The continuous growth of intellectual ecosystems leads to an environment populated by mutually uncomprehending hyperspecialised groups. That there are such groups can be taken, quite rightly, to reflect the way that human knowledge has expanded and deepened to such an extent that one person can only have a detailed grasp of a very narrow field. People no longer specialise in medicine and law but hyperspecialise in endocrinology and patent law. They occupy a niche that has its own epistemic standards and only those hyperspecialists make appropriate use of its standards. Enter gatekeepers whose sole responsibility it is to assist new entrants to navigate these standards, so that novices not be left stranded upon remote cognitive islands. Whilst they may be highly skilled practitioners, epistemic gatekeepers are not free from acting upon their own cognitive biases and prejudices. In this excerpt from my project, Why Facts Matter, I argue that facts should serve against not only stranding novices on cognitive islands but also privileging biased gatekeepers.