Drawing on literature on values in science and a case-study of UK cancer policy, this paper argues for a novel account of the demarcation project in terms of trustworthiness. The first part of the paper addresses the relationship between science, politics and demarcation. In 2010, the UK government decided to pay more for cancer drugs than for drugs for other diseases; in 2016, this Cancer Drugs Fund was reformed so as to lower the evidential standards for approving cancer drugs, rather than paying more for them. Are these two ways of treating cancer as “special” importantly different? This paper argues that, if we the argument from inductive risk seriously, they seem equivalent. This result provides further reason to doubt the notion of demarcating science from non-science. However, the second part of the paper complicates this story, arguing that considerations of epistemic trust might give us reasons to prefer epistemic communities centred around “broadly acceptable” standards, and which are “sociologically well-ordered”, regardless of inductive risk concerns. After developing these claims through the cancer case-study, the final section suggests how these concerns might motivate novel versions of the demarcation project.

本文借鉴有关科学价值的文献和英国癌症政策的案例研究，从可信度的角度提出了对划界项目的新解释。论文的第一部分讨论了科学、政治和划界之间的关系。2010年，英国政府决定为抗癌药物支付比其他疾病药物更高的费用；2016年，该抗癌药基金进行了改革，以降低批准抗癌药的证据标准，而不是支付更多的费用。这两种将癌症视为“特殊”的方法是否有重要的不同？本文认为，如果我们认真地从归纳风险论证，它们似乎是等价的。这一结果提供了进一步的理由来怀疑区分科学与非科学的概念。然而，论文的第二部分使这个故事复杂化了，认为认知信任的考虑可能让我们有理由更喜欢以“广泛接受”标准为中心的认知社区，这些社区是“社会学上有序的”，而不考虑归纳风险问题。在通过癌症案例研究提出这些主张之后，最后一部分提出了这些担忧如何激发新版本的划界项目。