Something We Know

Artist / Composer
Marc Hannaford
Time signature

Sheet music

Sheet music is accessible for subscribers only. To access all sheet music, take a subscription from only $35. Subscription options.

Samuel Hällkvist and I met at the 2011 International Rhythm Studies annual conference In Sweden. Needless to say, a conference with a focus on rhythm makes for many inspiring talks on various aspects of making rhythm a central focus of composition and improvisation (see also Ronan Guilfoyle's first part of an interview with Steve Coleman here).

The talk I gave outlined various processed by which I had incorporated quintuplets and septuplets into my improvisation language. I presented an analysis of my tune Something We Know.

Samuel recently asked me to write about how my trio worked on playing this music. In our performance we move between making 5:2, eight-note triplets, 7:2 and sixteenth-notes (what I'm going to call 5's, 6's, 7's and 8's from hereon out) the primary rhythmic subdivision, and there are plenty of times where these co-exist. Here I plan to share some of the things we did on order to feel comfortable playing this music.

Let me start out by saying that my primary aim working on this music was not to have it so we all would play the same thing all the time. I enjoy the sound of band members each working on their own ideas simultaneously. Rhythmic exercises were used to create a common understanding of rhythm so that we could aurally recognise each other's phrasing in 5's, 6's, 7's and 8's. My thinking was that if we can each recognise each others phrasing in each subdivision, then we're each free to not play that subdivision while maintaining our place in the form.

The points I'm about the present were not necessarily done in strict order; I didn't demand that we know 1 before moving on 2. We worked on most of them all the time.

So, some the ideas we worked on were:

1. Everyone learnt the rhythm of the melody, both "as a melody" and "as a system."

The "system" that provides the rhythm of the melody is based on magic squares.

The magic square for Something We Know reads:

13   17   21  (total of 51)
17   28   39  (total of 84)
21   39   57  (total of 117)

The total of this square is 9 x 28 = 252

Placing a "gap" of 9 between each line brings the total up to 270, which is divisible by 5.

Breaking up each of the numbers in the magic square into patterns (influenced by listening and transcribing some Karakudi R Mani) yields:

3 3 7    5 5 7    7 7 7   

gap of 9

5 5 7    7 7 7 7    9 9 7 7 7

gap of 9

7 7 7    (3 3 7) x 3    (6 6 7) x 3

If you go through the chart and put accents correlating with these numbers you'll see that each accent lines up with beginning and ends of phrases in the melody.

So, as a trio, we learnt this pattern, first in isolation, and then using the subdivision of 5:2, as in the chart.

2. I did this mainly with the drummer, James McLean, but it was also a useful drill to work through as a band: we picked a relatively simple rhythm that outlines 5 - if crotchets (quarter notes) are a represented by "2" and quavers (eights) by "1," then some patterns we used were 221, 212, 122, 2111, 1112 - played them through the form. At first we would play one rhythm for a whole form before moving onto the next, but as we got more familiar with them we would aim to change every four bars or something similar.

3. We practiced playing 5:2, 5:3 and 5:4 over the form, all together in rhythmic unison. The idea was to keep our place in the form and change harmony as the piece suggested, but move between these three speeds. We organised this in a similar way to point 2, first a whole form, or perhaps multiple form, and then changing some smaller number of bars. We practicing moving between 5:2 to 5:3 to 5:4 to 5:3 (and repeating that cycle), as well as moving between 5:2 to 5:3 to 5:4 (and repeating that cycle).

4. We then ran over points two and three again, but with one or two members of the trio either playing 4 the bar (as in walking, "jazz" time) or a different rhythm based on the quintuplet subdivision, and the other member of the trio improvising. We would nominate any of the polyrhythms or "5-patterns" outlined in points 2  and 3 as the rhythmic ostinato for the "accompanist(s)." Again, we'd do it over the form, and change it each musician's role at regular intervals.

5. At this point we came to notice our tendency to race or drag, mostly drag. Quintuplets in general seem to drag, as players fall back on a speed closer to 4. It's hard to have a specific exercise for this, but I suppose that I can say that I was quite militant in stopping the band whenever I felt we dragged. Not in a nasty way, but I would try and articulate what I think happened in the band that resulted in the drag. Luckily, James, Sam and I are all selfless enough to not get offended during these types of conversations, and we're all good at expressing ourselves without sounding accusative. 

6. Go through steps 2-5 again, but using rhythms based in subdivisions of 6, 7, and 8.

7. Practice combining all rhythms from steps 2-6, through improvisation. In this part, we would often have one person improvise, while the other two members would move between 4, 5, 6, 7 and 8, in various combinations. The distribution of the improviser/time keepers was cycled through the band.

8. The entire time our idea was that, whenever a rhythm came up in improvisation that caused us to lose our place, we would stop and analyse what rhythms were going in that band at the time, and then form some sort of exercise around that. At this stage our philosophy of rehearsal was very much about getting to ideas through improvisation. That way, things stay fresh and in context; we develop our language for improvisation through improvisation, and would only revert to exercises when we came up against something that was completely alien to us. It also keeps the sound of the group flexible, and avoids it sounding too "fusion-y."

9. We practiced improvising together while referencing the rhythm of the melody. The idea here was that each member of the band could make this reference and everyone could recognise it. This also worked as a way to bring the band back together if we were falling apart (when that was not what we wanted) or to begin the tune (I prefer to not count tunes off, but rather have one member simply begin playing and the others catch on).

10. We noticed that the bass figures in bars 6 and 15-16 often caused the band to drag, so we practiced having one member keep a solid 4 going while the others played the rhythm of these figures. Each member took turns playing the figure and improvising.

11. We rehearsed often, and played it in gigs a lot. Thinking back, I know James and I had been playing it from about 2008, Sam joined us around the end of 2009, and we recorded at the end of 2010. I know I worked on these rhythmic drills over piece every day for quite a while, and made a point of committing everything to memory.

I can't think of anything much more right now, but perhaps if I do I'll come back and add more. I hope some of this stuff is helpful. One of the main things I feel I need to convey is how much work this is. I spent a good 2 years working on this tune, but have been working on playing the rhythms contained in this article continuously since 2006. In my experience students often practice some of this stuff for a few weeks, or perhaps a month, and then ask why they can't get it into their playing. Maybe I'm just not so bright, but I remember it was about 10 months of playing 5:2 before I could put it into my solo, and much longer before I felt like it just "came out."

More recently I have started exploring 9's. I started at the beginning of 2012, and am just now (April, 2013) to feel like they're coming into my playing in a way that I like. Just some perspective I suppose . . .

Good luck, and I'd be happy to address any questions in the comments.

This track is from my album Sarcophile. We put a lot of work into this album, and would appreciate your support in buying it. 

You can check it and my other releases out at