Generator-offset property: Difference between revisions

Inthar (talk | contribs)
mNo edit summary
Inthar (talk | contribs)
mNo edit summary
Line 1: Line 1:
A scale satisfies the '''alternating generator property''' (also '''AG''' or '''alt-gen''') if it satisfies the following equivalent properties:
A scale satisfies the '''alternating generator property''' (also '''alt-gen''' or '''AG''') if it satisfies the following equivalent properties:
* the scale can be built by stacking alternating generators
* the scale can be built by stacking alternating generators
* the scale is generated by two chains of generators separated by a fixed interval, and the lengths of the chains differ by at most one.
* the scale is generated by two chains of generators separated by a fixed interval, and the lengths of the chains differ by at most one.


[[Diasem]] is an example of an AG scale, because it is built by stacking alternating 7/6 and 8/7 for [[chirality|left-handed]] diasem, or 8/7 and 7/6 for right-handed diasem.
[[Diasem]] is an example of an alt-gen scale, because it is built by stacking alternating 7/6 and 8/7 for [[chirality|left-handed]] diasem, or 8/7 and 7/6 for right-handed diasem.


More formally, a cyclic word ''S'' (representing a [[periodic scale]]) of size ''n'' is '''AG''' if it satisfies the following equivalent properties:
More formally, a cyclic word ''S'' (representing a [[periodic scale]]) of size ''n'' is '''alt-gen''' if it satisfies the following equivalent properties:
# ''S'' can be built by stacking a single chain of alternating generators ''g''<sub>1</sub> and ''g''<sub>2</sub>, resulting in a circle of the form either ''g''<sub>1</sub> ''g''<sub>2</sub> ... ''g''<sub>1</sub> ''g''<sub>2</sub> ''g''<sub>1</sub> ''g''<sub>3</sub> or ''g''<sub>1</sub> ''g''<sub>2</sub> ... ''g''<sub>1</sub> ''g''<sub>2</sub> ''g''<sub>3</sub>.
# ''S'' can be built by stacking a single chain of alternating generators ''g''<sub>1</sub> and ''g''<sub>2</sub>, resulting in a circle of the form either ''g''<sub>1</sub> ''g''<sub>2</sub> ... ''g''<sub>1</sub> ''g''<sub>2</sub> ''g''<sub>1</sub> ''g''<sub>3</sub> or ''g''<sub>1</sub> ''g''<sub>2</sub> ... ''g''<sub>1</sub> ''g''<sub>2</sub> ''g''<sub>3</sub>.
# ''S'' is generated by two chains of generators separated by a fixed interval; either both chains are of size ''n''/2, or one chain has size (''n'' + 1)/2 and the second has size (''n''&nbsp;&minus;&nbsp;1)/2.
# ''S'' is generated by two chains of generators separated by a fixed interval; either both chains are of size ''n''/2, or one chain has size (''n'' + 1)/2 and the second has size (''n''&nbsp;&minus;&nbsp;1)/2.
Line 12: Line 12:
== Theorems ==
== Theorems ==
=== Theorem 1 ===  
=== Theorem 1 ===  
If a 3-step-size scale word ''S'' in L, M, and s is both AG and unconditionally [[MV3]] (i.e. MV3 regardless of tuning), then the scale is of the form ''ax by bz'' for some permutation (''x'', ''y'', ''z'') of (L, M, s); and the scale's cardinality (size) is either odd, or 4 (and is of the form ''xyxz''). Moreover, any odd-cardinality AG scale is unconditionally MV3.
If a 3-step-size scale word ''S'' in L, M, and s is both alt-gen and unconditionally [[MV3]] (i.e. MV3 regardless of tuning), then the scale is of the form ''ax by bz'' for some permutation (''x'', ''y'', ''z'') of (L, M, s); and the scale's cardinality (size) is either odd, or 4 (and is of the form ''xyxz''). Moreover, any odd-cardinality alt-gen scale is unconditionally MV3.
==== Proof ====
==== Proof ====
Assuming both AG and unconditionally MV3, we have two chains of generator ''g''<sub>0</sub> (going right). The two cases are:
Assuming both alt-gen and unconditionally MV3, we have two chains of generator ''g''<sub>0</sub> (going right). The two cases are:
  CASE 1: EVEN CARDINALITY
  CASE 1: EVEN CARDINALITY
  O-O-...-O (n/2 notes)
  O-O-...-O (n/2 notes)
Line 35: Line 35:
Choose a tuning where ''g''<sub>1</sub> and ''g''<sub>2</sub> are both very close to but not exactly 1/2*''g''<sub>0</sub>, resulting in a scale very close to the mos generated by 1/2 ''g''<sub>0</sub>. (i.e. ''g''<sub>1</sub> and ''g''<sub>2</sub> differ from 1/2*''g''<sub>0</sub> by ε, a quantity much smaller than the chroma of the ''n''/2-note mos generated by ''g''<sub>0</sub>, which is |''g''<sub>3</sub> &minus; ''g''<sub>2</sub>|). Thus we have 4 distinct sizes for ''k''-steps:
Choose a tuning where ''g''<sub>1</sub> and ''g''<sub>2</sub> are both very close to but not exactly 1/2*''g''<sub>0</sub>, resulting in a scale very close to the mos generated by 1/2 ''g''<sub>0</sub>. (i.e. ''g''<sub>1</sub> and ''g''<sub>2</sub> differ from 1/2*''g''<sub>0</sub> by ε, a quantity much smaller than the chroma of the ''n''/2-note mos generated by ''g''<sub>0</sub>, which is |''g''<sub>3</sub> &minus; ''g''<sub>2</sub>|). Thus we have 4 distinct sizes for ''k''-steps:
# ''a''<sub>1</sub>, ''a''<sub>2</sub> and ''a''<sub>3</sub> are clearly distinct.
# ''a''<sub>1</sub>, ''a''<sub>2</sub> and ''a''<sub>3</sub> are clearly distinct.
# ''a''<sub>4</sub> &minus; ''a''<sub>3</sub> = ''g''<sub>1</sub> &minus; ''g''<sub>2</sub> != 0, since the scale is a non-trivial AG.  
# ''a''<sub>4</sub> &minus; ''a''<sub>3</sub> = ''g''<sub>1</sub> &minus; ''g''<sub>2</sub> != 0, since the scale is a non-trivial alt-gen.  
# ''a''<sub>4</sub> &minus; ''a''<sub>1</sub> = ''g''<sub>3</sub> &minus; ''g''<sub>2</sub> = (''g''<sub>3</sub> + ''g''<sub>1</sub>) &minus; (''g''<sub>2</sub> + ''g''<sub>1</sub>) != 0. This is exactly the chroma of the mos generated by ''g''<sub>0</sub>.
# ''a''<sub>4</sub> &minus; ''a''<sub>1</sub> = ''g''<sub>3</sub> &minus; ''g''<sub>2</sub> = (''g''<sub>3</sub> + ''g''<sub>1</sub>) &minus; (''g''<sub>2</sub> + ''g''<sub>1</sub>) != 0. This is exactly the chroma of the mos generated by ''g''<sub>0</sub>.
# ''a''<sub>4</sub> &minus; ''a''<sub>2</sub> = ''g''<sub>1</sub> &minus; 2 ''g''<sub>2</sub> + ''g''<sub>3</sub> = (''g''<sub>3</sub> &minus; ''g''<sub>2</sub>) + (''g''<sub>1</sub> &minus; ''g''<sub>2</sub>) = (chroma ± ε) != 0 by choice of tuning.
# ''a''<sub>4</sub> &minus; ''a''<sub>2</sub> = ''g''<sub>1</sub> &minus; 2 ''g''<sub>2</sub> + ''g''<sub>3</sub> = (''g''<sub>3</sub> &minus; ''g''<sub>2</sub>) + (''g''<sub>1</sub> &minus; ''g''<sub>2</sub>) = (chroma ± ε) != 0 by choice of tuning.


By applying this argument to 1-steps, we see that there must be 4 step sizes in some tuning, a contradiction. Thus ''g''<sub>1</sub> and ''g''<sub>2</sub> must themselves be step sizes. Thus we see that an even-cardinality, unconditionally MV3, AG scale must be of the form ''xy...xyxz''. But this pattern is not unconditionally MV3 if ''n'' ≥ 6, since 3-steps come in 4 sizes: ''xyx'', ''yxy'', ''yxz'' and ''xzx''. Thus ''n'' = 4 and the scale is ''xyxz''.
By applying this argument to 1-steps, we see that there must be 4 step sizes in some tuning, a contradiction. Thus ''g''<sub>1</sub> and ''g''<sub>2</sub> must themselves be step sizes. Thus we see that an even-cardinality, unconditionally MV3, alt-gen scale must be of the form ''xy...xyxz''. But this pattern is not unconditionally MV3 if ''n'' ≥ 6, since 3-steps come in 4 sizes: ''xyx'', ''yxy'', ''yxz'' and ''xzx''. Thus ''n'' = 4 and the scale is ''xyxz''.


In case 2, let (2, 1) &minus; (1, 1) = ''g''<sub>1</sub>, (1, 2) &minus; (2, 1) = ''g''<sub>2</sub> be the two alternating generators. Let ''g''<sub>3</sub> be the leftover generator after stacking alternating ''g''<sub>1</sub> and ''g''<sub>2</sub>. Then the generator circle looks like ''g''<sub>1</sub> ''g''<sub>2</sub> ''g''<sub>1</sub> ''g''<sub>2</sub> ... ''g''<sub>1</sub> ''g''<sub>2</sub> ''g''<sub>3</sub>. Then the generators corresponding to a step are:
In case 2, let (2, 1) &minus; (1, 1) = ''g''<sub>1</sub>, (1, 2) &minus; (2, 1) = ''g''<sub>2</sub> be the two alternating generators. Let ''g''<sub>3</sub> be the leftover generator after stacking alternating ''g''<sub>1</sub> and ''g''<sub>2</sub>. Then the generator circle looks like ''g''<sub>1</sub> ''g''<sub>2</sub> ''g''<sub>1</sub> ''g''<sub>2</sub> ... ''g''<sub>1</sub> ''g''<sub>2</sub> ''g''<sub>3</sub>. Then the generators corresponding to a step are:
Line 49: Line 49:
(The above holds for any odd ''n'' ≥ 3.)
(The above holds for any odd ''n'' ≥ 3.)


Now we only need to see that AG + odd cardinality => unconditionally MV3. But the argument in case 2 above works for any interval class (unconditional MV3 wasn't used), hence any interval class comes in at most 3 sizes regardless of tuning.
Now we only need to see that alt-gen + odd cardinality => unconditionally MV3. But the argument in case 2 above works for any interval class (unconditional MV3 wasn't used), hence any interval class comes in at most 3 sizes regardless of tuning.


== Conjectures ==
== Conjectures ==
Line 58: Line 58:
# is not of the form ''mx my mz'',
# is not of the form ''mx my mz'',
# and is not of the form ''xyxzxyx'',
# and is not of the form ''xyxzxyx'',
then it is AG.
then it is alt-gen.
[[Category:Theory]]
[[Category:Theory]]
[[Category:AG scales| ]]<!--Main article-->
[[Category:alt-gen scales| ]]<!--Main article-->