Davies 2012 Monists
Davies 2012 Monists
Davies 2012 Monists
net/publication/271044767
On Defining Music
CITATIONS READS
13 31,981
1 author:
Stephen Davies
University of Auckland
176 PUBLICATIONS 2,074 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Stephen Davies on 04 August 2015.
1
inherent
in
the
task,
I
do
not
offer
a
counter
definition.
Instead,
I
consider
four
strategies
that
might
be
adopted
by
the
would-‐be
definer.
I
argue
that
none
of
these
alone
can
succeed,
though,
in
combination,
they
might.
2
usually
understood
as
absorbing
all
the
sounds
that
take
place
in
the
performance
period
as
its
contents,
so
that
none
count
as
ambient.
In
that
case,
Cage's
piece
fails
the
necessary
condition
for
music
that
I
have
proposed.
I
conclude
that,
as
an
artwork,
4'33"
is
a
performance
piece,
a
theatrical
work
if
you
will,
about
the
performance
of
music,
not
a
musical
work
as
such.
The
main
difficulty
with
Levinson's
definition,
I
think,
is
that
music
does
not
and
is
not
always
intended
to
call
attention
to
itself.
I
think
"Happy
Birthday"
is
music,
but
I
doubt
that
its
composer
intended
either
that
it
enrich
the
experience
of
its
singers
or
that
it
be
regarded
primarily
as
sounds.
Many
baroque
composers
wrote
Tafelmusik
(table
music)
–
that
is,
music
to
provide
a
background
accompaniment
to
their
employers'
meals
–
and
it
is
possible
that
this
was
sometimes
not
even
registered
as
music
by
those
engaged
in
eating
and
conversation.
A
composer
who
wrote
table
music
to
draw
attention
to
itself
simply
did
not
understand
such
music's
function,
which
is
to
be
self-‐effacing.
While
we
might
dismiss
the
claim
to
music-‐status
of
some
forms
of
Muzak,
it
seems
unduly
harsh
to
demote
Telemann's
Servizio
di
tavola
on
the
grounds
of
its
aptness
for
its
purpose.
Similarly,
nowadays
much
film
music
is
intended
only
for
liminal,
perhaps
unconscious,
awareness
on
the
part
of
the
engrossed
viewer.
And
composers
of
the
themes
that
introduce
much-‐repeated
TV
shows
probably
do
not
intend
them
to
be
attended
to.
I
think
Levinson
is
too
focused
on
Western
concert
music,
even
if
his
definition
is,
as
he
says,
intended
to
apply
cross
culturally.
Admittedly,
Levinson's
account
leaves
a
space
between
art
music
and
merely
musical
sounds,
and
he
might
allow
that
this
gap
may
be
occupied
by
something
quite
properly
called
"music"
despite
its
lack
of
pretension
to
art
status.
In
that
case,
however,
the
definition
excludes
the
greater
part
of
what
is
correctly
identified
as
music
and
leaves
open
the
question
of
what
art
and
non-‐
art
music
share
and
in
virtue
of
which
they
are
music.
Kania
(2011)
is
also
troubled
by
the
intended-‐aesthetic-‐experience
component
of
Levinson's
definition.
As
a
counterexample,
he
offers
the
practicing
musician,
who
makes
music
not
intended
for
an
audience.
To
dodge
the
obvious
reply,
we
might
change
the
example
to
the
doodling
jazz
improviser,
so
that
the
performer
is
also
the
composer.
Kania
also
notes
the
range
of
non-‐
3
aesthetic
functions
to
which
music
can
be
put,
such
as
waking
up
a
friend,
and
defends
Muzak's
status
as
music,
albeit
as
bad
music.
Kania
proposes
his
own
definition:
"Music
is
(1)
any
event
intentionally
produced
or
organized
(2)
to
be
heard,
and
(3)
either
(a)
to
have
some
basic
musical
features,
such
as
pitch
or
rhythm,
or
(b)
to
be
listened
to
for
such
features"
(2011:12,
also
2010:348).
The
disjunctive
third
condition
is
there
to
allow
for
avant-‐garde
works
that
lack
musical
features
but
that
are
intended
to
be
approached
as
if
they
are
music.
His
example
is
Yoko
Ono's
Toilet
Piece,
the
sound
of
a
flushing
toilet
on
the
album
Fly
(1971),
which
Kania
regards
as
music
because
he
thinks
Ono
intended
that
it
be
heard
as
such,
as
is
apparent
from
its
inclusion
on
an
album
of
otherwise
uncontentious
examples
of
music.
In
addition,
the
(b)-‐clause
distinguishes
unmusical
avant-‐garde
music
from
non-‐musical
sound
art,
even
where
they
might
sound
the
same,
because
the
latter
is
not
projected
for
an
"as
if
music"
regard.
Kania
(2010)
classes
John
Cage's
4'33"
in
the
category
of
sound
art,
not
music.
But
a
genuinely
silent
piece
can
qualify
as
music
if
it
meets
the
(b)-‐clause.
Kania
offers
his
own
composition,
Composition
2009
#3,
as
an
example:
"The
score
reads
as
follows:
'Indicate
a
length
of
silence,
using
the
usual
cues
with
which
you
would
signal
the
beginning
&
end
of
a
single
movement,
song,
etc.
(The
content
of
this
work
is
the
silence
you
frame,
not
any
ambient
noise.)'"
(2010:351).
Though
Levinson,
Kania,
and
myself
have
no
ideologically
motivated
desire
to
exclude
avant-‐garde
and
experimental
works
from
the
realm
of
music,
we
turn
out
to
take
different
views
on
Cage's
4'33":
Levinson
has
it
as
music,
Kania
as
sound
art
but
not
music,
and
I
rate
it
as
a
non-‐musical
work
of
theater.4
It
would
be
a
mistake,
however,
to
judge
the
accounts
solely
in
terms
of
how
they
4
I
currently
lean
toward
Kania's
position:
Cage
intended
the
audience
to
attend
to
the
sounds
more
than
to
the
ritual
signs
of
performance
and
he
did
not
want
the
audience
to
try
to
hear
the
sounds
as
if
they
are
music.
Moreover,
Kania
can
appeal
to
the
conventions
of
sound
art
(as
readily
as
Levinson
as
I
might
appeal
to
those
for
musical
performance)
to
explain
both
how
the
performance
setting
is
necessary
to
prime
the
audience's
attention
and
to
establish
the
work's
boundaries.
4
handle
controversial
borderline
cases
like
4'33",
Toilet
Piece,
or
Composition
2009
#3.
Previously
I
objected
to
Levinson's
definition
by
reference
to
what
I
take
to
be
uncontroversially
mundane
pieces
of
music
that
it
excludes,
such
as
"Happy
Birthday"
and
Telemann's
Servizio
di
tavola.
If
it
is
clearly
to
be
defeated,
Kania's
definition
should
be
treated
similarly.
Whereas
Levinson's
definition
seems
to
me
to
be
too
narrow,
so
I
sought
music
that
it
inappropriately
omits,
Kania's
definition
strikes
me
as
too
broad,
so
the
relevant
counterexample
is
of
nonmusic
that
it
includes.
Morse
code
is
one
such
because
it
plainly
meets
(1),
(2),
and
(3)(a):
it
is
intentionally
produced
to
be
heard
in
terms
of
rhythm.
There
is
a
hint
that
Kania
might
reject
the
counterexample.
He
allows
that
speech
displays
periodicity
but
suggests
that,
in
a
definition
of
music,
"the
term
['rhythm']
would
be
restricted
to
a
division
into
stricter
units
of
time,
such
as
characterized
by
measures
of
two
or
three
beats"
(2011:8).
If
this
implies
that
rhythm
in
music
must
be
structured
in
terms
of
a
regular
(or
any)
meter,
it
is
surely
far
too
strong.
Some
music
employs
irregular,
constantly
changing
meters
and
some
other
music
is
in
free
rhythm.
But
if
the
main
point
is
only
that
the
periodicity
of
speech
is
unlike
the
rhythmic
organization
of
music,
it
will
not
rule
out
Morse
code,
which
is
much
more
tightly
and
systematically
organized
according
to
duration
than
utterance
is.
In
any
case,
I
do
not
think
it
is
difficult
to
come
up
with
many
other
examples
that
we
would
not
count
as
music
though
they
satisfy
Kania's
proposed
definition.
Sirens,
the
chord
that
sounds
when
I
start
my
Macintosh,
the
drumming
of
my
fingers
on
the
desk
as
I
contemplate
what
to
write
next,
the
wind
chimes
that
hang
above
my
porch,
and
tone
languages
satisfy
the
definition
without
being
music
(though
they
may
be
musical).
Kania
can
respond
by
arguing
that
my
examples
count
as
music,
or
alternatively,
that
they
are
not
sufficiently
like
music
to
challenge
his
view.
Obviously
I
have
chosen
examples
intended
to
defeat
both
approaches.
Though
I
allow
that
the
Macintosh
start-‐up
chord
could
be
replicated
in
a
deliberately
composed
musical
work
that
consists
of
nothing
more,
I
doubt
that
the
start-‐up
chord
qualifies
as
such
a
work
or
is
generally
regarded
as
music,
though
it
is
plainly
intended
to
be
heard
as
exhibiting
musical
features.
And
while
the
wind
chime
is
made
with
the
intention
that
it
produce
music-‐like
sounds
when
5
activated,
I
doubt
that
we
regard
the
outcome
as
music.
Kania
applies
the
second
strategy
to
the
case
of
tone
languages.
He
cites
evidence
that
pitch
is
treated
more
freely
in
tone
languages
than
in
most
music,
but
I
doubt
that
that
fact
gets
him
off
the
hook.
On
the
one
hand,
the
person
who
speaks
a
tone
language
certainly
intends
relative
differences
in
pitch
to
be
important,
and
on
the
other
hand,
some
non-‐Western
music,
such
as
Australian
aboriginal
singing,
relies
more
on
slides
and
portamentos
than
on
fixed
pitch
points.
I
hope
that
by
now
some
of
the
difficulties
are
apparent:
across
the
globe
music
takes
many
diverse
forms
and
serves
a
variety
of
functions.
Meanwhile,
we
are
surrounded
by
sonic
phenomena
in
which
we
can
take
a
range
of
interests,
including
listening
to
them
as
if
they
are
music.
Some
of
these
phenomena
are
plainly
musical
though
they
are
not
usually
thought
of
as
music,
birdsong
for
example.
And
many
unmusical
sonic
phenomena,
nevertheless,
share
properties
in
addition
to
that
of
being
audible,
with
music.
And
note,
finally,
that
music
can
incorporate
quotidian
soundmakers
(anvils,
cowbells,
typewriters,
sirens,
etc.)
and
can
be
composed
to
mimic
the
sounds
of
natural
events
(such
as
wind
and
thunder),
mechanical
devices
(trains),
or
the
sounds
of
physical
labor
(forges,
for
instance).
Rather
than
attempting
a
definition
that
will
negotiate
these
traps
and
pitfalls,
I
consider
at
the
broadest
level
how
the
definition
of
music
might
be
approached.
I
distinguish
four
ways
in
which
we
might
try
to
define
music
–
functionally,
operationally,
historically,
and
structurally
–
and
show
that
none
of
these
alone
can
produce
a
definition.
I
suggest,
however,
that
some
combination
of
the
characterization
of
music
in
terms
of
its
structure
and
its
traditions
holds
out
the
hope
of
definitional
success.
6
these
functions
are
directed
to
the
self
(self-‐expression,
for
example),
sometimes
to
intimates
(lullabies,
love
songs),
and
sometimes
to
the
group
(national
anthems)
or
even
to
the
enemy
(war
chants).
It
is
worth
reminding
ourselves
of
the
many
functions
music
is
employed
to
serve.
It
is
the
target
of
aesthetic
contemplation
pursued
for
its
own
sake,
or
it
is
a
pleasant
pastime.
Alternatively,
it
supplies
a
background
"wallpaper"
for
other
activities,
such
as
dining
or
shopping.
It
features
in
courtship.
It
functions
variously
to
regulate
mood
and
emotions,
sometimes
cheering
up,
sometimes
calming
down,
and
sometimes
ramping
up
anger
and
aggression.
It
is
used
to
induce
and
sustain
trance,
dissociation,
or
inward
focus,
in
other
words,
to
cocoon
the
individual
in
a
wall
of
sound
that
excludes
others
and
the
world.
And
it
is
employed
to
stimulate
social
involvement,
other-‐directedness,
and
shared
identification.
It
serves
to
coordinate,
entrain,
and
synchronize
the
group's
shared
attention,
emotions,
dancing,
marching,
or
laboring.
It
is
used
both
in
the
celebration
of
achievement
and
in
the
mourning
of
loss.
More
generally,
it
intensifies
public
ritual
–
consider
hymn
singing,
for
instance
–
and
augments
the
significance
and
effects
of
dramas
and
entertainments.
It
labels
or
symbolizes
various
types
of
social
event:
consider
fanfares,
Reveille,
the
Last
Post,
wedding
marches,
hoedowns.
It
is
a
badge
of
social
status
(grand
opera)
or
group
membership
(Dixie!).
It
plays
a
central
role
for
adolescents
in
identify
formation
and
more
generally
is
treated
as
a
significant
measure
of
each
person's
individuality
according
to
their
distinctive
musical
preferences
and
tastes.
It
is
an
offering
to
gods
and
a
mirror
of
heavenly
beauty,
harmony,
or
passion.
It
has
been
treated
as
a
sub-‐branch
of
mathematics
and
as
an
analog
of
astronomical
order.
The
Greeks
and
early
Christians
noted
its
long-‐term
effects
on
human
character
and
its
modeling
of
virtuous
dispositions.
It
is
adopted
for
its
therapeutic
and
medicinal
benefits.
It
is
given
mnemonic
functions:
an
especially
graphic
case
is
that
of
the
long
song
cycles
of
Australian
aborigines
that
not
only
relate
mythical
history
but
trace
a
map
from
waterhole
to
waterhole
across
the
7
deserts.
It
is
a
form
of
communication,
as
in
the
"talking
drums"
of
African
fame.
And
it
is
a
commodity
that
can
be
sold
and
traded.5
There
appear
to
be
few
limits
to
the
functions
music
has
been
put
to
and
I
doubt
that
any
single
function
stands
out
as
the
primary
one.
Moreover,
as
I
noted
before,
music
serves
apparently
opposed
functions:
to
pacify
and
to
incite,
to
exclude
others
and
to
bring
the
group
together,
to
mark
individuality
and
to
cement
the
collective's
shared
identity.
In
fact,
music
plays
a
central
role
in
all
aspects
of
life
–
work
and
ease,
self-‐amusement
and
group
enterprises,
informal
events
and
structured
ritual
–
and
at
all
social
levels
–
private
and
public,
personal,
family,
village,
and
state.
So
much
so,
that
it
is
not
a
candidate
for
functional
definition.
8
processes
and
circuits
that
music
draws
on
must
be
the
result
of
a
complex
interaction
of
environmental
musical
inputs,
already
existing
brain
structures
that
serve
non-‐musical
aural
functions,
and
physical
or
other
limitations
on
the
brain's
plasticity.
The
complexity
of
this
picture
makes
it
impossible
to
distinguish
evolved
and
innate
neural
structures
dedicated
to
music
from
ones
that
are
created
through
exposure
and
training
in
music
(McDermott
and
Hauser
2005,
Patel
2008:402–11).
And
to
complicate
matters
yet
further,
modular
functioning
need
not
be
subserved
by
specific
neural
circuits
(Barrett
and
Kurzban
2006).
It
is
to
be
expected,
then,
that
there
is
disagreement
over
the
interpretation
of
relevant
data.
Some
psychologists
think
there
are
music-‐
specific
neural
domains
(Brown
2000,
Huron
2003,
Peretz
and
Coltheart
2003,
Levitin
2006),
others
do
not
(Ball
2010,
and
with
minor
qualifications,
Patel
2008),
and
some
of
the
most
detailed
reviews
regard
the
evidence
as
indecisive
(Justus
and
Hutsler
2005,
McDermott
and
Hauser
2005).
It
certainly
is
true
that
music
lights
up
areas
all
over
the
brain
–
Ball
calls
it
"a
gymnasium
for
the
mind"
(2010:241)
–
many
of
which
are
not
music-‐specific.
(We
might
have
guessed
this
from
the
earlier
list
of
the
multitude
of
functions
music
can
serve.)
Equally
contested
are
claims
about
the
origins
and
evolutionary
functions
of
music
behaviors.
Many
argue
that
music
is
adaptive
(for
reviews,
see
Huron
2003,
Cross
2007),
though
there
is
significant
disagreement
over
what
it
is
alleged
to
be
an
adaptation
for.
Others
suggest
that
it
is
a
by-‐product
(Pinker
1999)
or
a
technology
(Patel
2008,
2010).
(For
a
critical
overview
of
these
positions,
see
Davies
forthcoming.)
To
illustrate
the
problem,
consider
music
in
relation
to
language.
There
is
a
significant
overlap
in
the
neural
areas
that
process
music
and
language
(Fitch
2005,
Justus
and
Hutsler
2005,
Koelsch
and
Siebel
2005,
Ball
2010)
and
there
are
many
parallels
between
the
way
the
brain
processes
both
language
and
music
(Fenk-‐Oczlon
and
Fenk
2009-‐2010).
But
there
is
disagreement
over
which
came
first.
The
view
that
music
originated
in
pre-‐linguistic
emotional
vocalizations
goes
back
to
the
early
days
of
evolutionary
theory
(Darwin
1880,
Pt.
3,
Ch.
19:572,
Grosse
1897)
and
has
many
modern
adherents
(for
instance,
Brown
2000,
Mithen
2005,
Fenk-‐Oczlon
and
Fenk
2009-‐2010).
Equally
venerable,
9
however,
is
the
argument
that
music
is
a
by-‐product
that
came
as
a
bonus
out
of
the
evolution
of
language
(Spencer
1966,
Vol.
14
[1857],
Pinker
1999,
De
Smedt
and
De
Cruz
2010).
Given
uncertainty
about
the
timing
of
music's
and
language's
respective
origins,
and
about
the
relation
if
any
between
these
historical
processes,
it
is
not
clear
which
could
claim
priority
over
or
ownership
of
neural
structures
they
share
in
common.
I
am
not
opposed
to
a
scientific
or
operational
approach
to
the
definition
of
music.
I
doubt,
however,
that
there
is
sufficient
clarity
over
matters
that
would
have
to
be
settled
were
such
a
definition
to
be
appealing.
At
this
time,
music
is
not
a
candidate
for
a
plausible
operational
definition.
10
needs
fleshing
out.
Emulate
or
resemble
their
style,
their
artistic
achievement,
their
use
of
the
cor
anglais,
their
expressive
effects,
or
their
structural
principles?
Possibly
all
of
these,
though
clearly
there
must
be
some
resemblances
not
relevant
to
the
conferral
of
music-‐status.
In
any
case,
I
think
the
stress
on
works
here
gets
the
emphasis
wrong.
A
musical
tradition
might
have
no
works,
only
free
improvisation.
And
where
there
are
works,
the
primary
intention
is
likely
to
be
to
create
new
ones
that
are
to
be
valued
for
their
individuality
as
well
as,
or
instead
of,
their
similarity
to
canonic
models.
A
central
intention,
one
that
ties
the
present
piece
of
music
making
to
the
prior
tradition
that
led
up
to
and
makes
it
possible,
is
to
use
the
idiom
and
resources
of
that
tradition
in
a
fashion
that
adds
something
to
it.
This
is
consistent
with
the
composer's
rebelling
against
or
trying
to
change
some
aspects
of
the
tradition,
even
as
he
relies
on
other
of
its
aspects
to
bring
that
intention
into
effect.
And
we
might
come
to
a
definition
that
gets
nearer
to
the
heart
of
music
by
considering
how
to
characterize
the
idioms
and
resources
that
mark
musical
traditions
as
such.
This
naturally
leads
to
the
next
definitional
approach.
11
something
equivalent
to
both
the
syntactic
rules
and
semantic
principles
that
generate
higher
level
linguistic
meaning:
just
as
phonemes
form
words,
which
form
phrases,
which
form
sentences,
all
with
transformational
rules
that
can
deconstruct
how
each
level
supervenes
on
the
one
below,
so
we
are
after
an
account
of
the
elements
of
music,
such
as
the
pitched
tone,
and
of
the
rules
by
which
these
can
be
combined
as
phrases,
which
are
joined
into
melodies,
which
are
combined
in
expositions,
developments,
and
the
like,
from
which
whole
movements
are
comprised.
Lerdahl
and
Jackendoff
(1983)
have
proposed
a
generative
grammar
for
tonal,
teleologically
organized
music.
Admittedly,
a
great
deal
of
music
is
of
this
kind,
but
not
all.
Some
music
is
cyclic
and
does
not
seek
closure.
Some
music
is
not
tonal;
it
uses
orchestras
of
untuned
instruments
(such
as
some
percussion
bands)
or
deliberately
avoids
tonality.
And
it
is
not
as
if
we
can
dismiss
these
exceptions
as
unconventional
or
marginal
if
they
are
as
easily
recognized
to
be
music
as
are
the
more
frequent
kinds.
Some
might
deny
this;
that
is,
they
might
deny
that
such
"music"
can
be
easily
recognized
as
such.
For
instance,
they
claim
that
what
Anton
Webern
composed
is
not
music.
But
I
think
such
views
are
often
ideologically
motivated
and
that,
if
we
are
seeking
a
global
perspective,
we
should
be
willing
to
accept
as
music
a
great
many
practices
that,
because
of
our
cultural
backgrounds
and
previous
musical
exposure,
do
not
strike
us
as
accessible.
When
I
claimed
we
easily
recognize
foreign
music
as
such,
I
did
not
intend
to
imply
that
we
therefore
enjoy
it
or
appreciate
it
as
such.
Much
highly
sophisticated
art
music,
not
only
in
the
West
but
also
in
other
musical
traditions
is
"difficult".
That
does
not
take
away
from
its
recognizability
as
music,
which
I
think
can
hardly
be
denied
to
Webern
either.
The
fact
that
Lerdahl
and
Jackendoff's
account
of
musical
grammar
is
limited
to
teleological,
tonal
music
is
not
the
only
problem,
however.
There
are
further
crucial
difficulties
faced
by
any
putative
structural
definition.
These
arise
not
only
at
the
level
of
identifying
the
atomic
units
of
music
but
also
sometimes
in
identifying
transformational
rules
that
transpose
one
hierarchical
level
to
the
next.
Music
need
not
contain
pitched
notes
or
tones
of
fixed
pitch,
and
it
frequently
contains
a
great
deal
of
silence
and
noise.
Non-‐musical
sound
events
–
12
sirens
and
digitized
phone
numbers,
for
example
–
often
do
contain
patterns
generated
from
notes
of
fixed
pitch.
So,
it
is
not
at
all
clear
that
what
characterizes
music
as
such
is
an
exclusive
set
of
sound
elements.
Equally,
it
is
not
clear
that
music's
generative
rules
go
beyond
the
statistical
regularities
of
syntax
to
account
for
content
or
significance
at
higher
levels.
In
other
words,
they
cover
organization
as
sequence
but
not
the
significance
that
characterizes
more
abstract
gestalts.
Consider
melodies,
which
are
often
central
units
of
higher-‐level
musical
content.
Most
people
think
of
them
as
rhythmically
inflected
patterns
of
pitched
tones
(or,
more
correctly,
intervals)
ordered
as
to
sequence
and
as
to
relative
stability
or
tension.
But
the
fact
is
that
the
identity
of
melodies
can
survive
alterations
in
their
interval
sequences,
and
this
occurs
not
only
when
filigree
decoration
is
added
or
structurally
weak
notes
are
elided
but
also,
for
example,
in
transpositions
from
the
major
to
the
minor.
To
follow
minor
key
sonata
form
movements,
one
must
hear
that
the
minor
key
statement
of
the
second
subject
in
the
recapitulation
is
a
restatement
of
the
very
tune
that
was
in
the
major
in
the
exposition,
despite
differences
in
their
interval
sequences.
This
is
a
common
effect.
Melodies
are
often
reconfigured
without
thereby
losing
their
identity:
the
new
statement
is
a
version
of
the
original,
not
a
variation
on
it
nor
a
different
but
related
theme.
So,
the
rules
that
one
might
have
hoped
would
allow
us
to
explain
the
identity
of
specific
melodies
in
terms
of
the
lower-‐level
elements
they
comprise
do
not
exist.7
A
further
difficulty
with
this
approach
concerns
the
fact
that
not
all
music
is
hierarchically
structured
to
the
same
degree.
It
is
possible
to
get
music
that
provides
a
sequence
of
base-‐line
elements
that
cannot
be
synthesized
into
higher-‐order
musical
manifolds.
There
is
an
ordered
array
or
list
of
elements,
but
they
resist
attempts
to
pattern
them.
(Perhaps
there
are
long
silences
and
wide
intervals
between
successive
notes.)
And
it
is
also
possible
to
get
music
without
melodies,
or
without
repetition
of
sections
or
parts,
or
without
multiple
movements.8
7
For
a
more
detailed
discussion,
see
Davies
2001,
2010.
13
In
other
words,
music
does
not
always
display
a
generative
grammar
and
when
it
does
show
the
appearance
of
this
it
turns
out
to
be
virtually
impossible
to
specify
transformational
rules
of
the
kind
the
theory
needs
and
expects.
In
addition
to
these
concerns
there
is
another:
this
approach
tries
to
define
music
as
a
formal
system.
Admittedly,
it
can
allow
that
the
human
purpose
of
music
might
be
to
apply
this
system
in
the
generation
of
humanly
significant,
higher-‐order
musical
content.
But
it
is
likely
there
is
more
to
this
significance
than
can
be
explained
solely
by
reference
to
music's
elements
and
generative
principles,
just
as
there
is
more
to
linguistic
meaning
than
is
covered
by
syntax
and
semantics:
in
both
cases
contextual
considerations
and
wider
pragmatic
conventions
can
be
equally
important.
And
it
seems
that
music's
human
significance
should
figure
in
its
definition.
The
structural
approach
to
music's
definition,
with
its
focus
on
musical
elements
and
how
they
can
be
combined,
tends
to
bypass
the
seemingly
crucial
fact
that
music
is
a
culturally
embedded
human
practice,
as
I
elaborate
below.
I
have
argued
elsewhere
(Davies
2007)
that,
whereas
a
theory
of
art's
human
value
must
take
account
of
its
place
in
human
life,
a
definition
of
art
perhaps
need
not
do
so.
I
do
not
mean
to
renege
on
this
general
point
in
insisting
that
a
definition
of
music
should
take
account
of
it
as
a
human
practice.
My
point
is
that,
unlike
mathematics,
which
may
be
characterized
as
a
closed
system
without
regard
to
its
human
applications,
music
is
not
a
closed
formal
system.
As
I
argue
in
the
next
section,
to
understand
what
music
is
necessarily
involves
considering
the
non-‐musical
as
well
as
musical
social
practices
that
constrain
its
historical
development.
If
this
does
not
take
us
back
full
circle
to
consider
again
music's
multi-‐functionality,
it
at
least
takes
us
back
to
the
story
about
musical
traditions
and
socio-‐historical
human
practices.
Outside
of
these
it
is
doubtful
that
organized
musical
sounds
would
count
as
music
and,
hence,
a
structural
definition
cannot
suffice.
14
The
emphasis
was
both
on
the
musical
"language"
and
structural
conventions
of
that
tradition
and
on
the
present
musician's
intention
to
use
these
to
add
new
music
to
the
tradition.
But
there
are
two
other
respects
in
which
the
relation
to
tradition
might
be
important
in
defining
music.
One
difficulty
with
the
structural
approach
was
that
not
all
music
is
melodic,
teleological,
assembled
in
terms
of
successive,
sometimes
repeated
or
reprised,
sections.
But
by
contextualizing
music
historically,
we
may
be
able
to
see
the
presently
diverse
array
of
possibilities
as
emerging
in
a
regular
fashion.
If
we
look
at
the
history
of
Western
music,
it
seems
plausible
to
suggest
that
song
(and
dance
accompanied
by
rhythmically
regular
music,
whether
vocal
or
not)
preceded
instrumental
ensemble
music,
that
monodic,
antiphonal,
and
verse–
chorus
forms
of
singing
were
succeeded
by
harmonic
or
polyphonic
vocal
music,
that
short
one-‐movement
forms
came
before
longer
multi-‐movement
forms,
that
tonically
and
harmonically
simple
music
was
progressively
complexified,
and
that
tonal/modal
music
preceded
atonal
music.
One
can
see
this
as
change
and
development
in
the
"language"
that
the
composer
had
at
his
or
her
disposal
at
any
given
moment.
Awareness
of
the
direction
of
historical
musical
change
might
also
allow
us
to
generate
more
context-‐sensitive
and
perspicuous
generative
principles
of
musical
construction.
In
effect,
rather
than
trying
to
define
music
per
se,
we
might
do
better
by
defining
it
as
what
could
be
created
as
music
at
a
given
time
in
a
given
tradition.
To
take
one
example
noted
previously,
there
can
indeed
be
music
that
involves
only
sequence
and
not
pattern,
but
this
possibility
became
available
in
the
West
only
in
the
twentieth
century,
through
the
piecemeal
rejection
and
deconstruction
of
some
generative
principles
that
previously
excluded
this
option.
The
second
consideration
holds
that
the
tradition
was
too
narrowly
conceived
anyway.
The
metaphor
of
a
musical
language
can
be
misleading
in
many
ways,
not
only
because
it
can
deflect
attention
from
the
historical
relativity
of
musical
modes
of
expression,
as
was
just
noted,
but
because
it
fails
to
notice
how
far
beyond
the
immediate
business
of
creating
music
that
tradition
extends.
The
tradition
must
be
taken
to
include,
for
instance,
the
methods
of
making
musical
instruments,
limitations
to
technical
innovation
in
the
field
of
15
instrument
making,
and
how
the
use
of
instruments
is
transmitted.9
It
will
notice
the
development
of
notations
and
how
this
affected
musical
practice.
Similarly,
it
should
consider
wider
attitudes
to
the
various
purposes
of
music
making.
And
the
important
point
to
realize
is
that,
at
every
level,
such
matters
were
affected
by
conventions,
pragmatic
concerns,
and
considerations
that
were
not
at
all
musical
or
aesthetic.
As
people
who
now
regularly
access
music
through
mechanical
reproduction
devices,
we
are
often
indifferent
to
or
ignorant
of
the
practical
business
of
making
music
and
how
that
was
regulated
as
only
one
domain
within
a
wider
social
web.
Only
someone
desensitized
in
this
way
could
think
that
Platonic
formalism
could
provide
for
music's
definition.
Music
is
not
patterned
sound,
it
is
the
bodying
forth
of
sound
through
human
action,
and
the
relevant
actions
are
conditioned
not
only
by
musical
factors
but
by
socio-‐cultural
arrangements
and
resource
availabilities
that
have
nothing
to
do
with
music
as
such.
Musical
"universals"
Earlier
I
identified
the
disparate
nature
of
musical
traditions
as
a
barrier
to
a
socio-‐historical
definition
of
music:
the
difficulty
lies
in
characterizing
what
shared
characteristics
make
music
traditions
music
traditions,
despite
their
manifest
differences.
The
hybrid
approach
to
music's
definition
that
I
have
advocated
faces
the
same
problem.
If
the
course
of
a
music
tradition
is
molded
arbitrarily
by
non-‐musical,
socio-‐historical
circumstances,
so
that
distinct
traditions
need
not
closely
resemble
each
other,
how
can
they
be
jointly
identified
as
musical
traditions
without
begging
the
definitional
question?
The
answer,
I
suggest,
is
by
appeal
to
musical
"universals,"
though
I
place
the
term
in
scare
quotes
because
my
account
of
these
universals
is
perhaps
not
the
orthodox
one.
Psychologists
and
others
have
identified
a
number
of
musical
universals.
For
example,
music
is
based
on
tonal/modal
scales
that
share
a
variety
of
features:
octaves
are
inevitably
recognized
as
equivalents;
starting
with
the
most
9
For
discussion,
see
Godlovitch
1998.
16
stable,
tonic
note,
the
octave
is
divided
into
five
to
seven
intervals;
these
intervals
are
not
all
equal;
and
the
scale
usually
includes
a
major
fourth
or
fifth
above
the
tonic.10
Or
to
take
a
different
example,
lullabies
share
certain
melodic
features
cross-‐culturally
(Unyk
et
al
1992).
The
manner
in
which
they
are
discussed
often
implies
that
musical
universals
apply
to
all
music.
This
will
be
true,
I
suppose,
of
those
that
are
a
consequence
of
hardwired
neural
circuits.
For
example,
the
experience
of
octave
equivalence
is
humanly
universal
and
is
shared
with
some
non-‐musical
animals,
so
is
a
likely
function
of
auditory
processing
of
a
rather
general
kind.
But
I
have
already
noted
that
not
all
music
is
based
on
tonal
scales
of
the
kind
identified
above
as
universal.
Serial
music
in
the
Webernian
style
employs
an
atonal
scale
in
which
the
octave
is
divided
into
twelve
equal
intervals.
And
the
slendro
scale
of
classical
central
Javanese
music
employs
five
equal
intervals.
Moreover,
though
the
notion
of
universality
is
sometimes
taken
to
entail
a
normative
notion
of
musical
naturalness,
so
that
departures
of
the
kinds
just
noted
might
be
viewed
as
unnatural
or
unmusical,
that
attitude
surely
is
not
justified.
Consideration
of
the
variability
of
musical
norms
from
culture
to
culture
and
reflection
on
the
historical
malleability
in
the
Western
tradition
of
what
counts,
say,
as
an
unacceptable
discord
provide
ample
warning
against
underestimating
the
flexibility
and
arbitrariness
of
musical
conventions
and
practices.
In
other
words,
we
cannot
define
music
simply
in
terms
of
its
exemplifying
the
"universals"
identified
by
psychologists.
A
weaker
but
more
plausible
account
of
universality
would
tie
it
to
what
is
common
not
to
all
music
but
to
all
musical
traditions.
On
this
view,
at
least
some
of
the
music
in
a
tradition
will
conform
to
universal
norms,
even
if
the
tradition
also
licenses
other
kinds
or
styles
of
music
that
do
not.
This
weaker
view
is
sufficient
for
our
purposes:
we
can
define
musical
traditions
as
those
displaying
the
appropriate
universals
in
at
least
some
of
their
music,
and
music
as
what
is
produced
in
musical
traditions.
What
is
more,
at
first
glance
the
10
For
discussion
of
these
and
other
universal
features
of
music,
see
Justus
and
Hustler
2005,
McDermott
and
Hauser
2005,
Higgins
2006,
Stevens
and
Byron
2009,
Thompson
and
Balkwill
2010.
17
proposition
that
all
musical
traditions
include
music
displaying
the
appropriate
universals
appears
to
be
true.
For
instance,
the
Western
classical
tradition
includes
much
tonal/modal
music
based
on
scales
with
the
previously
mentioned
"universal"
features,
and
Javanese
music
also
employs
the
pelog
scale,
which
has
unequal
intervals.
Nevertheless,
even
this
notion
of
universality
might
be
unacceptably
strong.
Though
I
cannot
think
of
a
real-‐world
example,
it
is
conceivable
that
a
tradition
would
come
to
reject
types
of
music
displaying
universals
–
that
is,
other
than
those
that
are
hard-‐wired
as
part
of
ordinary
auditory
analysis
–
and
entrench
styles
that
do
not,
so
that
none
of
its
current
music
shares
these
universals
with
other
traditions.
Perhaps
pelog
comes
to
sound
over-‐scented
and
effete
to
Javanese
ears,
so
that
they
abandon
it
in
favor
of
the
exclusive
use
of
slendro.
Here
is
the
yet
weaker
relation
that
I
expect
to
hold
between
musical
traditions
and
musical
universals:
though
musical
universals
may
not
be
apparent
in
all,
or
even
any,
of
the
music
in
a
developed
musical
tradition,
it
is
reasonable
to
speculate
that
every
tradition
based
its
music
on
those
universals
at
the
outset.
In
other
words,
what
unifies
musical
traditions
as
musical
is
their
initial
reliance
on
the
kinds
of
features
identified
as
"universal"
to
music.
Accordingly,
there
is
a
direction
to
music's
historical
development.
This
maps
onto
the
position
I
presented
earlier:
monodic
music
preceded
polyphony,
simple
bipartite
or
repeated
forms
came
before
multi-‐part
forms,
and
so
on.
Further
reflection
on
how
musical
universals
arise
demonstrates
how
psychologists'
and
neuroscientists'
studies
bear
on
the
topic.
As
I
have
noted
for
the
case
of
the
experience
of
octave
equivalence,
musical
universals
can
be
by-‐
products
of
processing
biases
in
systems
developed
for
ordinary,
non-‐musical
auditory
processing.
In
addition,
they
may
reflect
widely
shared
music-‐specific
preferences,
either
based
on
hardwired
neural
circuits
generated
through
processes
of
evolution
that
saw
the
emergence
of
music
as
an
adaptation
in
its
own
right,
or,
alternatively,
based
on
music's
incidental
capacity
to
engage
pleasure-‐inciting
brain
systems
that
evolved
originally
to
motivate
non-‐musical
behaviors.
But
not
all
musical
universals
need
be
grounded
in
inherited,
hardwired
neurophysiological
systems,
and
for
this
reason
the
need
to
consider
18
the
social
histories
of
non-‐musical
practices
is
again
apparent.
For
instance,
the
universal
structure
of
lullabies
might
reflect
not
strictly
musical
preferences
but
exploitation
of
elements
of
infant-‐directed
speech,
which
has
distinctively
musical
qualities
(Fernald
1992).
Or
to
return
to
the
more
general,
our
species'
evolved
sociality
and
reliance
on
both
cooperation
and
coordination
might
have
conspired
to
create
conditions
under
which
all
societies
were
bound
to
produce
(some)
socially
shared
music
that
universally
displays
features
(of
transparency,
memorability,
restricted
melodic
range
etc.
etc.)
necessary
for
the
group
all
to
take
part
in
its
production.
19
produce
pieces
with
the
same
sound
structure,
though
one
aimed
to
write
an
avant-‐garde
musical
work
and
the
other
to
create
a
piece
of
sound
art.
The
obvious
move
here
is
to
appeal
to
the
artist's
intentions
as
determining
whether
music
is
the
product.
And
I
am
happy
to
take
this
direction.
But
it
is
important
to
be
clear
that
the
social
context
enables
the
relevant
intentions,
which
are
to
be
thought
of
as
institutional
decisions
rather
than
merely
as
acts
of
willing.
If
the
wind
chime
maker
intended
that
the
chime
produce
music,
his
intention
would
fail.
He
does
not
have
the
credentials
to
bring
off
the
result.
And
it
would
not
be
music
that
was
sounded
even
if,
by
chance,
the
chime
produced
something
structurally
identical
to
the
Mozart
sonata.
But
if
an
established
avant-‐garde
composer
included
the
wind
chime
in
his
next
piece,
its
sounds
would
contribute
to
the
musical
content
of
that
piece.
We
can
follow
Levinson
and
Kania
in
appealing
to
composers'
intentions
as
authorizing
that
something
is
music,
provided
we
are
aware
of
the
role
limits
and
other
social
constraints
relevant
to
determining
the
efficacy
of
such
intentions.
Conclusion
The
best
bet,
I
think,
for
defining
music
would
appeal
to
the
intentional
use
of
structural/generative
principles
viewed
historically
against
the
background
of
musical
traditions
that
are
construed
sufficiently
broadly
that
they
take
in
not
only
the
immediate
practices
connected
with
music
making
but
also
the
cultural
forces
that
facilitate
and
structure
this.
This
involves
acknowledging
that
many
of
these
forces
direct
the
development
of
musical
traditions
in
culturally
arbitrary
directions.
To
explain
how
quite
distinct
historically
developed
music
traditions
all
are
music
traditions,
some
account
of
the
role
of
musical
universals
is
in
order.
Rather
than
identifying
the
defining
characteristics
of
all
music,
they
are
relevant
for
indicating
the
starting
point
common
to
the
world's
diverse
array
of
musical
traditions.
To
be
successful,
a
definition
of
music
must
balance
these
various
stories,
defining
music
in
terms
of
its
traditions,
the
universal
elements
and
common
generative
principles
that
these
traditions
licensed
at
their
origins,
the
musical
and
non-‐musical
pressures
that
constrain
music's
20
development
and
enable
its
diversity,
and
the
intentions
and
understandings
of
music
makers
and
listeners.11
11
For
their
helpful
comments
I
thank,
Julian
Dodd,
Andrew
Kania,
Justine
21
References
Ball,
Philip
2010.
The
Music
Instinct,
London:
Bodley
Head
Barrett,
H.
Clark
and
Kurzban,
Robert
2006.
"Modularity
in
Cognition:
Framing
the
Debate,"
Psychological
Review,
113:
628–47.
Brown,
Steven
2000.
"The
'Musilanguage'
Model
of
Music
Evolution,"
in
The
Origins
of
Music,
edited
by
N.
L.
Wallin,
B.
Merker
and
S.
Brown,
Cambridge,
MA:
MIT
Press,
271–300.
Cross,
Ian
2007.
"Music
and
Cognitive
Evolution,"
in
The
Oxford
Handbook
of
Evolutionary
Psychology,
edited
by
R.
I.
M.
Dunbar
and
L.
Barrett,
Oxford:
Oxford
University
Press,
649–67.
Darwin,
Charles
1880.
The
Descent
of
Man
and
Selection
in
Relation
to
Sex,
London:
D.
Appleton.
Revised
and
augmented
edition.
First
edition
1871.
Davies,
Stephen
2001.
Musical
Works
and
Performances:
A
Philosophical
Exploration,
Oxford:
Clarendon
Press.
———.
2003.
"John
Cage's
4'33":
Is
it
Music?"
in
S.
Davies,
Themes
in
the
Philosophy
of
Music,
Oxford:
Oxford
University
Press,
11–29.
———.
2007.
"Essential
Distinctions
for
Art
Theorists,"
in
S.
Davies,
Philosophical
Perspectives
on
Art,
Oxford:
Oxford
University
Press,
23–38.
———.
2010.
"Perceiving
Melodies
and
Perceiving
Musical
Colors,"
Review
of
Philosophy
and
Psychology,
1:
19–39.
———.
Forthcoming.
The
Artful
Species:
Aesthetics,
Art,
and
Evolution,
Oxford:
Oxford
University
Press.
De
Smedt,
Johan
and
De
Cruz,
Helen
2010.
"Toward
an
Integrative
Approach
of
Cognitive
Neuroscientific
and
Evolutionary
Psychological
Studies
of
Art,"
Evolutionary
Psychology,
8:
695–719.
Dissanayake,
Ellen
2006.
"Ritual
and
Ritualization:
Musical
Means
of
Conveying
and
Shaping
Emotions
in
Humans
and
Other
Animals,"
in
Music
and
Manipulation:
On
the
Social
Uses
and
Social
Control
of
Music,
edited
by
S.
Brown
and
U.
Volgsten,
Oxford:
Berghahn,
31–56.
Fenk-‐Oczlon,
Gertraud
and
Fenk
2009-‐2010.
"Some
Parallels
between
Language
and
Music
from
a
Cognitive
and
Evolutionary
Perspective,"
Musicae
Scientiae,
13:
201–26.
22
Fernald,
Anne
1992.
"Meaningful
Melodies
in
Mothers'
Speech
to
Infants,
in
Nonverbal
Vocal
Communication:
Comparative
and
Developmental
Aspects,
edited
by
H.
Papousek,
U.
Jurgens,
and
M.
Papousek,
Cambridge:
Cambridge
University
Press,
262–82.
Fitch,
W.
Tecumseh
2005.
"The
Evolution
of
Music
in
Comparative
Perspective,"
Annals
of
New
York
Academy
of
Sciences,
1060:
29–49.
Godlovitch,
Stan
1998.
Musical
Performance:
A
Philosophical
Study,
London:
Routledge.
Grosse,
Ernst
1897.
The
Beginnings
of
Art,
New
York:
Appleton.
Hamilton,
Andy
2007.
Aesthetics
and
Music,
London:
Continuum.
Higgins,
Kathleen
Marie
2006.
"The
Cognitive
and
Appreciative
Impact
of
Musical
Universals,"
Revue
Internationale
de
Philosophie,
60:
487–503.
Huron,
David
2003.
"Is
Music
an
Evolutionary
Adaptation?"
in
The
Cognitive
Neuroscience
of
Music,
edited
by
I.
Peretz
and
R.
J.
Zatorre,
Oxford,
Oxford
University
Press,
57–75.
Justus,
Timothy
and
Hutsler,
Jeffrey
J.
2005.
"Fundamental
Issues
in
the
Evolutionary
Psychology
of
Music:
Assessing
Innateness
and
Domain
Specificity,"
Music
Perception,
23:
1–27.
Kania,
Andrew
2010.
"Silent
Music,"
Journal
of
Aesthetics
and
Art
Criticism,
68:
343–53.
———.
2011.
"Definition,"
in
The
Routledge
Companion
to
Philosophy
and
Music,
edited
by
T.
Gracyk
and
A.
Kania,
London:
Routledge,
3–13.
Koelsch,
Stefan
and
Siebel,
Walter
A.
2005.
"Towards
a
Neural
Basis
of
Music
Perception,"
Trends
in
Cognitive
Sciences,
9:
578–84.
Lerdahl,
Fred
and
Jackendoff,
Ray
1983.
A
Generative
Theory
of
Tonal
Music,
Cambridge,
MA:
MIT
Press.
Levinson,
Jerrold
1990.
Music,
Art,
and
Metaphysics,
Ithaca:
Cornell
University
Press.
Levitin,
Daniel
J.
2006.
This
is
your
Brain
on
Music:
the
Science
of
a
Human
Obsession,
New
York:
Dutton.
McDermott,
Josh
and
Hauser,
Marc
D.
2005.
"The
Origins
of
Music:
Innateness,
Uniqueness,
and
Evolution,"
Music
Perception,
23:
29–59
23
Mithen,
Steven
J.
2005.
The
Singing
Neanderthals:
the
Origins
of
Music,
Language,
Mind,
and
Body,
London,
Weidenfeld
&
Nicolson.
Patel,
Aniruddh
D.
2008.
Music,
Language,
and
the
Brain,
Oxford:
Clarendon
Press.
———.
2010.
"Music,
Biological
Evolution,
and
the
Brain,"
in
Emerging
Disciplines,
edited
by
M.
Bailar,
Houston:
Rice
University
Press,
91–144.
Peretz,
Isabelle
and
Coltheart,
Max
2003.
"Modularity
of
Music
Processing,"
Nature
Neursocience,
6:
688–91.
Pinker,
Steven
1999.
How
the
Mind
Works,
London:
Penguin
Books.
Spencer,
Herbert.
(1966).
The
Works
of
Herbert
Spencer,
21
Vols.
Osnabrüch:
Otto
Zeller.
Stevens,
Catherine,
and
Byron,
Tim
2009.
"Universals
in
Music
Processing,"
in
The
Oxford
Handbook
of
Music
Psychology,
edited
by
S.
Hallam,
I.
Cross,
and
M.
Thaut,
Oxford:
Oxford
University
Press,
14–23.
Thompson,
William
Forde
and
Balkwill,
Laura-‐Lee
2010.
"Cross-‐cultural
Similarities,"
in
Oxford
Handbook
of
Music
and
Emotion:
Theory,
Research,
Applications,
edited
by
P.N.
Juslin
and
John
Sloboda,
Oxford:
Oxford
University
Press,
755–88.
Unyk,
Anna
M.,
Trehub,
Sandra
E.,
Trainor,
Laurel
J.,
and
Schellenberg,
E.
Glenn
1992.
"Lullabies
and
Simplicity:
A
Cross-‐cultural
Perspective,"
Psychology
of
Music,
20:
15–28.
Ziff,
Paul
1953.
"The
Task
of
Defining
a
Work
of
Art,"
Philosophical
Review,
62:
58–78.
24