变很多东西，包括人们对他的看法（周二他暗示可能是他的“公关团队”出了问题）。随着中期选举临近，他似乎没有打算去解决这些问题。

好了，是时候下线了。读者们，临别之际：一年的“特朗普二世”也意味着一年的《The Logoff》。感谢你们的支持和阅读时间。希望这份简报能帮助你们更好地了解特朗普政府的重要新闻，同时保持理智。如果你希望支持Vox的新闻报道，最好的方式是成为会员。我也很乐意听听你们的想法：你们对特朗普世界有什么迫切的问题？有没有什么内容希望我们多报道？欢迎通过电子邮件[email protected]与我联系。

最后，一件好事：今天是企鹅日。你知道吗？这是麦哲伦企鹅的声音。

This story appeared in The Logoff, a daily newsletter that helps you stay informed about the Trump administration without letting political news take over your life. Subscribe here.

Welcome to The Logoff: One year after returning to the White House, Donald Trump is more unconstrained — and unpopular — than ever. 

What’s happening? Trump dedicated nearly two hours on Tuesday to a rambling, dishonest press conference touting his record over the past year and free-associating about everything from gas prices to his childhood in Queens. Tonight, he will depart for Davos, Switzerland, where he will speak to many of the leaders of a world order in shambles. 

Why does this matter? About a month ago, Trump gave an equally strange primetime address, albeit in a different register (shouty, rather than mumbling). At the time, my colleague Zack Beauchamp wrote that it revealed Trump was flailing against political gravity. That remains objectively true — Trump’s approval ratings are abysmal — but whether Trump realizes he’s losing the public (or cares) is less certain. And a delusional Trump may be even more dangerous than a flailing one.

What’s the context? Over the weekend, Trump reiterated his threats against Greenland in a message to Norway’s prime minister, citing the Nobel Peace Prize as part of his rationale. He also threatened tariffs against European nations that opposed his move on Greenland. We additionally learned more details about Trump’s proposed Board of Peace, which would resemble a quasi-UN with Trump at its head; and on Tuesday, Canadian Prime Minister Mark Carney warned about a “rupture” of the US-led world order.

What’s the big picture? In the short term, Trump clearly feels that he can do what he wants. This is as true for Greenland as it is for his attack on Venezuela, or his heavy-handed use of force against Minneapolis residents. 

But his increasingly total assertions of power at home and abroad can distract from the fact that there’s a lot Trump can’t change, including how the country feels about him (on Tuesday, he suggested maybe his “bad public relations people” were at fault). With the midterms approaching, it doesn’t seem like he has any plans to figure it out.

And with that, it’s time to log off…

Hi readers, before we go: One year of Trump II also means one year of The Logoff. We’re so thankful for all of your support and for taking the time to read and share this newsletter. Hopefully, it’s been a helpful and sanity-preserving way to keep up with the most important news out of the Trump administration. If you want to support Vox’s journalism, the best way to do so is by becoming a member.

I’d also love to hear from you: What burning questions do you have about what’s happening in Trumpworld? Is there something you want to see us cover more often? You can shoot me an email at [email protected] to let me know how we’re doing. 

Lastly, a good thing: It is Penguin Awareness Day. Did you know this is what Magellanic penguins sounds like? 

2026-01-21 04:00:25

美国总统唐纳德·特朗普再次将格陵兰岛置于其外交政策的核心位置，重新推动了这一长期举措，令美国盟友感到不安。最新举措是任命路易斯安那州州长杰夫·兰德里为格陵兰岛特别大使，负责推进特朗普将该地区纳入美国控制范围的目标。特朗普称该岛对美国国家安全至关重要。丹麦和格陵兰迅速驳回了这一提议，发表联合声明强调主权和边界受国际法保护。其他欧洲领导人，包括法国总统埃马纽埃尔·马克龙，也支持丹麦和格陵兰。特朗普自上任以来就曾提出购买或兼并格陵兰岛，但其言论在第二个任期中变得更加激烈。这一重新关注的举措反映了美国政府的国家安全战略，强调美国在西半球的主导地位，并限制中国和俄罗斯在北极地区的影响力。尽管美国官员与格陵兰和丹麦已有密切的军事合作，但特朗普的做法引发了对其盟友关系的新担忧。以下内容将为您提供最新动态和分析：国会能否阻止特朗普试图占领格陵兰？特朗普对格陵兰的推动简要解释谁能阻止特朗普夺取格陵兰？特朗普的外交政策尚未跨越的一条红线特朗普再次提及格陵兰一位格陵兰居民解释格陵兰对特朗普的感受*特朗普对格陵兰的冒险举动真正的危险性

President Donald Trump has again elevated Greenland to a central place in his foreign policy, renewing a long-running push that has unsettled US allies. The latest move came with the appointment of Louisiana Gov. Jeff Landry as special envoy to Greenland, charged with advancing Trump’s goal of bringing the territory under US control. Trump has argued the island is vital for American national security.

Denmark and Greenland swiftly rejected the idea, issuing a joint statement underscoring that sovereignty and borders are protected by international law. Other European leaders, including French President Emmanuel Macron, voiced support for Denmark and Greenland.

Trump has floated buying or annexing Greenland since his first term, but his rhetoric has intensified during his second. The renewed focus reflects the administration’s National Security Strategy, which emphasizes US dominance in the Western Hemisphere and limiting Chinese and Russian influence, particularly in the Arctic. While US officials already enjoy close military cooperation with Greenland and Denmark, Trump’s approach has raised fresh concerns about relations with America’s allies.

Follow along here for the latest updates and analysis.

• Can Congress stop Trump from trying to take Greenland?
• Trump’s Greenland push, briefly explained
• Can anyone stop Trump from seizing Greenland?
• The one line that Trump’s foreign policy still hasn’t crossed
• Trump is talking about Greenland again
• How Greenland feels about Trump, explained by a Greenlander
• The real danger of Trump’s Greenland gambit

2026-01-21 03:05:00

2019年12月2日，支持枪支管控和枪支安全措施的支持者在最高法院外举行集会，当时大法官们正在听取“State Rifle and Pistol v. City of New York”一案的口头辩论。| Saul Loeb/AFP via Getty Images

在周二上午，最高法院的共和党多数大法官花费了大量时间试图理解如何让两个相互矛盾的原则同时成立。一方面，所有涉及第二修正案的案件都必须使用专门适用于第二修正案的法律规则进行判断；另一方面，持枪权不应与其他宪法权利受到不同的对待。2022年，在“New York State Rifle & Pistol Association v. Bruen”一案中，共和党大法官推翻了一项已有百年历史的纽约法律，该法律要求任何希望在公共场所携带手枪的人都必须证明“正当理由”才能获得许可证。而在周二的“Wolford v. Lopez”一案中，法院听取了对夏威夷州一项法律的挑战，该法律似乎是故意针对Bruen案而设计的。虽然Bruen案中的法律直接禁止大多数人携带枪支上街，但夏威夷的法律则通过要求枪支拥有者在进入某场所前必须获得该场所业主或经理的明确许可，间接达到同样的目的。由于很少有企业会给予这种许可，而枪支拥有者也很少会空手进入场所，因此夏威夷的法律实际上可能在大多数公共场所有效地禁止了枪支。

然而，Bruen案还宣布了一个奇怪的法律规则，仅适用于第二修正案案件。根据这一规则，只有当政府能够证明某项枪支管制措施与美国历史上枪支管制的传统一致时，该措施才是合宪的。因此，政府律师必须将现代法律与宪法制定时期“类似”的法律进行比较，以证明其一致性。如果法院认为旧法律与新法律足够相似，那么新法律就不会违反Bruen案的裁决。然而，这一专门针对第二修正案的规则过于模糊和不明确，以至于来自不同政治立场的法官都抱怨其难以适用。

在Wolford案中，夏威夷的律师提出了一个非常有力的论点，认为他们的法律应被Bruen案所接受。他们的简报列举了大量与夏威夷法律相似的旧法律，例如1771年新泽西州的一项法律，禁止人们在非自己拥有的土地上携带枪支，除非获得土地所有者的书面许可；1763年纽约州的一项法律，规定在“封闭土地”上携带枪支必须事先获得业主或持有者的书面许可。这些只是18世纪存在的一些类似法律的例子。

但事实证明，这些历史背景并不重要，因为最高法院的六位共和党大法官——包括对案件双方律师提出严厉质疑的艾米·康妮·巴雷特大法官——都表明他们很可能推翻夏威夷的法律。这表明，共和党大法官似乎希望在涉及枪支的案件中使用双重标准。当现代枪支法律与1790年代的枪支法律不相似时，他们可以使用Bruen案的规则来推翻该法律；而当现代法律与旧法律相似时，他们又会抱怨政府律师对第二修正案的处理与其他宪法权利不同，并据此推翻法律。

另一个迹象表明夏威夷法律可能面临挑战，是几位共和党大法官试图让律师Neal Katyal难堪，因为他简报中引用的一些旧法律可能是出于不良目的而制定的。例如，Katyal引用了重建时期后由路易斯安那州颁布的一项法律，据称该法律旨在剥夺黑人在私人土地上的持枪权。当然，针对种族的法律是违宪的，但它们并不违反第二修正案。这些法律违反了第十四修正案关于平等保护的条款。

无论如何，如果路易斯安那州是唯一要求枪支拥有者在进入私人土地前获得业主许可的州，那么这一历史案例将大大削弱夏威夷的法律论点。但这一被指控具有种族主义色彩的法律只是众多历史法律中的一项，而路易斯安那州可能只颁布了一项具有种族歧视性质的枪支法律，并不能否定Katyal简报中其他相似法律的合法性。

总的来说，历史上多个州都曾颁布过与夏威夷法律相似的法律，而这些法律中绝大多数似乎都是出于正当目的。如果Bruen案的裁决被诚实地应用，这些历史法律似乎要求法院支持夏威夷的法律。但最高法院的共和党多数派似乎并不愿意在他们不喜欢结果时适用Bruen案的规则。换句话说，当Bruen案的规则不利于枪支法律时，他们可以据此推翻该法律；而当该规则支持枪支法律时，他们又认为不能适用，因为这会意味着对第二修正案的处理与其他宪法权利不同。

The Supreme Court’s Republican majority spent much of Tuesday morning trying to figure out how two mutually exclusive principles can both be true at the same time. One principle is that all Second Amendment cases must be judged using a bespoke legal rule that only applies to the Second Amendment. The other principle is that the right to bear arms must not be treated differently than other constitutional rights.

Four years ago, in New York State Rifle & Pistol Association v. Bruen (2022), the Republican justices struck down a century-old New York law that required anyone who wishes to carry a handgun in public to demonstrate “proper cause” before they could obtain a license allowing them to do so. On Tuesday, the Court heard Wolford v. Lopez, a challenge to a Hawaii state law that appears to have been designed intentionally to sabotage Bruen. 

While the law at issue in Bruen directly banned most people from carrying a gun in public, Hawaii’s law tries to achieve this same goal indirectly by requiring gun owners to obtain explicit permission from a business’s owner or manager before they can bring a gun into that business. Because few businesses are likely to grant such permission — and few gun owners are likely to go into a business unarmed, ask the manager for permission, and then return with their weapon — Hawaii’s law is likely to operate as an effective ban on firearms in most public spaces.

But Bruen also announced a bizarre legal rule that applies only in Second Amendment cases. Under Bruen, a gun regulation is constitutional only if the government can “demonstrate that the regulation is consistent with this Nation’s historical tradition of firearm regulation.” Thus, government lawyers must prove that consistency by comparing the modern-day law to “analogous regulations” from the time when the Constitution was framed. If the courts deem the old laws to be sufficiently similar to the new law, then the new law does not violate Bruen.

This bespoke rule for Second Amendment cases is so vague and ill-defined that judges from across the political spectrum have complained that it is impossible to apply. But, in Wolford, Hawaii’s lawyers made a very strong argument that their law should survive Bruen. Their brief names an array of old laws that are very similar to the Hawaii law at issue in Wolford. 

A 1771 New Jersey law, for example, barred people from bringing “any gun on any Lands not his own, and for which the owner pays taxes, or is in his lawful possession, unless he has license or permission in writing from the owner.” A similar 1763 New York law made it unlawful to carry a gun on “inclosed Land” without “License in Writing first had and obtained for that Purpose from such Owner, Proprietor, or Possessor.” And these are just two examples of the kinds of laws that existed in the 1700s that resemble Hawaii’s law.

But it turns out that none of this history actually matters, as all six of the Court’s Republicans — including Justice Amy Coney Barrett, who did have some tough questions for lawyers on both sides of the case — signaled Tuesday that they are likely to strike the law down.

The Republican justices want to apply a double standard in Second Amendment cases

One of the Republican justices’ primary arguments against the Hawaii law was that the law would be unconstitutional if, instead of applying Bruen’s historical test, the Court were to apply a more normal approach to constitutional interpretation. 

Chief Justice John Roberts, for example, suggested that the First Amendment does not permit a state to forbid people from knocking on a private property owner’s door and asking for their vote. So why should the Second Amendment be read to allow states to bar this person from carrying a gun? As Roberts argued, one of the “motivating concerns” behind decisions like Bruen is that the right to bear arms has historically been treated as a “disfavored right.” And thus there shouldn’t be disparities between how the Court treats the First Amendment and how it treats the Second Amendment.

Similarly, Justice Samuel Alito accused Neal Katyal, the lawyer for Hawaii, of “just relegating the Second Amendment to second-class status.”

But if Roberts and Alito don’t like the fact that Second Amendment cases are treated differently than First Amendment cases, they have no one but themselves to blame. Again, Bruen announced a bespoke legal test, which fetishizes history, and which applies to no other constitutional right. So a court that fairly applies the Bruen test will sometimes reach different results than they would if they applied the legal rules that apply in First Amendment cases. 

If Roberts and Alito don’t like this reality, the obvious solution is to overrule Bruen.

The Republican justices, in other words, appear to want a double standard to apply in gun cases. When a modern-day gun law is not similar to gun laws from the 1790s, the Republican justices can apply Bruen and strike down the modern-day law under Bruen’s good-for-the-Second-Amendment-only legal standard. But when a modern-day gun law is similar to gun laws from the 1790s, then they can complain that the government’s lawyers are treating the Second Amendment differently than other constitutional rights — and strike down the law.

One other sign that the Hawaii law is in trouble is that several of the Republican justices tried to embarrass Katyal, because one of the many examples of old laws cited in his brief was probably enacted for nefarious reasons. One of the old laws Katyal cites in his brief is a post-Reconstruction law, enacted by Louisiana, which allegedly was enacted in order to disarm Black people on private land.

Of course, laws that target people because of their race are unconstitutional, but not under the Second Amendment. They violate the 14th Amendment’s guarantee that no one may be denied the equal protection of the laws. 

In any event, if Louisiana were the only state to require gun owners to obtain a property owner’s permission before bringing a gun onto their land, then that historical example would undercut Hawaii’s legal argument significantly. But this allegedly racist law is but one example of an historical law similar to Hawaii’s. And the fact that Louisiana may have enacted one racist gun law does not invalidate all of the other examples of similar laws in Katyal’s brief.

The bottom line is that several states historically enacted laws similar to Hawaii’s law, and all but one of those laws appear to have been enacted for benign reasons. If Bruen were applied honestly, this web of old laws seems to require courts to uphold Hawaii’s law.

But the Court’s Republican majority does not appear interested in applying Bruen when they do not like the outcome it produces. Again, when Bruen’s unique test cuts against a gun law, they can strike that law down under Bruen. And when Bruen’s historical test cuts in the other direction, the Republican justices appear to believe that they cannot apply Bruen, because that would mean treating the Second Amendment differently than other constitutional rights.