On Chord Symbols: Discussing Steve Coleman’s Cell Notation System

A shorthand system for depicting chords is a useful element in music composition, improvised music [spontaneous composition], and in music analysis. Chordal instruments might use these symbols to craft an accompaniment, composers might use these symbols to map form as changes in harmony, improvisers might use these symbols to determine melodic choices for tunes with which they are not yet aurally familiar. The deficiencies of traditional chord symbol notation are clear: no provision for notating chord voicing [the specific order and intervallic spacing of tones within a “chord”], presumption of tertian [built by stacking thirds] harmony, presumption of major/minor (or as has been common in jazz styles, lydian/dorian) modality, no (conventional) provision for the motion of inner voices or upper structures, amongst others. As a pioneer of new concepts in music composition, analysis, notation, and beyond, Steve Coleman is commendable in his efforts to develop a system for efficiently notating specific harmonic structures [two or more tones sounding simultaneously]. 

Steve Coleman’s Cell Notation System

Steve Coleman’s system follows these basic guidelines:

  1. The lowest pitch in a cell is notated as a traditional pitch class.
  2. Additional voices within a cell are notated numerically, as semitones above the lowest pitch.
  3. For additional voices, a “+” symbol indicates one octave higher than otherwise notated, “++” indicates two octaves higher, etc.
  4. A horizontal line divides multiple cells within a single symbol [a “composite” cell].
  5. Moving voices are indicated by a horizontal arrow and a new semitone relationship to the lowest pitch.
  6. Slashes indicate the rhythm of those moving voices.
  7. Traditional chord symbols, when mixed with cell notation, appear in brackets.
  8. [Generally, it appears that additional voices are notated in their actual vertical position, regardless of octave displacement. However, this is conjecture and is not directly addressed by Coleman in his original post.]
[Cell notation example: courtesy of http://m-base.com/] More information and additional detail can be found at http://m-base.com/scores/cell-notation/.


Coleman’s system efficiently addresses most of the deficiencies of traditional chord symbol notation (that efficiency in communication is reflected in the innovative nature of his recorded music). Nevertheless, this cell notation system can be built upon, with consideration for efficient conventions of improvised music notation and chordal accompaniment texture tendencies. Particularly, the following arenas leave room for improvement:

  1. Semitones as a means of analyzing harmony is somewhat clunky. Yes, this could afford the opportunity to gain freedom from pre-conceived notions towards the major/minor (lydian/dorian) system and tonality in general, but the practical application seems contrary to the goal. If the goal is spontaneity and improvisation, yet overwhelming familiarity with another system (We are used to seeing “maj 7” and “#11” and not “11” and “6+”. Intervals rather than semitones have been the means of harmonic analysis for hundreds of years, harkening back to figured bass notation.) precludes such spontaneity and improvisation, will the innovation ever become pragmatic?
  2. Chordal instruments playing improvised music tend not to limit themselves to playing just one voicing per given chord, unless the chord is transient or the tempo is fast. If Bill Evans had played just one voicing (rather than variations on a voicing type) over “So What”, it would be a different tune. Traditional chord symbols afford some ambiguity regarding voicing in order to suggest a wide palette of voicings which might fit over that given harmony (implied “horizontally” by melody instruments, bass, etc.). However, there is little control over that ambiguity. What if there were an option for specifying voicing types, while allowing freedom in the actual voicing itself?
  3. Several voicing types are pervasive and have earned the right to their own notational “shortcut”. Quartal voicings, rootless, drop-2, A/B/C/D voicings, etc. If voicing is specified within the chord symbol, it would behoove the designer of an efficient system to account for the prevalence of these voicing types.
  4. Slashes as a means of communicating rhythm (especially in musical styles where odd meters abound) might leave room for ambiguity where it is not intended. In the example above, could each slash represent a quarter note? An eighth note? A half note? Upon closer investigation, one can easily divide the length of the whole note by the four slashes provided and ascertain that they represent quarter notes. However, considering that the length of a given chord lasting less than a single measure is already an implication, this layering of implication on top of implication may hinder fluid improvisation.
  5. The proposed system of integration with traditional chord symbols could be visually confusing.

A Modified System

  1. Cells are enclosed in brackets, to distinguish from traditional chord symbols.
  2. The lowest pitch in a cell is notated as a traditional pitch class.
  3. Additional voices within a cell are notated numerically, as modified diatonic intervals above the lowest pitch. Within the span of two octaves (generally, the largest possible span of a single hand on most given instruments), major intervals above the lowest pitch in a cell are indicated numerically, as intervals. Non-major intervals above the lowest pitch in a cell are indicated with a “flat” or “b” symbol before the numerical interval. Raised or “sharp” intervals should not be used, except in the case of #9 and #11.
  4. A minus (“-“) symbol indicates the given interval is played one octave smaller than indicated. A plus (“+”) symbol indicates the given interval is played one octave larger than indicated. This might become advantageous when a #9 is most easily read, but is intended to be just 3 semitones above the lowest pitch in the cell.
  5. A horizontal line divides multiple cells within a single symbol [a “composite” cell].
  6. Moving voices are indicated by a horizontal arrow and a new intervallic relationship to the lowest pitch. Diamond notehead rhythmic notation indicates the rhythm of those moving voices.
  7. The notation “[4]” may be used as a shortcut to indicate that any quartal voicing can be applied to a traditional chord symbol. The notation “[344]” (where the 3 is positioned vertically in line above both 4s, to indicate a third played over a stack of fourths) after a traditional chord symbol indicates the use of “So What” voicings.
  8. The notations “[A] –>”, “[B] –>”, “[C/D] –>”, etc. can be placed before a ii-V-I progression using traditional chord symbols to indicate the use of those specific LH rootless voicings. (The use of an arrow serves to distinguish these as specific voicing types, as opposed to cells of one or two pitches. Additionally, it serves as a visual indication that the notation applies over several chords.) The use of a dashed line terminated by a slash in brackets, “[/]”, can be helpful to extend the use of A/B/C/D voicings over a larger section of the composition.
  9. The notation “[d2]” can be placed after a traditional chord symbol to indicate the harmonization of a given melody with drop-2 voicings. Additionally, an alternating chord quality for harmonizing non-chord tones can be suggested within the brackets, i.e. ” Fmaj7 [d2 dim7]”. (The lower case “d” within brackets is helpful for discerning the drop-2 notation from a cell composed of the pitches “D” and “E”.)
Revision of the above shown cell notation example using the modified system.

Example of traditional chord symbol with multiple possible voicings of one given voicing type.

Example of drop-2 harmonization.

The processes with which we compose or notate do not/should not have to be static. Nevertheless, musicians (and jazz musicians are more guilty of this than any) seem fearful to discuss in an authentic manner the technical aspects of innovation in their creative process, almost as if this would “give away their secret”. To those like Steve Coleman, Vijay Iyer, Jason Moran, and others, thank you for using public forums as a think tank to increase access to the open-source development of our art form and inclusion within the community. In this age of information, “artists ought to be writing about what they do and what kinds of procedures they go through to realize a work; what their presuppositions in making the work are, and related things. If artist’s intentions and ideas were more accessible to the general public, I think it might break down some of the barriers of misunderstanding between the art world and artists and the general public. I think it would become clear the extent to which artists are just as much a product of their society as anyone else with any other kinds of vocation.” (Adrian Piper)



“Artists Ought To Be Writing”, Adrian Piper

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s