AbstractSome series can go on indefinitely, others cannot, and epistemologists want to know in which class to place epistemic chains. Is it sensible or nonsensical to speak of a proposition or belief that is justified by another proposition or belief, ad infinitum? In large part the answer depends on what we mean by “justification.” Epistemologists have failed to find a definition on which everybody agrees, and some have even advised us to stop looking altogether. In spite of this, the present essay submits a few candidate definitions. It argues that, although not giving the final word, these candidates tell us something about the possibility of infinite epistemic chains. And it shows that they can short‐circuit a debate about doxastic justification.