[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: On "learned" A/P, lattice / grid

Dear Kevin,

I think it is very dangerous to conclude from what a subject can do, what it is he cannot do.

Labeling pitches can only be done with the speed of internal speech and this is, as we know from the famous one page publication by Tom Landauer, as fast as really pronounced speech. However in music there are many more voices and often notes that are faster than syllables. All these notes are perceived in a holistic manner, exactly as happens in a non-AP subject.

If it AP existed in Brady in a covert way chances are that it existed in you and all other people too. Did you spend a year to find it back?

Kind regards,


On 02 Sep 2009, at 16:53, Kevin Austin wrote:

Thank you. The article is very interesting and informative.


From Parncutt & Levitin, Absolute Pitch: (abstracted)

They, AP possessors, may work out integrated qualities (intervals and chords) by reconstructing them from the notes (note names), rather than perceiving an 'integrated' sonority (Miyazaki, 1992, 1993). There is the proposition that "melody" is not heard, but rather a string of pitches passing by.

http://books.google.com/books?id=IhFOe_bBb1UC&pg=PA39&lpg=PA39&dq=%22I+don't+hear+melodies,+I+hear+pitch+names+passing+by.%22&source=bl&ots=4qII4vElNL&sig=6ZjUS-5KDXtyq_8PzuBX90p5EGQ&hl=en&ei=a36eSp6kLoq7lAe9oNSbDQ&sa=X&oi=book_result&ct=result&resnum=2#v =onepage&q=%22I%20don't%20hear%20melodies%2C%20I%20hear%20pitch %20names%20passing%20by.%22&f=false

An equivalent to this for non-AP people would be to read a chinese text, with a knowledge of how chinese characters are constructed (radicals and combinations), but having no sense of what the characters mean, or how they relate.

This seems to correspond to the statement:

Date:    Tue, 1 Sep 2009 17:20:33 +0200
From:    Leon van Noorden <leonvannoorden@xxxxxxx>
Subject: Re: Linearity as pitch perception: was Perception as memory

It was true in the time that I had to make music dictations, very long ago, that I had to reconstruct the interval or chord from the notes. It does not mean that you could not say what kind of chord it was from the sound, such a major or minor.

As a non-AP listener, I hear (only) sets of relationships, and I have developed large numbers of 'musical hierarchies' to categorize these relationships. My hearing is (more or less) pattern-based, what I refer to as 'process-oriented listening'. I note repetition, variation, transformation, lattice / pattern matching etc.

In the article cited, P T Brady says that he 'taught himself' AP -- an idea that many have difficulty with. Count me in this group. As with Eliot Handelman, my alternate interpretation is that the AP existed, but some aspects of it had been extinguished for some reason, and he had re-discovered it with a one year re-training.

The visual model I use for this distinction is that used in certain kinds of cross-country car racing where the instructions read:
Drive to the second traffic light, turn left
After five stop signs, turn right
One block after the school on your left, turn right

Rather than:
Drive to Maple St, turn left'
At Walnut Street, turn right
After Oak Street School, turn right on Oak Street

The first of these (similar to how I hear music), I build a kind of chain, or lattice. It floats freely (usually about a P4 below the absolute pitch level). The lattice is hierarchical, and contains multiple referents. In tonal music, the scale degree ^3 will largely identify a <t> tonic function, and ^7 will identify a <d> dominant. ^4 or ^#4 I hear largely as <dp>, dominant preparation. Knowing (or guessing) the key of the piece, I will visualize a score, and through reverse-engineering, would write out the pitches I think I hear. This appears to be the opposite method used by possessors of AP.

FWIW, my understanding of Chomsky's transformational linguistics is built upon the idea of relationships and hierarchies; language as a form of variation and hierarchies. In (many? most?) Indo-european languages, there are clusters of word types (verbs, gerunds, adverbs,nouns etc), often based around some kind of root that undergoes transformations: to think, thinking, thoughtfully, thought ...), and while the 'absolute' form [well-formed] is learned (with the exceptions), the mind is able to read that:
 Ta-daye eye had gotten inta ma kar n wint duntoun
The translation of this, or machine reading of this, would likely be more difficult.

Jazz, if it is to be more than a stream of notes going by, can be heard as song-form variation, and enjoyed for the inventiveness of the musician. This may also be the case in other 'language delimited' improvisations -- I think of north Indian and Persian particularly, both of which have extensive ancient theories of music.

Many years ago I worked on creating a database with the analysis of Bach Chorales, and very quickly understood how Heinrich Schenker came to his conclusions that western music is about structure, relationships and hierarchy rather than objects -- relationships in a lattice (being a spatial relationship), rather a grid (map) upon which objects are fixed [absolutely].