for the listener.If we want briefly and intuitively to describe the analytic
principle proposed by Lerdahl,salience most often consists of the obses-
sive and contextually dominant repetition of a sonority (chord).Domi-
nance of the salient element plays a role in atonal music analogous to
the role of the tonally stable element in tonal music and gives rise to
what we may call prolongation by iteration.
This poses a major problem for the cognitive psychology of memory
and perception,as Lerdahl noted himself:“The crux of the theory out-
lined above is the decision to regard contextual salience in atonal music
as analogous to stability in tonal music.This step amounts to an acknowl-
edgment that atonal music is not very grammatical.I think this is an accu-
rate conclusion.Listeners to atonal music do not have at their disposal
a consistent,psychologically relevant set of principles by which to orga-
nize pitches at the musical surface.As a result,they grab on to what they
can:relative salience becomes structurally important,and within that
framework the best linear connections are made.Schoenberg had reason
to invent a new system”(1989:84).This is exactly the problem that con-
cerns perception and interpretation of completely atonal pieces.Rela-
tive salience determines a temporary,always modifiable,hierarchy.One
of the characteristics of atonal music,and serial music in particular,is
that it is extremely fluent for the listener,that it does not have a definite
structure,thus opening itself to the creative fantasy of the interpreter.Dynamic Aspects of Salience Clues and the Concept of Macrostructure
In the case of atonal music,the question immediately arises of how
relative salience can create,for listener and interpreter,the equivalent
of alternations of tension and relaxation that underlie the emotional
dynamism of tonal music.More exactly,does the relative contextual
salience of an event also determine its degree of stability?
Until now,most theoretical and experimental studies responded in
the affirmative,but dealt only with cognitive and structural aspects of
salience clues,thus demonstrating stability only as abstract and concep-
tual.Deliège (1989,1990,1993) developed in detail concepts of clue
extraction from the musical surface,and of imprint in memory.In a series
of experiments on Berio’s Sequenza VIand Boulez’s Eclat,she showed
that listeners,while listening repeatedly to pieces they do not know and
for which there are no tonal reference points,create a simplified schema
of what they hear in the form of an imprint stored in memory.Details of
this imprint develop into a model-type in relation to numerous varia-
tions of successive listenings.This imprint,which is a sort of image the
listener keeps of the musical piece,is progressively elaborated through
clues taken from the musical surface.At first these clues are anything
that can capture the listener’s attention and make certain events salient457 Innate Competencies in Musical Communication