I am a small-group leader who likes a rhythm section to consist of guitar, bass and drums. What's unusual about that, you ask? Nothing - except for the fact that I play piano. Pianists are supposed to like playing with horn players, not guitar players. The reason for this is that, since the piano and the guitar are both chord-playing instruments, there is a real danger that there can be too many chords at the same time, resulting in too thick a sound that takes away from the melody and the solo. Or, so goes the thinking of many players I know.

But, I can't help it. I like guitar players - I like the fact that they can rhythmically groove in a way that horn players just cannot do. This may be due to the fact that a guitar has a built-in capacity for being strummed. Things that can be strummed usually groove well with the beat, because of all the different possible ways that they can be strummed. Both guitarists and pianists are practiced in the art of strumming and, consequently, groove in a kindred way. Because of all this, I like to play in quartet situations with a guitar player.

However, before all you guitarists go out and form hornless bands, you should be aware of a certain problem that will present itself : Two chordal instruments in one band can produce a sound that should normally be avoided, a sound I call middle register density. What I mean by this is that in a quartet situation, each member of the group usually operates in a distinct and separate registral zone, or timbral zone. These are not purely registral zones, nor are they purely timbral zones. They are zones of location, i.e. when a listener hears four instruments he is able to distinguish where the separate sounds are in relation to one another. The listener can keep track of a lot of musical information if these zones are properly placed. For example, the upright bass, naturally, occupies the lowest registral zone. The drums cymbals (and perhaps the snare drum) are in the highest zone, and the bass drum and other drums are somewhere near the upper end of the lower register (in the case of the floor tom) or the lower end of the high register (in the case of the other toms.) The pianist is usually right in the middle of this registral timbre, as is the guitar, although they can both play in extremely wide variants of the middle.

If the pianist and the guitarist are both playing chords, there is usually too much of the middle registral timbre for human ears to take. (There are times when a creative musician may want that specific sound, so to have two chordal instruments play chords at the same time is not strictly verboten. It's just recommended that you avoid it if you want to keep getting paying gigs.)

The challenge, then, is that while I desperately want the sound of a guitar to complement my playing, I also want the guitarist to be sensitive and discrete in their use of chords. I am sure they want the same thing from me as well. And the solution is very simple: both the pianist and the guitarist should stop thinking of themselves as just chordal instruments. We should think of ourselves as one-note instruments who can occasionally play some other notes at the same time. The music we make if we adopt that attitude will be much more interesting for both us as musicians and for our audiences. It will be more contrapuntal.

It is easy to think how we might apply this philosophy to the head of a typical jazz tune. Here are some suggestions:

• The guitarist plays the melody, while the pianist plays the chord changes
• The pianist plays the melody, while the guitarist plays the melody
• The guitarist is tacet, letting the pianist play both the melody and the chords
• The pianist is tacet, letting the guitarist play both the melody and chords
• The pianist plays the chords in the left hand, while doubling the guitar melody with their right hand
• The A section is comped by the guitarist, while the B section is comped by the pianist