Dave Stewart's Music Seminar (Part 2)
Man Versus Machine
Part 2: Famed keyboardist Dave Stewart continues his monthly column, designed to get you thinking about your music.
One of the choices facing us late 20th century keyboardists is whether or not to automate our playing — to sequence or not to sequence, that is the question. Sometimes, this is presented to us as a moral choice, particularly by American musicians, who rant on in the pages of magazines such as this one about how they never sequence anything, how they always play all their parts live (no matter how long it takes), and how they spend nine hours a day practicing fantastically difficult scales and exercises, to which end they have insisted that their road managers install Bosendorfer concert grands in the tour bus, lest their precious "chops" become a tad rusty. (The strange thing is, if one can be bothered to unearth the recordings of these hirsute individuals, they can usually be heard banging out extremely simple A minor triads somewhere in the background while the singer screeches about the powers of darkness, or some such nonsense, a paradox I cannot explain — but let's not dwell on that.)
Anyway, to these musical athletes I have this to say: Take your chops and bugger off! The only people interested in technique for its own sake are other neurotic musicians with the same hangup. The rest of us are more concerned about whether the music has any decent ideas in it than whether the keyboard player's little finger can play ascending chromatic scales at the rate of one note per millisecond.
Perhaps the technique-displayer's hatred of sequencers is based in part on confusing sequencing with quantising. Quantising (or 'auto-correction'), as I'm sure most of you know, is a function of sequencers whereby the rhythmic placement of the sequenced notes can be made mathematically accurate. For example, you can 'quantise' all notes to play as exact 16th notes, or shift every note of a chord (which may have been played in a slightly arpeggiated style) so that all the notes sound exactly together on the downbeat. You can sequence music without quantising it, but the over-use of quantisation — along with other deathly habits that tend to make music sound mechanical, over-repetitive, and lifeless — is rife in popular music, and may have helped to give sequencers a bad name.
I personally believe you can use sequencers in an organic, musical way, even with the dreaded quantisation which so offends the musical purists. It's not so much a moral as a musical question — the secret is to get the machine to do what you want, to make it conform to your musical will, rather than it dictate what you can do. There is too much music around where the operating system of the machine has evidently imposed itself on the composition — too many two-bar drum loops, quantised bass lines, parts that repeat over and over again in strict, unwavering tempo and unchanging metre. Before this mechanical sounding stuff takes over the world completely, we need to redress the balance by establishing who is boss... sequencers, like fire, are a good servant but a bad master; the friendly robot that can easily become the murderous juggernaut. (Especially if you lose its manual.)
One problem facing the novice keyboardist is that the sequencer can become too much of a crutch, thereby robbing the player of the opportunity to improve his or her playing. These days, we often have to consider the possibility of working with a 'click track'. These things are hard to play with: they slow down (ie. don't speed up when you do), speed up (ie. brutally ignore your rallentandos), and fail to stop if you make a mistake.
Faced with the prospect of trying to play music in time with one of these things, it is little wonder that nervous beginners prefer to enter their parts into a sequencer, maybe even one note at a time — quantise them, then synchronise the sequencer and the click-track with one of those awfully clever little gadgets that proliferate in the pages of this magazine. Is this such a bad idea?
Well yes, it can be. Given the editing facilities, it's all too easy to copy the part you've played for Chorus 1 into Choruses 2, 3, and 4. The part then becomes both fixed and repetitive, and the danger is that the player will never learn to play that part properly (ie. fluently and in time). Worse still, the player might never consider other ways of playing or developing the musical idea behind the part. Thus the automation process, in the name of 'convenience' (a god to whom we are currently sacrificing large portions of this planet), can have a destructive side, rendering the composition inorganic and standing between the musician and the music.
So much for the dangers, but what of the good side? Luckily for me, I was no novice when I bought my first sequencer in 1984. By then, I had already learned to play a bit, and my ability to compose music was not compromised by the machine. After its acquisition, I still wrote music in the way I had always done — by playing and listening, imagining melodies in my head or sometimes whistling them in a ghastly, tuneless monotone that to me represented pitches, but to anyone else present resembled the noise made by an old kettle! My sequencer enabled me to capture ideas that previously I would have lost, and to store them for further exploration on small disks — this seemed terribly modern to me in 1984. Now, I use it on all my recordings, to a greater or lesser extent, and the relationship between its performance and my performance (in reality, of course, both mine) has become so intertwined that it seems to me a kind of symbiosis — so much so, that I have christened my sequencer 'Little Dave'... (pauses typing for a while, awaiting the arrival of men in white coats with concerned expressions).
Good Lord! I've written over 1000 words without once mentioning a piece of my music! Let's get back immediately to our surgical dissection of my song Grey Skies, which we commenced last month, and see how the two Daves shared the labour of recording some of the parts.
Remember the 9/8 keyboard rhythm that repeats over a 4/4 backing? It is shown again in Figure 1.
After a while, I grew tired of hearing these notes play over and over, so I substituted percussion sounds for the keyboard pitches. By accident, the selection of percussion sounds I had chosen (seven or eight conga and tabla drum sounds) had certain keyboard zones where no sound was allocated, so I ended up with the rhythm shown in Figure 2.
This was accompanied by another keyboard part in the scale of D flat major (also transcribed last month), part melodic, part rhythmic, which was actually improvised and left fairly intact. Both parts were sequenced, and quantised to 16th notes — I felt no sense of rigidity in the quantised versions, whereas the originals had some flams, clunks, and questionable placements which would have caused problems with the other rhythm parts if they had been left untreated.
It was over this background of two keyboard rhythm parts, along with a drum machine playing hi-hat 16th notes, that I improvised the chord sequence which was to become the basis for the song. Last month, I showed you the chord voicings (which include some harmonic movement within the chords) to the chorus — these voicings are critical to the mood of the song and cannot be changed. (I don't believe in the idea of 'substitutions' — this seems to me to be a system devised by tired old jazz musicians who are too lazy to learn the precise chords.) Figures 3 and 4 show the accompanying bass line and string parts respectively, again both were initially improvised.
The bass part was sequenced and quantised (though I took care not to quantise the pitch bends — that always sounds terrible!). As the sound I had chosen for the bass line had a very hard bass marimba sample at the front-end of each note, even tiny discrepancies of rhythm sounded bad, so quantisation was my only option — it sounded fine, though. The strings were played live. Little Dave (my sequencer) can't do everything, and sometimes his timing resolution on unquantised parts (about five milliseconds at this tempo) worries me. Can anyone hear a timing discrepancy of five milliseconds? Who knows? I dunno. Anyway, playing string parts is so enjoyable that I wasn't going to let some cocky little machine do it. The same went for my keyboard solo (Figure 5), which is my favourite on the whole album.
The combination of all these parts gives the song a slightly eccentric feel (especially in the rhythm department), but somehow it retains a lyrical quality. Our sound engineer described it as "an English folk song played by sequencers" — I look forward to hearing your own interpretations!
First published in Keyboard magazine, Japan.
Excerpts from 'Grey Skies' by kind permission of Budding Music © 1990.
'Grey Skies' comes from the CD 'The Big Idea' by Dave Stewart & Barbara Gaskin. It is available via mail order from: Broken Records, (Contact Details).
Dave Stewart is happy to receive your questions, comments, criticisms, and other phenomena associated with the enquiring mind. You can write to him at this magazine or c/o Broken Records.
Feature by Dave Stewart
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!