I would propose that the movement towards a big head in hominids came first with the movements of dance that developed upright posture and a different relationship to gravity and changed the pelvis so babies with larger heads could be born. As for singing, the movement form of the temporalis muscle [from above the eyebrows where it was anchored for a strong bite,] became less necessary as humans used rhythms and resonant sounds to coordinate movement in hunting and better teamwork. [The dance helped with that as well.] Rhythm creates coordination in groups both in games and music. Many animals have a type of music and rhythm but don't speak. Why shouldn't humans? As better learning groups evolved through team hunting, better hunting teams meant better diets and evolving bodies and psychologies. Still the brain was blocked by the thick bone where the pre-temporalis was anchored at the front of the forehead. As skills advanced there was less of a need to bite and more of a need to communicate. More sophisticated rhythms gave us vowels, long sounds, and consonants (articulation). The need to fight with the mouth receded and the temporalis muscle migrated to the present position. The frontal bones of the skull thinned and the brain grew in the frontal lobes. More brain power gave more variety of music and the evolution of language. Many cultures in the world have stories about the world beginning with the song of the Creator. Not just indigenous cultures but the sounding God sounded the world into existence through thought and word. Later he sang to Moses and became the model for the evolution of Jewish music and later, Christian music. For a discussion of this you can refer to

http://www.geoffreyshisler.com/Shallach.html

So the idea of sound being the beginning of modern humanity and the precursor to concept is probably pretty over simple but I don't find it hard to believe that such a supremely rhythmic creature as humans would sing before they form the concept of a word and since sound is the "stuff" of words, what's the big deal? Also I would encourage one to read Edward T. Hall's book called "The Dance of Life" and his discussions of human rhythms and what he learned from extensive research in proximics with modern humans. I would say that concepts in modern humans begin with dance, rhythm and song in early childhood as well. Mithen's observation was documented in the 1980s by Hall and his team. I did, however, enjoy the Mithen book. It's nice to see the idea evolving given the current status of the Arts in American thinking. Meanwhile, what does it mean that we teach music so sparsely and poorly in our schools? We have feral sociopaths documented on Wall Street. Could that be an indicator of an anesthetized culture that has lost its foundations? Walter Bernard and the Positivists were making that argument until Hitler stopped it cold with a war and the destruction, at Auschwitz, of the people making the argument. It's true more than ever, in my opinion. REH

I believe Mithen goes into the why, being that variations in intonation and rhythm help develop emotional cognition through non-verbal elements of communication.

Daniel, From cradle to grave, the working definition I use for all of the arts is from the late composer/pianist Bela Rozsa. He was my teacher and a student himself in the Liszt Academy in Budapest, Juilliard, Phd in music psychology from Iowa, Boulanger in France and Schoenberg in California. He evolved the definition that "music is a psycho-physical pursuit of values in sound." I extend that to all of the perceptual sensorium. It's both specific and generic enough that it opens doors for my classes.

REH

I'm not an expert in this area, but came across an interesting take on this subject when reading the excellent book "The Master and his Emissary" by Iain McGilchrist - which is all about the divided brain and how this affects individuals and society. He references "the singing neanderthals" and also pulls in other research and his own findings.

In terms of development he thinks that music came before singing which came before speech.

He suggests that although the hemispheres are involved equally in many tasks, the way in which they do them is very different. The right is more expansive, the left very specific. The right hemisphere is more intimately connected with the "real" world - and emotional connections, the left is more responsible for pigeon holing things as specifics - for use - making mental maps and models. The left is primarily responsible for vocabulary, and the right for meaning. People with damaged left hemispheres often have trouble remembering words for speech, but often can use the very same words in song.

In music, individual notes mean almost nothing, it is the relationships between notes in the overall context that matters to get the emotional connection. And this is the realm of the right hemisphere. If you have a right hemisphere stroke, you can get "amusia" - an inability to understand or perform music. The left hemisphere does contribute to musical understanding - it gets simple rhythm. But the right comprehends syncopation, and complex, off beat patterns.

In infants, the corpus colossum that links the hemispheres is 10 years away from being fully developed, so the brain is getting itself organised. But the right hemisphere is thought to be much more active in newborns. Because to learn, we need to have things recorded in the left brain with which to compare new information gathered by the right. There is no vocab. Very early development is about close connections / emotions / love etc. This is the right hemisphere's patch - as is the overall understanding of musical meaning.

So, I think you can relate infant directed speech as musical, because you are connecting primarily with the right hemisphere of the infant. I would say that part of the emotional connection between the you and the infant is empathy. And empathy, along with most higher order musical meaning is subserved only by the right hemisphere - according to the 20 years worth of research summarised in this book. Which is well worth a read - it is on Kindle. I'll attach a taster, although it doesn't mention music specifically ..

http://vimeo.com/31780637

According to my personal view and also from the studies of Human Development by Piaget's theory and Hurlock and others, It was understood that singing comes first during the developmental Tasks of Babyhood period. BABBLING is a way of singing of Rhythmic sounds of vowels later stages the vocabulary and speaking developed in Human Development. I am not an expert in the field of your topic or subject.

Louis Lucas, in his "L'acoustique nouvelle," held that imitation of natural sounds comes first, with the vocal mimicking of birdsong, animal noises, and the sounds of matter in nature. Spanish composer Manuel de Falla and his student Spanish poet Federico Garc'ia Lorca applied Lucas's opinion to "deep song' (cante jondo), ancient southern Spanish folk music heard among the gypsies with their musical trills and hand-clapping. Falla's teacher Felipe Pedrell maintained that lullabies, with their primitive melodies, should receive deep study by folklorists.

Music (and singing) was created 500,000 years ago while speech and language were developed 200,000 years ago; according to some researchers, language evolved from music: http://www.theguardian.com/education/2014/feb/27/musicians-better-language-learners

Regarding the reasons of this sequence (and consequence) singing - language: probably it's because singing (and music) is based on imitating nature's sounds and patterns, animals, birds while language needed time to evolve as a distinct system of communication.

This is an interesting and challenging question. Probably, as most questions, it depends on how one defines specific terms, especially language and singing. And I guess, it can't be answered, as we have to speculate.

As I understand it, language and music have common roots, common aspects, such as timbre, rhythm, melody, dynamics etc. If we take a "hmm" as a sound and add pitch and duration and dynamics, it can express very different meanings and could as such already qualify as language. If we have an angry hmmm, it would mean don't (continue what you started or intended to do). In a questioning melody (upwards) it could mean, what do you want. So from this perspective, I would say, language comes first, but already contains elements of singing. Another angle to be considered would be to watch what one is doing when they hold a baby for the first time. Do they sing? Or do they talk to it?

I think that you should start with a definition. Since we can name it - singing when should call just speaking.

Like some of the other respondents have said, the question posed is too broad and it doesn't indicate how the questioner himself understands/defines the key terms. In addition, it seems satisfied to limit the scope of the inquiry to a discursive horizon defined by S. Mithen, Steven Brown, Daniel Levitin, Olivier Sacks, etc. If that is so, I'm afraid I cannot join the discussion. I do nonetheless offer  some suggestions on other authorities who have looked into this enigma. I'm referring, inter alia, to Thrasybulos Georgiades, Musik und Rhythmus bei den Griechen, Johannes Lohmann, Mousiké und Logos, Gregory Nagy, Pindar's Homer, Jean Molino, Le singe musicien (etc.). They all present very convincing arguments for the case that, in the Western Tradition, singing (aoidia) predates speaking. They also make the case that the phonetics, semantics and semiotics of ordinary, "idiotic" speech remained dependent on musical arrangments of sound until relatively recently.

Which raises a separate question: when and how did demotic uses of speech "liberate" itself from mousiké? With the advent of "New Attic Dithyramb"? The sophistics of Gorgias of Leontini? The arguments presented by Plato's in The Sophist? ...