In an earlier article a reconstruction was proposed of the stemma of the primary manuscripts of Livy. If such a stemma has been correctly drawn up, it must work, that is, it must enable an editor to arrive by routine methods at the reading of the archetype. The archetype itself need not have good readings—it may have bad ones, emended by later manuscripts—but, good or bad, it gives the tradition from which all correction must start. If these readings make grammatical, (...) linguistic, and contextual sense and if there is no external tradition such as citations in grammarians or scholia, they may be taken to constitute what Livy wrote; if they do not make sense, then the editor must resort to correction on the basis of them. (shrink)