The author tells us how medieval and early Renaissance music acquired its voices-cum-instruments interpretation, then shows that the recent shift to unaccompanied voices in music of these periods is the recovery of an earlier understanding. The first chapter, "The invention of the voices-and-instruments hypothesis," shows that those who rediscovered old music in the 19th century considered it purely vocal music. Even Hugo Riemann in his Musik Lexicon (1892 and 1893 editions) agreed completely. But in his 1905 edition, Riemann insisted that untexted parts of polyphonic songs were played on instruments, and many phrases of texted melodies were actually instrumental preludes, interludes and postludes. This soon became the standard point of view, for Riemann's book was a popular and influential reference. Guido Adler was one of the few scholars who disagreed. Medieval illustrations of singers and instrumentalists were misinterpreted to reinforce this thinking. In chapter 2, "The re-invention of the a cappella hypothesis," he recounts the early reaction to this orthodoxy in the 1950s and the turning point of Christopher Page's 1977 article in Early Music. Since then, Page, Andrew Parrott and Paul Hillier have led many performances and recordings that demonstrated the sound of early music without instruments, and David Fallows supported their efforts in his writings. Two more chapters fill in the background of the arguments. National prejudices and Nazi ideology come into play, and details are provided. No one who listens to early music will want to pass up this magisterial treatment of its 20th-century evolution. The fight is not yet over, for most recordings of this music still use instruments in the fashion that Leech-Wilkinson thoroughly discredits in this engaging book.