I just finished the strangest, most disconcerting little book. It's called Why Do People Sing?: Music in Human Evolution by Joseph Jordania.
If the title hasn't already piqued your interest, its thesis surely will. The thesis is wild, bold, and original, but makes an eerie amount of sense. If true, it would be a revolution — and I don't use the term lightly — in how we understand the evolution of music, cooperation, warfare, and even religion.
I have my reservations about Jordania's theory (and his book), but I'll save them for a later time. As Daniel Dennett once wrote about another remarkable theory:
I think first it is very important to understand [the] project, to see a little bit more about what the whole shape of it is, and delay the barrage of nitpicking objections and criticisms until we have seen what the edifice as a whole is. After all, on the face of it, [the project] is preposterous... [but] I take it very seriously.
These are exactly my feelings about Jordania's project. Seemingly preposterous, but worth taking very seriously.
0. Stylized facts
I'm going to share Jordania's theory with you, but first I want to present a set of "stylized facts" — curious, disparate, and nearly inexplicable phenomena that would seem to have little relation to each other. Then I'll present the theory that (uncannily) links them all together and explains everything.
OK, brace yourself. Here come the facts:
- When our ancestors  first moved from the forest to the savannah, we were not yet capable of making tools. But early hominid evolution tended away from a physiology that would have helped us hunt and/or defend ourselves from predators. Our canine teeth receded, we became slower and weaker, and we didn't develop tough skin (in fact the opposite).
- Lion evolution and migration seems to have mirrored early hominid patterns, both spatiotemporally and (in some ways) behaviorally and morphologically. Lions, for example, are the only social species of cat.
- Humans are the only ground-dwelling species that sings. There are over 4000 singing species — mostly birds, but also gibbons, dolphins, whales, and seals. But they all sing from water or the trees. When a bird lands on the ground, it invariably stops singing.
- Of all singing creatures, humans are the only ones who use rhythm.
- When we sing, we almost always dance, even if it's just nodding along or tapping a foot. Both singing and dancing (whether together or separate) are group activities used across the world in tribal bonding rituals. Isolated ethnic groups have remarkably similar styles of song and dance.
- Rhythmic chanting and dancing induce trance states.
- Early hominids quite possibly ate their dead, and (some while later) definitely started burying them. The instinct to preserve a dead human body from mutilation, and then to dispose of it, is fairly universal. E.g. we strive to retrieve corpses even from a battlefield.
I hope you are intrigued. Each of these facts is hard to explain even in isolation. So a theory that can unify and account for all of them will have to be either profound or crazy — or both.
At this point I'm going to present Jordania's theory as clearly and comprehensively as I can. I'll interpolate a bit and add my own explanatory flare, but the ideas come straight out of his book.
1. Hunters or Scavengers?
When human ancestors first descended from the trees and stepped out onto the grasslands, they faced two critical problems: acquiring food and defending themselves from predators. We'll discuss food acquisition in this section and defense in the next section, but as you'll see they're linked by a similar mechanism.
I hadn't thought deeply about these problems until I read Jordania's book. I always imagined, naively, that early humans had been "hunter-gatherers." While this is true of later humans, it's almost certainly not true of our earliest savannah-dwelling ancestors. Gathering? yes. But hunting, especially big-game hunting, was out of the question. As I mentioned, our earliest ancestors hadn't yet learned how to make or use tools beyond simple rocks and sticks, and we were fairly weak.
Yet we certainly ate meat — the archaeological record is pretty clear on that. So there's a growing consensus that we were actually scavengers (or perhaps "scavenger-gatherers").
Now there are two types of scavenging, two strategies for "carcass acquisition": passive and confrontational. Passive scavenging occurs when the scavenger comes across an undefended carcass. If the carcass was the result of a natural (non-predatory) death, there will be plenty to eat, assuming the scavenger finds it it before putrefaction sets in (about 48 hours). But even if the carcass was killed and eaten by a predator, there's going to be some meat left on the bone. And, just as important, inside the bone. Marrow was an excellent source of calories for our early ancestors, and we were in a unique position to access it, since we could use rocks as primitive tools to break the bones open. The archeological record shows bones that were repeatedly smashed with "hammering stones," as well as bones that had scrape-marks, indicating that we also used sharp stones to carve off some of the clingier meat.
Now, confrontational scavenging is where it gets interesting — and also more circumstantial. Confrontational scavenging, also known as power or aggressive scavenging, happens when an animal (or group of animals) comes across a carcass that's in the process of being eaten, whether by the original predator or another scavenger. In this case, a confrontation ensues, and to the victor go the spoils.
Whether our ancestors practiced confrontational scavenging is hard to determine, given the evidence available to us at this point, but it's not a crazy idea. It happens all the time in the (non-human) animal kingdom, as when a pack of hyenas scares a lion away from its kill.
2. Defense from Predators
Along with finding food, defense from predators is crucial to the survival of any species, and ours presents a puzzle.
We definitely had predators — lions, principally, but also other big cats (jaguars, leopards), as well as hyenas and maybe crocodiles. (Or more accurately, the ancestors of those species, since they've been evolving just as we have. Hyena-ancestors, for example, were much larger than their modern descendants.)
Physically, early humans were quite weak — and getting weaker with each generation. As I mentioned earlier, our canines were receding and our skin was growing softer. And we were getting bigger, it's true, but not stronger. Our size developed in service of bipedalism, which made us both weaker and slower(!). A chimpanzee, for instance, is roughly twice as strong as a modern human, and can run just as fast if not faster. In the general case, bipedalism results in slower sprinting speeds, even if it increases efficiency (allowing us to walk/run for longer stretches).
Across the animal kingdom there are various strategies for evading or defending against predators, known in the literature as antipredator adaptations. Each adaptation defends against one of the four stages of predation: detection, attack, capture, and consumption. To evade detection, for example, most species pursue a strategy of crypsis, aka hiding. To evade capture, species rely on speed, burrowing, climbing into trees, etc. And to evade consumption, species develop physical defenses like claws, fangs, horns, shells (or at least a tough hide), venom, etc.
But early humans weren't doing any of these things. Physiologically, we weren't getting stronger or faster, or developing sharper claws or teeth or tougher skin. Nor were we trying to hide. Habitual bipedalism has its benefits, but going unnoticed in the grass certainly isn't one of them.
In fact, we seem to have evolved to become more noticeable, more conspicuous. We grew taller, we sang and made noise (the only animal who sings from the ground), we painted our bodies, and we developed strong body odor. 
The strategy of being conspicuous is known as aposematism: apo (away from) + sematism (signalling) = counter-signalling. Aposematism is an antipredator adaptation which prevents an attack from happening in the first place, by making it easier for the predator to recognize you as unprofitable.
Species who counter-signal (aposematically) abound in nature, but they all have something up their sleeves, so to speak. Being conspicuous is a viable strategy only when you're advertising an otherwise invisible weapon. Consider the following aposematic species from across the animal kingdom:
- skunks and zorillas
- bees, wasps, hornets, and velvet ants
- black widows and redback spiders
- poison dart frogs and fire salamanders
- all manner of snakes, e.g. the coral snake
- flamboyant cuttlefish
Each of these species has a powerful defensive weapon — often some kind of venom/poison, but also noxious odors. And most species advertise this with distinctive coloration (bright and high-contrast), but some use other signals. Rattlesnakes and bees use sound, for example, and many species, including zorillas, 'mark their territory' using scents. 
Aposematism (being conspicuous) doesn't help an organism in a single encounter with a predator; in fact it's a liability. But it helps with repeated encounters. When a predator gets sprayed by a skunk or stung by a bee, it quickly learns to avoid future encounters with all members of the species.
All of this raises a most important question: how did early humans manage to get away with aposematism? What defensive weapons did we have up our sleeves?
The answer probably won't surprise you: we used stones. It's how we used them (section 4) that's so unusual.
Stones were the most fundamental weapon at our disposal — the only means we had of causing actual physical damage to another animal.
When we first started living on the savannah, we hadn't yet developed the hand/eye coordination necessary to hunt with stones (or to make and use other tools for that purpose). But self-defense is a much simpler proposition, for a number of reasons.
First, using stones to hunt requires a lot of force and a lot of dexterity. You need to hit a target that's moving away from you at high speed, and gravity is not on your side. But defense is different. Defense happens up close. It's much easier to bash a lion with a large rock when the lion is coming toward you. And bringing the rock down allows gravity to work for you rather than against you.
Moreover, attack and defense are fundamentally asymmetrical. Like in a lawsuit, where the prosecution assumes the burden of proof, in a hunt the predator assumes the burden of actually killing its prey. But to succeed at defense, you don't need to kill your would-be predator — you only need to injure it. If a lion or other competitor gets injured while attacking a human, it will flee the scene, opting to lose a meal rather than risk its life by continuing to fight.
And finally, defense was easier for early humans because we helped each other. In most of these circumstances (while being attacked and during confrontational scavenging raids), we had group members by our sides, ready to throw their own stones if others in the group became incapacitated.
Cooperation was indeed our biggest advantage, and our early ancestors refined it, quite literally, to a high art.
4. Audio-Visual Intimidation Display
Here, finally, is the crux of Jordania's thesis. His claim is that early humans developed a unique defense he calls the Audio-Visual Intimidation Display:
My suggestion is that our ancestors turned loud singing into a central element of their defence system against predators. They started using loud, rhythmic singing and shouting accompanied by vigorous, threatening body movements and object throwing to defend themselves from predators. 
Before we go any further, I'd like you to see the AVID in action. Amazingly, some 'primitive' ethnic groups have maintained an AVID-like tradition into modern times. The kailao war dance of the Wallisian people and the Samoan cibi are two examples, but the best-documented tradition is the haka of the New Zealand Māori.
Wikipedia describes the haka as
a traditional ancestral war cry, dance or challenge from the Māori people of New Zealand. It is a posture dance performed by a group, with vigorous movements and stamping of the feet with rhythmically shouted accompaniment....
Various actions are employed in the course of a performance, including facial contortions such as showing the whites of the eyes and the poking out of the tongue, and a wide variety of vigorous body actions such as slapping the hands against the body and stamping of the feet. As well as chanted words, a variety of cries and grunts are used.
For a demonstration we now turn to YouTube. Here's a video of a haka by the New Zealand army, showing just how visually and auditorily intimidating it could be:
And here's a more traditional version:
The point of the AVID (of which the haka is the best-known example) is:
- To intimidate competitors with loud sounds. Shouting in unison is clearly the best way to do this, because the result is far louder than any sound an individual human can make. But clapping, slapping, stomping, and banging rocks together would have added nicely to the cacophony.
- To intimidate competitors visually. The synchronized movements, body paint, and bizarre gestures (including facial expressions) would confuse and intimidate even the fiercest competitors.
- To project the image of a single, powerful organism. As Jordania puts it, "rhythmically well-organized group vocalizations [and movements] send a strong message to the predator about the unity and determination of the group." Although the group is composed of individual humans, during an AVID the group acts as if it were a single, multiheaded, many-armed creature. In some ways it's a fiction, but as far as the competitor is concerned, if the movements and actions are sufficiently coordinated, the group becomes a unitary beast for all intents and purposes.
Remember, the AVID would have been used for confrontational scavenging as well as defense against predators — and would have included rocks as physical weapons to fall back on when the intimidation display wasn't perfectly effective as a non-contact defense.
5. Battle Trance
Jordania also suggests that humans evolved a specific state of consciousness — the battle trance — to go along with the Audio-Visual Intimidation Display. This state has several characteristics:
- Analgesia and aphobia — no pain and no fear.
- Neglect of individual survival instincts.
- Loss of individual identity and acquisition of a collective identity.
Central to the battle trance is the notion of collective or group identity:
Like in a well-established combat unit, where in the heat of the battle one can sacrifice his own life to save a friend's life, human ancestors developed the feel of group identity. The feel of group identity is based on the total trust and dedication of each member of the group to the common interest. Group identity kicks in when there is a critical situation, a mortal danger for survival of the group or any of its members. In such moments, the noble principle of 'One for all, all for one' rules any individual self-preserving instinct, fear and pain.
This state of consciousness may have originally evolved for parents (especially mothers) to defend their children when threatened, but it was repurposed for group defense and confrontational scavenging. And in the process, we developed a new trigger for it: rhythmic chanting and synchronized body movements.
Evolution supplied powerful neurological mechanisms to make this feeling a positive experience. Going into group identity brings the most exhilarating feelings to every member of the group. Every member of the group feels bigger, feel stronger, and virtually feels immortal.... Group members in such an altered state of mind, when they share total trust with each other, emotionally believe that the group cannot be defeated. This unique altered state of mind is supported (and most likely caused) by the powerful neurological substances such as endorphins and oxytocin, which are momentarily released in the brain when a critical survival situations arises. As the neurological substances are released into the brain, feelings of pain and fear are blocked, and total trust and exhilaration of being part of a supernaturally strong unit becomes overwhelming.
Actual combat isn't as central to our lives in the 21st century, but the battle trance and feelings of collective identity still echo in many of our modern rituals of solidarity, which I wrote about last year, and they're particularly pronounced in religion/politics and team/spectator sports.
6. Cannibalism (and burial) of the dead
As if Jordania's theory didn't cover enough ground already, he has one last surprise in store for us. He claims that early humans practiced cannibalism of their dead as a key part of their comprehensive antipredator strategy.
If you recall from section 2, the goal of aposematism is to advertise that, as a piece of prey, you are decidedly unprofitable for the predator. If a predator can easily recognize you (and other members of your species), and remembers getting burned during past encounters, it will quickly learn to stop attacking you in the first place.
Given this strategy, it's very important not to let the lions (or any other predator) get away with killing and eating a human. The more our ancestors were able to reinforce the message that humans are not a (good) meal, the safer they would be across repeated encounters. This was especially important for early humans because, unlike an actually poisonous species, human meat is worth eating, if a predator can get away with it.
This implies a heavy selection pressure for the following behaviors among our ancestors:
- If a predator attacks during a confrontation, make sure it gets injured.
- If a predator manages to kill one of your fellow humans, don't let it eat. Retrieve the body of your fallen comrade or your whole tribe will be in danger.
- When a human dies naturally, make sure the corpse is properly disposed of. The corpse is a liability because a predator will associate even a scavenged meal (of human meat) with 'profitability'.
None of this implies that our ancestors had to practice cannibalism of their dead. Any means of keeping human meat out of the mouths of predators would have been effective. But there's some evidence that suggests cannibalism (stone scrapings on human bones similar to the scrapings on the bones of other animals), and if calories were hard to come by, it might have been an ecological necessity.
There's so much more to say about this theory, but I'll save most of it for another time.
I'd just like to end by showing how some our beliefs and behaviors take on new significance in light of Jordania's theory, especially those that relate to how we handle the bodies of our dead.
Funerary traditions vary widely around the world, but all have one thing in common: disposal of the body. Mechanisms include burial, entombment, mummification, burial at sea, sky burial , and ritual cannibalism, and even more exotic mechanisms like hanging coffins or tree burial. The common reasons given for disposal practices are all public-health-related, but intentional burial is at least 225,000 years old. Of course our ancestors wouldn't want a corpse rotting in their camp, but there's quite a leap from disposal to burial. Why not just drag the corpse away from camp and expose it to the elements?
Jordania's theory doesn't predict how exactly we should dispose of our dead, but it predicts that we should care an awful lot about it (i.e. that it should be something sacred), and that we should be especially concerned that the body doesn't fall into the wrong hands. In Paleolithic times, this ensured that our ancestors retrieved the bodies of their comrades when they were killed by predators. But you can see vestiges of this in historic times — e.g. in our concern for salvaging bodies of the war dead. Mutilating or otherwise desecrating the war dead is an ancient practice, a ghastly way for the victor to show utter dominance over the loser.
And finally, Jordania's theory helps explain the the religious nature of our funerary practices. Burial has always been a quintessentially religious practice. For example, we date the earliest religious behavior in our ancestors by when they started burying their dead. But religions are fundamentally about the living — a set of beliefs and practices that relate to collective identity and tribal cohesion.
Why do religions care about the disposal of corpses? This has always puzzled me. It's always seemed like such a mundane concern. When someone dies, that should be the end of what we care about, and removing the body should be no more sacred than taking out the trash. Chimpanzees, for instance, can perceive when another chimp passes away (and mourn), but they soon lose interest in the body.
But if Jordania is right, it's no coincidence that death rituals are intimately bound up with collective identity, because they're two parts of the same system.
Update 2015/10/16. For the record, I don't think Jordania's theory explains (or even tries to explain) the lyrical aspects of music — only rhythm. But it's the only plausible account I've read that explains how music may have helped us survive. There's always the mate-selection hypothesis, of course.
 human ancestors. I'm going to write about "human ancestors" and "early humans" — or sometimes simply "us" — knowing that there are more precise terms to describe the different stages (and branches) of our evolutionary path. Please forgive me — I'm not particularly steeped in the distinctions, and I doubt many of my readers are either.
 singing, body painting, body odor. It's unclear whether early humans actually sang habitually, painted their bodies, or had B.O., but modern humans certainly do, and there's a case to be made for our ancestors as well. Red ochre is noted as far back as 100,000 years ago, but earlier humans could easily have used simpler preparations like blood or berries, which would have left no trace in the archeological record.
 aposematism. A particularly interesting example is stotting — when a gazelle springs into the air by lifting all four feet off the ground simultaneously. Stotting is an aposematic display because (1) it makes the gazelle more visible, but (2) it's an honest signal of unprofitability. By stotting, a gazelle signals to a potential predator, "I'm in peak physical condition. Don't both chasing me, because I can easily outrun you."
 Audio-Visual Intimidation Display. I'm taking a couple liberties here. For one, Jordania calls it an "intimidating" display, but I prefer the noun form. But more substantively, Jordania lumps our morphological changes (bipedalism, hair, etc.) into the AVID, whereas I'm presenting the AVID as just the behavioral part (rhythmic chanting and dancing).
 sky burial. Yes, during a sky burial we let another animal eat our corpses, but they are always scavenging birds (e.g. vultures), never a potential predator. See also Dakhma for the Zoroastrian tradition.