Combinatoriality and The Aggregate
Combinatoriality and The Aggregate
Let A = hexachord a b c d e f
Let A + n = transposition of A by n semitones =
a+n b+n c+n d+n e+n f+n
(where n is unknown and may stand for the principal intervals 1 2 3 4 5 6 or their
complements E T 9 8 7)
then: A + A + n are complementary hexachords iff A contains no interval of n
semitones
Milton Babbitt first demonstrated that the pitch invariance of transposed collections
could be predicted from the interval classes within the collection or its complement. A
'semi-combinatorial set' is one for which only one transformation other than retrograde
(prime, inversion, retrograde inversion), under some transposition (T), will generate a
new set form in which the first hexachord will contain only those pcs not contained in
the first hexachord of the original set form.
Each semi-combinatorial hexachord omits at least one interval or at least one odd sum,
among all possible pairs of pcs (allowing the creation of an aggregate by inversion
using complementation to that odd sum).
Each all-combinatorial hexachord omits at least one interval class (allowing the creation
of an aggregate by transposition), and is inversionally symmetrical with itself
(equivalent to a transposition of ints own inversion).
All-combinatorial tets:
0369 4-28 0167 4-9 0123 4-1 0235 4-10 0257 4-23 0127 4-6
All-combinatorial trikes:
048 3-12 012 3-1 024 3-6 027 3-9
An all-combinatorial tetrachord contains one and only one member of each augmented
triad (it omits ic4), and is inversionally symmterical to itself; a semi-combinatorial
tetrachord obeys only the first condition.
An all-combinatorial trichord must contain one and only one member of each
diminished triad (it omits ic3), and is inversionally symmterical to itself. All other
trichords except the diminished triad (036 or 3-10) are semi-combinatorial. Each such
trichord and one of its inversions will form an all-combinatorial hexachord.
For every 12-note transformation whose segments generate aggregates with the
corresponding segments of the original set form (i.e., the transformation shares no
invariant pitches with the original set), there is a transformation whose segments will
completely duplicate the pc content of the original segments.