This chapter turns to the semantic level of reference and the True. Logical Investigations barely mentions reference at all, so I introduce the sense-reference distinction the way I believe Frege would have done had he continued the series. I present the distinction as primarily justified by the differences between cognitive and expressive acts in scientific as opposed to artistic contexts. This explanation requires a thorough examination of his views regarding reference in artistic contexts, which provides insights into his understanding of semantics in general. Against this backdrop, I then present what I believe would be Frege’s justification for presenting truth-values as the referents of sentences had he continued the series of essays, along with his ‘more appropriate way of speaking’ about truth. While doing so, I discuss Frege’s conception of facts and why he both identifies them with true thoughts and would not consider them suitable entities as referents of sentences. In the brief final section, I bring together the various theses regarding truth presented in this book and suggest how they fit with his claim, made early in Thoughts, that the laws of truth ‘spell out’ or ‘unfold’ truth.