I am very excited that my paper that starts by quoting English poetry has been published in The Auk: Ornithological Advances --- our first two lines are: "Percy Bysshe Shelley wrote in 1821 that the nightingale “sits in darkness and sings to cheer its own solitude with sweet sounds.” However, unlike English romantic poets, ornithologists know that nightingales sing not to stave off despair, but for sex—a primary reason birds sing is to attract and maintain mates." Graham Montgomery and I (ok, mainly Graham) have completed a huge number of playback experiments in the past year. Our aim is to study how songs evolve, and our protocol is to measure whether territorial individuals respond to (angrily approach the speaker) or ignore song from related populations that live somewhere else. We assume that song discrimination is a proxy for premating (not "permeating", as autocorrect wanted me to type) reproductive isolation. That is, two related populations that live in different places and ignore each others' song are unlikely to mate with each other should they ever come into contact. In this paper, we highlight 21 cases where related populations ignore each others' songs but are classified as a single species. We think this means the populations deserve species status; thus, the "21 new species" title! We also argue that playback experiments that "ask the birds themselves" are a better method to measure how different songs are than using computer software to compare audio recordings. Last, though this paper took a ton of fieldwork (again, mostly Graham there), it's not too hard to conduct these kind of playback experiments. We hope that our study motivates folks around the world to get out into the field with smartphones and wireless speakers in hand - we need this data to shed light on which populations should be considered distinct biological species!