A comparative evaluation of current theories of the origin of music (OoM)

1
A comparative evaluation of current theories of
the origin of music (OoM)

• Richard Parncutt
• Centre for Systematic Musicology, University of
  Graz, Austria
• International Summer School on Systematic
  Musicology and Sound and Music Computing (ISSSM),
  Genova, Italia
• 15 March 2014

SysMus Graz
2

• Across cultures and periods, music is social
  (coordinating group behavior, strengthening group
  identity), emotional (evoking basic and everyday
  emotions but especially pleasure, wonder,
  tenderness, nostalgia), and melodic/rhythmic
  (within physical constraints of the human body).
  It is generally accepted that music and speech
  had a common ancestor called musilanguage the
  two may have split between 106 and 105 years
  ago. The split must have occurred because music
  acquired a special non-linguistic function, e.g.
  to facilitate group cohesion (like grooming),
  mate selection (flirting), or cognitive skill
  acquisition (childplay). In other approaches,
  music developed in a continuous line from
  protomusical primate behaviors, or emerged
  accidentally because it was somehow pleasurable.
  Another candidate for music's origin is
  motherese, which emerged some 106 years ago
  because it promoted infant survival when
  gestation became shorter due to increased brain
  size and a narrower pelvis, in turn due to
  bipedalism (Mithen, 2009). Musilanguage and
  motherese may even be the same thing.
• We do not know enough about ancient environments
  and behaviors - let alone ancient music or
  protomusic - to evaluate such theories directly.
  But we can evaluate how well each theory predicts
  the apparently universal social functions,
  emotional qualities, and structural features of
  music as we know it today. That may be the most
  objective way that we have to evaluate and
  compare competing theories.
• In this presentation I will attempt to
  systematically list the main characteristic
  features of music as we know it today and
  evaluate the extent to which each theory predicts
  each feature. A list of such features is
  essentially just a definition of music. A major
  difficulty in defining music has always been to
  distinguish it from speech, with which it shares
  many characteristics. A possible solution is
  first to list the common features of music and
  speech (comparable with a definition of
  musilanguage) and then to list the ways in which
  music and speech differ. Music and speech are
  both acoustic signals, and both are structured
  (gestural, rhythmic, melodic, syntactic) and both
  are social (meaningful, emotional, intentional).
  By comparison to speech, music is less lexical,
  less socially essential, more spiritual, more
  repetitive, more exact in pitch/time, less exact
  in timbre, and more expertise-oriented. A careful
  analysis of the extent to which each theory
  predicts these features does not clearly
  distinguish four leading theories from each other
  (group cohesion, mate selection, skill
  acquisition, motherese) but it does seem to
  eliminate primate behaviors and non-adaptive
  pleasure seeking.

3
Centre for Systematic MusicologyUni Graz, Austria
Bernd Brabec Ethnomusicology
Erica Bisesi Expression in piano music
Hande Saglam Music and Migration
Michaela Schwarz Secretary
Sabrina Sattmann Music psychology
Andreas Fuchs Music technology
4
Contents

• Definition of music
• What are we trying to explain, exactly?
• Overview of current theories
• How music might have begun
• Comparative evalution
• How well does each predict musical features?

5
Language and music universal features
Cecilia Bartoli
Cécile Kyenge
Language and music Music (compared to language)
acoustic meaningful gestural rhythmic melodic syntactic social emotional intentional less lexical less socially essential more spiritual more repetitive more exact in pitch/time, less exact in timbre more expertise-oriented DANCE???
6
Both language and music are 1. Acousticphysical
vibrations of voice, instruments, air, ear...

• Physical limitations
• physiology of voice and ear
• size ? resonances of vocal tract
• f0 range of music 100 1000 Hz

7
Both language and music are 2. MeaningfulThey
communicate information

• language obvious?
• music mysterious?
• philosophical psychological issues

8
Both language and music are3. GesturalBoth
involve body movement (corporality)

• Language
• theories of origin based on sign and gesture
• Music
• dance, conducting, musicians gestures
• talk about music rising melody, fast music

9
Both language and music are 4. Rhythmic and
melodic

• Speech timing intonation prosody
• Music rhythm contour melody

DonJohnson Rice University
10
Both language and music are 5. Syntactic

• Language
• nouns, verbs, subjects etc.
• Music
• motives, scale steps, beats etc.
• Both
• contextual probabilities
• hierarchical structures
• ambiguities (esp. music)

11
Both language and music are 6. Social

• Language and music
• give groups and participants identity
• enable motivate coordinated action
• Different natural constellations
• language pairs
• music groups

12
Both language and music are 7. Emotional

• Emotional communication through prosody (timing,
  pitch, loudness, timbre)

• Music focuses on
• wonder, transcendence, tenderness, nostalgia,
  peacefulness, power, joyful activation, tension,
  sadness
• (Zentner et al, 2008)

13
Both language and music are 8. Intentional

• Planning
• thinking about past and future
• Metacognition, reflection
• thinking about thinking
• Theory of mind
• others minds are different
• ? language lying
• ? music emotional manipulation

14
Differences
Music as compared to language Music as compared to language
1 not lexical
2 less syntactic
3 less socially essential
4 more spiritual
5 more repetitive
6 more exact in pitch/time, less in timbre
7 more expertise-oriented
15
1. Music is not lexical

• Words can be defined in dictionaries.
• Meaning does not depend on sound.
• Musical elements not defined in dictionaries.
  Meaning does depend on sound.

16
2. Music is less syntactic
Language Music
Is syntax integral to meaning? Yes No
Does scrambling destroy message? More Less
17
3. Music is less socially essential

• People with aphasia
• Relatively big social problem
• People with amusia
• Relatively small social problem

18
4. Music is more spiritualtranscendent,
connecting, life-changing

• Music is more ritualised
• special times, special places, special feelings
• Language is more everyday
• any time, any place, any feeling

19
4. Music is more repetitive

• people usually say things only once
• musical themes are usually repeated

I must not talk in class I must not talk in
class I must not talk in class I must not talk in
class I must not talk in class I must not talk in
class I must not talk in class I must not talk in
class
I must not talk in class I must not talk in
class I must not talk in class I must not talk in
class I must not talk in class I must not talk in
class I must not talk in class I must not talk in
class
I must not talk in class I must not talk in
class I must not talk in class I must not talk in
class I must not talk in class I must not talk in
class I must not talk in class I must not talk in
class
20
5. Music is more exact in pitch and time
language, in timbre

• Pitch-time patterns in music
• ? thousands of melodies
• Timbre in language
• ? thousands of words

21
6. Music is more expertise-oriented

• Speaking
• everyone can do it
• everyone practices
• Playing music
• experts are admired
• only experts practice

22
What is a good theory?

• Satisfies general criteria for good theories
• Accounts for many musical features

23
A good theory is...

• simple ? parsimonious, falsifiable
• general ? accounts for range of phenomena
• focused ? on the main issues
• concrete ? clearly defined terms, processes
• logical ? clear argument
• empirical ? observation-based, ecological
• seminal ? inspires new approaches
• Ockham Kuhn
  Popper Gibson

24
A good theory of OoM should explain or be
consistent with as many points as possible
Language and music Music (compared to language)
acoustic meaningful gestural rhythmic melodic syntactic social emotional intentional not lexical less syntactic less socially essential more spiritual more repetitive more exact in pitch/time, less in timbre more expertise-oriented
Music a cultural construct of singing, dancing,
playing instruments?
25
Kinds of theory of OoM

• Prerequisites or behaviors?
• Different prehistoric periods
• One of more origins

26
Prerequisites versus behaviorstwo kinds of
origin

• Prerequisites include
• physiology
• vocal tract, fast neural processing...
• psychology
• reflective language, theory of mind...
• latent abilities
• ability to synchronize to a beat, e.g. Snowball
• Behaviors include
• animal singing
• motherese

27
When did music begin?
Years ago Relevance for humanity Relevance for music
100 000 000 animals acoustic communication
10 000 000 primates sonic social coordination
3 000 000 tools
1 000 000 large brain, social cognition lang.-music separation
300 000 low larynx ? variety of sounds
100 000 reflection, theory of mind speech complex vocab syntax cultural explosion art, religion exit from Africa instruments scales, metre complex ritual
10 000 agriculture, villages, writing hierarchical social structures pentatonic/diatonic rich oral traditions
1 000 religious politics polyphony, notation
100 electrical technologies new tonalities sound recording
28
Origin or origins?

• Did different processes contribute to music in
  different periods?

29
Leading theories of OoM

• Musilanguage
• Emotional coordination
• Non-human animal behaviors
• Non-adaptive pleasure-seeking

30
MusilanguageEmotional-lexical vocal
communicationWere music and language once one?
Does music exist because it has different social
functions from language?

• My assumption
• Yes, obviously
• Not obvious
• Separation was
• functional and deliberate?
• start of consciousness?
• cf. phylogeny and ontogeny

31
Emotional coordinationDoes music exist because
it coordinates group emotion, which promotes
group survival?

• Theoretical foundation
• Music can coordinate emotions of many people
  (social control in ritual)
• Groups with music can be larger compete better
  with other groups
• Evidence
• Language music can fulfil social function of
  grooming (Dunbar)
• Growth of group size with brain size
• Accounts for musics...
• social function, emotion, intentionality
• NOT spirituality, strong emotion

32
Non-human animal behaviorsIs music an
elaboration of ancestral behaviors?

• Examples
• singing (birds, whales, gibbons)
• territorial marking (wolves...)
• synchronous chorusing (chimpanzees)
• For may explain unconscious drives
• Against isolated, impoverished skills
• Can account for musics...
• repetition, social function, emotion
• NOT complexity, intentionality, spirituality

Non-human animals are amazingly bad at music and
language - as the photos suggest
33
Non-adaptive pleasure seekingDoes music exist
because it activates neural reward systems?

• Uses multiple existing neural systems
• motor, ASA, language, social, aggression...
• Counterevidence
• based on accident could go in the wrong
  direction?
• musicians do not get addicted and suffer
  withdrawal!
• Accounts for musics...
• emotion, repetition, spirituality
• NOT gesture, structure, syntax
• NOT social function, intentionality

34
Music and stages of sexual reproduction

1. Flirting
2. Prenatal development
3. Motherese
4. Childhood

35
1. Flirting Mate attractionDoes music exist
because it indicates male reproductive fitness?

• The idea
• Males use music to attract females
• Females use music to judge male fitness
• Evidence (shaky)
• love songs
• musicians are mainly male
• we are more creative when sexually active
• Counterevidence
• no gender diff. in music ability androgeny
• other ways for females to evaluate fitness
• humans are smarter than peacocks
• Accounts for musics...
• emotion, intentionality, expertise orientation -
  NOT spirituality, social glue

36
2. Prenatal developmentDoes music exist because
it evokes the mother schema?

• The idea
• Fetus acquires mother schema that promotes
  postnatal survival and includes sound and
  movement
• ? motherese, play, ritual
• Evidence
• Musical skills of infants and children
• Music as persona (mother perceived by fetus?)
• Musical emotion is strong, spiritual (e.g. awe),
  changed states, enclosure, flow, non-lexical
• Counterevidence
• Hard to test causal relations
• Evidence is circumstantial
• Unfalsifiable ?

37
3. Mother-infant communicationIs music
elaborated motherese?

• Motherese babbling are musical
• prosodic exaggeration
• rhythmic, melodic, gestural
• emotional, meaningful
• Evidence studies on
• motherese
• infant musicality

38
4. Childhood Playing, trainingIs music
elaborated childplay?

• Childplay develops cognitive, social and motor
  skills, promoting survival
• performance, dance ? physical skills,
  coordination
• listening ? cognitive skills, language
• Doubts
• Does this kind of training really promote
  survival?
• Is there a Mozart effect?
• Accounts for musics...
• social functions, pleasureexpertise orientation
• NOT spirituality

39
THEORY/ CRITERION animal behav useless pleasure flirting play skill emotioncoord. mother-infant
acoustic 1 1 1 1 1 1
meaning 1 1 1 1 1 1
gesture 1 1 1 1 1 1
rhy mel 1 1 1 1 1 1
syntax 0.5 0 0.5 1 0.5 1
social 1 0 0.5 1 1 1
emotional 0.5 0.5 1 0.5 1 1
intentional 0 0.5 1 1 1 1
non-lexical 1 1 1 1 1 1
repetitive 1 1 1 1 1 1
spiritual 0 0.5 0 0 0 1
unessential 1 1 1 1 1 1
expertise 0 0 1 1 0 0.5
evidence 0.5 0 0.5 0 1 0.5
TOTAL 2.5 1.5 4.5 4.5 4.5 6.0
40
Conclusion

• Is music...
• 40 motherese
• 20 flirting
• 20 childplay
• 20 emotion coord?

41
Infant musicalitye.g. many studies by Trehub and
collaborators

• Sensitivity to...
• melodic contour
• relative pitch/duration
• specific musical intervals
• changes in unequal scales/rhythms
• pulse (Winkler Phillips-Silver)
• These are predispositions
• evident in infancy, before they have obvious
  utility (Trehub, 2001)

42
Origins of infant musicality

• Genetic (Trehub)
• selection for music (mate attraction, training,
  social glue, motherese)
• Learned (Parncutt)
• prenatal exposure to changing maternal sound,
  movement and hormone levels

43
Prenatal cognitionHuman gestation 40 weeks

• Taste from 15 weeks
• Hearing from 20 weeks
• Light sensitivity from 28 weeks (Lowery, 2002)
• Memory (Hepper, 1991) learning (Hepper, 1996)
• possible functions practice, infant-mother
  bonding, promotion of breastfeeding, language
  acquisition

44
Prenatal hearing in animals

• Yes humans, sheep, goats, guinea pigs
• No gerbils, rats, cats
• Function
• Prenatal bonding?
• Practice for postnatal perception?

45
Prenatal learning

• Fetal rats learn in response to intrauterine
  stimulation administered occasionally over 5
  days (Smotherman Robinson, 1990)
• Fetal humans learn from maternal sound, movement
  ... almost constantly for 20 weeks
• Trivial Why should a fetus be less good at
  learning than a non-human animal?

46
Infant ? fetus?

• Sudden physiological changes at birth
• Breathing
• Digestion
• No psychological or neurological change!
• Birth is a trivial event in development. Nothing
  neurologically interesting happens
• (Janet DiPietro, cited in Psychology Today, 1998)
• From about 32 weeks fetus behaves like newborn
• ? Extrapolate infant behavior to fetal behavior?

47
The mother schema(Parncutt, 2009)

• Infant schema (cuteness) promotes survival
  (Lorenz)
• Inverse Mother schema?
• Fetal knowledge of maternal sounds, movements and
  emotions could promote bonding and infant
  survival
• Multimodal and holistic baby cannot analyse it
• Evidence for mother schema
• very early hearing learning what function?
• fetus learns maternal smell, taste, diet ?
  preferences (Joy Brown, 2008)
• fetus distinguishes maternal voice from
  loudspeaker on mothers abdomen (Hepper et al.,
  2007)
• All of this without reflective awareness

48
The phylogeny of music The mother schema theory
Larger brain 2 to 1.5 million years ago (Falk,
2000 Mithen, 1996) Today, the first 3 postnatal
months are the 4th trimester
49
The ontogeny of musicThe mother schema theory

• Prerequisite Long-term multimodal recognition
  memory
• multimodal sound movement emotion
• for similar, repeated patterns
• not episodic memory! (e.g. no memory for birth)
• Play is like motherese , ritual is like play
• Both evoke motherese feelings
• Both are reinforced by operant conditioning

50
From motherese to music

• It is not surprising that societies all over the
  world have developed these nodes of culture that
  we call ceremonies and rituals, which do for
  their members what mothers naturally do for their
  babies engage their interest, involve them in a
  shared rhythmic pulse, and thereby instill
  feelings of closeness and communion. The inborn
  propensities for imitation, reciprocity, and
  emotional communion in infancy have become
  further elaborated and used in ritualized and
  ceremonial forms that themselves build and
  reinforce feelings of unity among adults, all of
  which ultimately serve to hold the group
  together.
• (Dissanayake, 2000, p. 64, cited by Davies)

51
Musilanguage and motherese

• Hypothesis They are the same thing
• Makes theory more parsimonious, concrete,
  tangible
• No evidence that they are different
• A concrete scenario for OoM

52
Music and religion/spirituality
Why the strong connection In all known human
cultures?
53
Prerequisites for religious belief

• Basic morality
• Reciprocal altruism
• Are gods supernatural police? (Matt Rossano)
• Mental images and causality
• Tool use ? (-2 000 kya)
• Formulation of beliefs
• Language, ability to construct narratives
• Large brain (500 kya)
• Construction of gods Theory of mind, agent
  detection
• Oral tradition Gullibility of children (Dawkins)

54
Monotheistic tendencies in world religions

• Hindu
• Poly- or mono-theistic?
• 3 aspects of universal supreme god Brahma
  (creator), Shiva (destroyer), Vishnu (protector)
• Buddhism
• A- or mono-theistic?
• 3 jewels Buddha, Dharma (teachings) and Sangha
  (community)

Religion Billions of people
Christianity 2
Islam 1.5
Hindu 1
Buddhist 0.5
folk 1
No religion 1
55
Origins of religious emotionAn approach from
anthropology and animal behavior

• the soulful need to pray to gods, to praise God
  with hymns, to shake in terror before the power
  of invisible spirits, to fear for ones life at
  the hands of the unknown or to feel bathed in
  all-enveloping love from the heavens. The
  inaccessibility to language of the sacred
  experience mirrors what Martin Buber writes God
  is wrapped in a cloud but reveals itself, it
  lacks but creates language. We hear no You and
  yet we feel addressed we answercreating,
  thinking, acting with our being we speak the
  basic word, unable to say You with our mouth
• Barbara King (2007). Evolving God A provocative
  view of the origins of religion. New York
  Doubleday.

56
Pre- perinatal mother schema

• multimodal representation
• behavior, states, predictions
• the first schema ever
• - in ontogeny and phylogeny
• mirror of mothers infant schema
• both schemata ? infant survival
• human qualities
• of gods and spirits?

Bharat Mata Mother India national
personification  of India as a mother goddess
57
God according to Wikipedia

• Not only the origin of everything, but also
• supernatural and eternal
• overseer of the universe
• omni-scient, -potent, -present, -benevolent
• divinely simple
• a culturally reconstructed mother schema?
• originally ungendered and unpoliticized
• cf. Jungs mother archetype
• transmitted by non-episodic, pre-linguistic
  memory
• not remembered in the everyday sense
• social and cultural - not individual

58
Chora - ???a

• Ancient / modern Greek
• womb popular dance / country, region
• Plato
• maternal space where forms materialize
• Julia Kristeva
• prelinguistic space
• place of mobility, dissolution, renewal
• source of emotion, instinctive drives
• Jacques Derrida
• feminine roles mother, nurse, receptacle

Canterbury Cathedral
Carnegie Hall
59
Acoustics of churches

• Appropriate for worship
• psychological connection to prenatal
  experience?
• 1. Reinforcement of low frequencies
• Prenatal sound is low-pass filtered
• 2. Suppression of directional cues
• (Criteria for good room acoustics include
  envelopment)
• Fetus cannot localise sound sources

60
Supplication or fetal position?
Islam
Jewish prayer Plaza Hall, Jerusalem youtube
Hinduism
Buddhism
Sikhism
Bahai
61
Origin of consciousnessreflective language ?
reflective awareness

• 100 000 years ago
• No fundamental neural change
• No clear benefits for survival and reproduction
• Scenario
• bigger brain ? earlier birth ? fragile infant ?
  motherese
• Infant survival depended on mother guessing its
  needs
• strong evolutionary pressure!
• empathy, role of mirror neurons
• theory of mind, metarepresentation, consciousness
• parallel linguistic development (motherese)

62
The origin of music in the mother-infant
relationship...maybe even the origin of language
and religion

• Larger brain bipedalism ? earlier birth
• infant fragility atriciality
• more parental care ? survival
• new, complex infant-mother interaction
• Reflective language, arts, culture
• Accounts for musics...
• structures
• functions
• emotions
• spirituality