Ever found yourself sitting at home, saying "I wish I could find a book with superficial retellings of science like Malcolm Gladwell combined with the ego of Jordan Peterson?" Well now, friends, you can!
In "The World in Six Songs," Levitin claims to provide an argument that music throughout history can largely be reduced to six types of songs: friendship, joy, comfort, knowledge, religion, and love. As one might expect with such grandiose claims, this taxonomy seems to do very little to help us understand these types of music, primarily because the categories bleed into each other so much as to be effectively useless as an analytical framework. This isn't to say that these categories don't represent some of the motives behind songwriting (which they do), but they don't end up working very well as mutually exclusive nodes for understanding genres of music.
That's all fine and dandy, though, because Levitin's book - despite his apparent thesis and his use of them to organize the chapters - isn't actually about those categories at all. Rather, it's a wading through equal parts of alternating personal memories, "I'm friends with Sting!" (or insert other famous musician, but Sting sure does get a lot of airtime as friend), questionable evolutionary psychology, and some useful neuroscience perspectives on music.
As a result, the best review of this book is to simply read some of the wide literature critiquing evolutionary biology. This isn't about denying evolution, at all! Rather, evolutionary biology is a particular approach to storytelling about 'how creatures have gotten X or Y traits' that puts evolution at the front and centre. Think, for instance, of classic examples about why humans have the stress reaction they do: our heart rates accelerate, our breathing speeds up, our digestion slows, all to get ready for evading that lion hidden behind the nearby bush.
The problem, in short, with evolutionary biology is that it becomes remarkably easy to tell whatever 'just so' story that you want. Indeed, this book is basically one 'just so' story after another, in which Levitin explains a multitude of reasons of why a musical brain would have conferred survivability (and, therefore, evolutionary advantage) upon its holders.
Again, this isn't to dispute evolution. Humans absolutely did evolve through processes of natural selection. But, the problem with this kind of retrospective storytelling is that it becomes very easy to justify /any/ current perspective on "what humans are" by creating a story about how that particular attribute conferred some advantage in passing on genes. At its core, it's basically unfalsifiable: there's simply no way to prove or disprove any of these stories; they can just be told ad nauseam for any factor you want to explain. (As an exercise, try this at home: come up with an evolutionary story for why good singers exist, then another for why bad singers exist. Magically, through the power of storytelling, these stories are equally persuasive and unfalsifiable.)
Levitin, of course, shouldn't be held responsible, singlehandedly, for the sins of evolutionary biology. But, there's just something so trite about this book; about the writing style that channels Gladwell and Peterson as 'magical white dude who can explain everything about humans.' There are certainly some nuggets of interesting neuroscience in here, but I can't imagine how a reader would know enough to be able to separate wheat from chaff without finding the book boring (by virtue of already knowing the real stuff).
I also do just have to give a bonus gold start to Levitin, his editor, or whoever decided it was a good idea to write a book with end notes, but include NO NUMBERING to the endnotes in text. There are a whole ton of citations and marginalia included in the notes afterwards, but you, as reader, have zero idea when to go looking for it, as there's simply no indication in text for when these points exist. What a remarkably useless format, likely justified by some harebrained idea that the book 'would be intimidating if it featured superscript numbers in text.'