(duration ~ 5:35)
Percussion solo for marimba, 3 tom toms and digital delay

To purchase the part for this work, please see the print music catalogues
Program note –
An electronic delay system is employed throughout the piece, serving to reproduce the ‘live’ signal 566 milliseconds (about  half a second ) after it has been played, thereby building a multi-marimba illusion & creating rhythmic counterpoint against the live performance.
The player is requested to play in tight synchronisation with the delayed signal & by moving through a number of tempo changes, different rhythmic effects are achieved.
The use of the delay gives rise to the title, which is a reference to the great Roman dictator, Fabius Maximus who was victorious due to the use of delay tactics in battle.
Fabian Theory was commissioned by Synergy Percussion with  financial assistance from the music board  of the Australia council.

Technical performance notes –

In some of the earlier editions of Fabian Theory there is a missing tempo marking.
Measure 193 should be marked with “Tempo Primo / quarter note = 106”. This tempo is maintined until the next tempo marking at measure 211. (I apologise for the confusion!)

Over the years I have had many emails from percussionists all over the world requesting details of how to manage the technology associated with this piece.
Here are a bunch of hints, suggestions, questions, answers & ideas from other percussionists concerning the performance of the piece.

Working with the delay on stage.
In order to perform the piece accurately, it is important to be able to clearly monitor the signal from the delay unit, so as to synchronise your playing with this signal.
One of the main problems encountered with Fabian Theory performances is feedback generated from monitoring this signal.
If you are using a foldback wedge on stage to monitor the delay signal, chances are it will set up a feedback loop within the delay (the microphone above the marimba picks up the signal from the foldback monitor & begins to feedback. )
You can get around this problem by monitoring your delay signal through headphones (mini earphones are sufficient – & less obtrusive visually).
Make sure you leave plenty of time to set yourself up on stage before the performance – it can often take a while to obtain the right balance through the P.A. system.
Also…..make sure you are able to practice the piece with the delay unit itself, as this is quite difficult & can take a while to get the hang of.

Re: the “LOOP” created from measures 49 -50
With some delays (i.e. the Roland SDE series, which I used when writing this piece, & which are probably now obsolete ), upon the depression of a footswitch connected to the unit, the material that has just been “recorded” into the delay is “held” & repeated. I’ll try to explain further…………..
The delay time = 1 quarter note & the idea of the loop is simply to catch a “quarter notes” worth of the stuff being recorded into the delay at this point, which repeats untill measure 87.
If the figure in measures 49-50 is fed into the delay with the right feedback, this should result in an effect similar to that of 4 or 5 marimbas all playing that sequence of notes – but each part being delayed by a quarter note.
By “capturing” (with the footswitch) the material in the delay at the end of measure 50 – you should end up with 1 quarter notes worth of this pulsating, multiple marimba effect.
One problem that has been encountered by various players is that some delay lines produce a nasty sounding “click” at the moment the footswitch is depressed (which keeps repeating until the loop is turned off!). The “click” can be minimized with practice, but some players have constructed the “loop” as a pre-recorded sample & triggered it (in sync) by some form of midi mallet device (i.e. KAT) . Percussionists the world over seem to be particularly rescourceful with regard to these kinds of issues.
Heres one particularly inventive solution to the “Loop issue” from a percussionist in West Virginia ……

“I recorded a few bars of the ostinato pattern at bars 49-50 into my laptop, then added the delay effect in an audio editing program. Then, I counted the total number of beats the ostinato should last (157 1/4-notes, plus an addtional 1/8th-note), and looped it that many times. (This avoids having to worry about turning off the sampled pattern—instead, it just ends at the correct time.) Finally, I used a DrumKat MIDI percussion controller to trigger the sampled pattern via MIDI (the sample plays back from the laptop). I have the DrumKat positioned right next to the bottom of the marimba, so I can easily trigger the sample with my left mallet.
To turn off the delay effect in bar 49 (and back on again in bar 87) I use a simple footswitch connected to the delay unit’s bypass jack.
I don’t know how others are doing it, but I perform the opening section of the piece (through bar 50) with 2 mallets, switching to 4 mallets when I trigger the recorded sample.
One final note: listening to Michael Askill’s recording was crucial in figuring out the relationships between the delay and the notes played. I’m not sure I would have been able to get it right if I couldn’t hear how it should sound. No criticism intended, but I’m not sure metronome markings alone are adequate. It might be helpful to add some addtional notes on what the “composite rhythm” with the delay should be. For example, you might add a small note at bar 215 stating that the delay—in conjunction with performed notes—should create a composite rhythm of 16th-note sextuplets. Or at bar 211, the composite rhythm should be 16th notes. Just an idea!”

Various questions (& responses) I have received from percussionists concerning the practicalities of performing Fabian Theory…….

Q. – “does the delay remain “on” (sounding) at measure 51, or is the delay bypassed (not sounding)?”

A – With the “loop” now active, the delay effect is bypassed until measure 87, when it starts up again (& the loop is turned off @ measure 87).

Q. – “In measure 87 it says “delay hold off” does this mean turn the delay back on (sounding once again)….if it has been off since measure 51?”

A. – Yes. The “loop” is de-activated & the delay is re-activated.

Q. – I noticed a discrepancy between the score I purchased from you a while ago
and a recording of the piece I have by an Australian percussionist (whose
name is Michael Askill, I believe?). My question regards bar 110, where
the time signature changes from 6/16 to 2/4. In the recording, Askill
performs this as if the dotted-eighth-note in 6/16 = the quarter note in
2/4 time. However, the score does not indicate this (therefore the
assumption is that the 16th-note remains at a constant tempo throughout,
which at this point in the piece should still be at mm=106). If the
1/4-note in bar 110 *should* equal the dotted-eighth-note in the previous
bar, then what happens in bar 195? (Where the meter changes from 6/16 to
3/4.) On the recording, Askill considerably slows the tempo at this point
(back close to the original tempo of 106). I’m happy to play it either
way, and just want to be true to the composer’s original intent. If you
could clarify this for me, I’d appreciate it!

A. – The tempi do in fact change just as Michael Askill played on the recording.
Changing the tempo has the effect of “virtually” changing the note-value of the digital delay.

Q. The “metric modulation” at measure 110 is marked as quarter note =134. If the dotted eighth from measure 109 equals the quarter in measure 110, shouldn’t the tempo be closer to quarter note =141? More importantly, what is the intended composite rhythm? Very slight changes in tempo (134 vs 141) have drastic effects on the composite through the delay. It seems the Askill recording settles around 134, but that wouldn’t be the “correct” metric modulation mathematically.

A. I see where you are coming from – but its not intended as an actual metric modulation – simply a strict tempo shift.

Q. I’m really curious about the transition between measure 214 and measure 215. The quater note =90 tempo makes sense with the previous marked tempo of quarter note=134 in terms of metric modulation, but if quarter note =134 is incorrect based on the fact that the dotted eighth at 106 becomes the new quarter note in the next section, is the quarter note =90 also incorrect?
Most importantly, what should the composite rhythm through the delay be at measure 215? One person on the FAQ suggested that the composite at measure 211 should be sixteenth notes (which doesn’t make sense if a metric modulation truly occurs) and that the resulting rhythm at measure 215 should be sixtuplets (which doesn’t quite seem to work based on the relationship between the digital delay’s 566 miliseconds and the rhythmic material being played at 90 bpm).

A. Again – there is no actual metric modulation intended here – just another strict tempo shift.
The effect of playing the triplet patterns through the delay at measure 215 at a tempo of quarter note = 90 should result in a constant flow of triplet sixteenths