copyright © 2001 by Larry J Solomon
In Ex A the black notes and slurs of the foreground are eliminated. Beams are consolidated, with some note stems combined on one beam; this is not essential, but simplifies the appearance of the graph.
A simplification of the foreground domains is achieved by grouping harmonic cycles and noting longer harmonic prolongations. Thus, the middleground domains reduce to tonic-dominant-tonic; e.g., the first cycle is I V ii6 I, prolonging tonic as a common sequence and therefore reducing to tonic.
In Ex B, all repeating notes are reduced to their first occurrence (shown in Ex A with the broken beams). One exception is the retention of the bass pedal, G, in m24-31. The reason for retaining this is that without the second G the appearance of its prolongation may be lost or obscured.
In Ex C all non-domain notes are blackened (filled in). Domain notes remain white.
In Ex D beams and slurs are re-introduced to show step progressions to and from domain notes, using the same principles as constructing the foreground. All white notes within each domain are beamed together, and the black notes are shown as step progressions linking the domain notes. Stems are added connecting with the white note beams to show the stepwise voice leading. Whenever possible a separate beam is used for each voice, but some are combined on a single beam, especially when voices move in parallel motion. Slurs, just as in the foreground graph, show step progressions between domains.
The bass voice may leap, but only when no step motion exists there. These leaps are normally fourths and fifths. Some notes are reinstated from the foreground in order to show essential step progressions. Only passing notes may be reinstated; no neighbors are necessary. Some back-tracking is involved here. The reinstated notes are circled in Ex D.
Duplicate lines, normally in parallel octaves, are shown above or below the staff with dotted lines; e.g., from m1-16 the line E-----D-----C is found both in the soprano and tenor voices, indicated ST in the Ex D. One of these is less essential and will be eliminated in the middleground. Generally, outer voices are kept as essential, and inner voices are eliminated when they are duplicates. If inner voices are involved in duplication, the upper voice is maintained on the graph, e.g., m29-35.
Any notes that do not have a stepwise connection are marked for deletion with parentheses or an X. Only the high G5 in m7 is so marked.
Subsequent construction of the middleground graph follows the same priciples as for the foreground. Doubled lines are reduced to a single voice as explained above. Non-stepwise movements are deleted except in the bass, but ancillary bass notes are eliminated.
Back to Schenkerian Primer
You are Visitor No: