Music
Home > Music > Composition > Tips for student composers > Composition method (Stephen Lalor)
Composition method
Stephen Lalor
Composer
Stephen Lalor writes about his composition process. To find biographical
information and a list of works composed by Stephen Lalor, go to http://www.amcoz.com.au/composers/composer.asp?id=103 
In this section I'll take you on a "nuts and bolts" tour of how I composed a piece for violin and piano. I've chosen this piece because it's on a small scale in terms of instrumental forces and thus easier to deal with in this format. It is also a short piece and was created from very simple elements. It provides a useful tool for discussing such issues such as:
- How do I start?
- After I've come up with the initial ideas, where do I go next?
- How can I make it all hang together and avoid jumping from one idea to another?
- How can I introduce twists and turns to make it interesting but not disjointed?
The piece, Dance (for violin and piano), is the second part of a piece called Prelude and Dance. It was composed mainly when I was a post-graduate student at the Tchaikovsky Conservatorium in Kiev (then part of the USSR) and has elements of the eastern European/Slavic music I was studying and playing then, amongst other things.
1. Before examining the piece in detail, listen to the whole of Dance and follow the score.
Print
the score l 1 l 2 l 3 l 4 l 5 l 6 l 7 l 8 l 9 l 10 l
Listen
to the complete composition (3':08").
The whole piece is based on extremely simple building blocks,
one of which are the use
of 5ths, used both melodically and chordally.
In the left-hand introduction they are used melodically, like a riff:
From bar 7, they are used chordally:
This is combined with the little descending idea found in the right hand of the opening bar (see above), the use of glissando and deliberate key "clashes", plus a strong formal outline. And that's about it!
A good thing to remember if you have trouble finding a starting point for your own composing: you don't need necessarily need dramatic themes or riffs.
2.
Intro (bars 1-6). Play and follow the excerpt from the score.
Bars
1-6. Page 1.
The Dance movement is meant to be a complete contrast to the
naïve, simple Prelude which precedes it. The Dance is strongly rhythmic, frenetic and a little weird in a cartoonish way.
The piece really kicks off at bar 7, but is preceded by an introductory
six bars played by the piano, linking the Dance to the preceding Prelude. The left hand of the piano is in 5ths (E and B, an interval
which is central to the piece). This gives the piece a cartoon-like
childishness, against which the right hand plays an equally simple 3-note
line in the "clashing" tonality of Eb (Eb, Db, Bb). This starts
brightly but winds down like a clock or toy running out of energy and
is one of two contrasting thematic moods used. The use of very simple
"cells" or ideas of varying tonality and rhythm against each
other, to create an unusual sound, is one which comes up in my music
quite a bit. It may have come from the influence of Russian composers
such as Shostakovitch and especially Prokofiev.
3.
Section 1 (bars 7-20). Play and follow the excerpt from the score .
Bars 7-20.
Page 1, 2.
0'.08"
- 0'.20"
The other main idea of the Dance is heard immediately in bar
7. This cell came about while I was experimenting on mandolin and guitar
during breaks in rehearsals for concerts and recordings. You'll probably
realise that it begins on the lower 2 open strings of the violin (G,
D), which are the same on the mandolin, a 5th apart. The effect is just
like "power chords" on guitar. Try playing it and you'll see
what I mean. Playing loud and accented at 110bpm gives a really rough
energy to this idea, while establishing the tonal centre (G) and the
folk dance character.
Rather than have the two instruments play this section through together
as one, I engaged them in a bit of question-and-answer dialogue, the
violin doing the "answering" with a bar (see bar 8) of country
or east-European fiddle sound, also based on 5ths. Bar 10's glissando
slide from another 5th (C# - G#) is something which crops up in my music
for strings, again probably due to my being involved with folk playing.
After the chromatic 3/4 bar 13, which breaks up the very regular 2/2 feel, the section is essentially repeated. Throughout the piece you'll find what seem like regular 4-bar phrases in 2/2 made irregular by a final 3/4 or 3/8 bar. Again, the music I was involved in at the time (Russian/Ukrainian and other) often had a similar feature.
Don't be afraid to repeat something that sounds good and feels like it needs repeating.
4.
Section 2 (bars 21-47). Play and follow the excerpt from the score .
Bars
21-47. Page 2, 3.
0'.21"
- 0'.45"
The introduction is recalled in the G-D ostinato of this section and
in the violin's line (now alternating between E and Eb tonality). The
much lighter, more playful nature of this section (leggiero) makes
a contrast to the previous section, as does the syncopated right-hand
piano part, again featuring 5ths, but in this case descending, with
longer values.
Everything holds together because of its organic nature; it is derived from earlier material. It's not necessarily the material itself, but what you do with it, that matters.
After a descending Eb run in 3/4, the preceding 6 bars are essentially the same, but the right-hand piano 5ths now ascend, as does the 3/4 pattern in bar 34. This builds up the energy level which was lowered at bar 21, and the instruments explode into a reprise of the opening from bar 35.
5.
Section 3 (bars 48-77). Play and follow the excerpt from the score.
Bars
48-77. Page 4, 5.
0'.46"
- 1'.31"
The
propulsion of the previous section seems to be building to a climax
at this point, so I decided to throw in a surprising change of style
and energy. As if a pumped-up balloon suddenly deflates, the violin
starts a long section of woozy, semi-improvised double-stopping glissandi.
This contrasts with the previous beat-driven section. It all takes place
on the A and D strings because I liked the timbre of these strings for
this glissando passage, as opposed to the more cutting, thinner E string.
The violinist is given general indications of where and when to move,
but the rest is up to him/her. The pianist is asked to play short, interjecting
parts, with a duller, thudding timbre created by plucking the strings
of the piano with the fingers. This thudding sound is reminiscent of
the home-made plucked basses I sometimes heard playing music in the
Carpathian mountain style of eastern Europe.
By bar 61 this apparently anarchic section is reeled in by brief, spluttering
returns of the 5ths pattern heard earlier, while the violin holds on
the same pitch for many bars, a kind of upper drone. The piano becomes
more regular and repeated in a riff-like way, while the violin slowly
rises in pitch, like a clock or toy winding up to full strength until
it plunges into a new section by bar 74.
6.
Section 4 (bars 77-98). Play and follow excerpt from the score.
Bars
77-98. Page 5, 6, 7.
1'.32"
- 1'.52"
The violin now takes a back seat to the piano. I think it adds a lot
to a piece if the musical interest is shared around. The violin glissando
rises to long notes at each new phrase, the second phrase adding a trill
(bar 81), then double-stopping creates a different texture and timbre
from bar 84. Throughout, the violin gradually rises in pitch, upping
the tension.
Meanwhile the piano has been playing another cartoonish part, recalling
the earlier 5ths accompanying figure, but moving down every second note.
This makes it slightly different but still connected to the earlier
material, an important point in maintaining the sense of the piece.
If I'd introduced something completely different, the piece would have suffered from too many unrelated ideas; in other words, it would have been a hotchpotch rather than a piece of music which "hangs together" convincingly.
The right-hand part from bar 77, while apparently new, is also derived from the opening right-hand idea (see bars 79-80) and the bar 7 "riff" (see bar 78).
Why pitch it so high? Two reasons, one an artistic choice, the second a practical consideration. Firstly, I wanted to create a cartoonish, slightly "cracked" effect (which the very high, fast piano notes provide). Secondly, it would have been lost in the violin part if it had been an octave lower.
At bar 91, to further broaden the scope, the right hand plunges from very high to the low bass range. This is a surprise to the ear, and also sets up the climax of this section, emphasised by the right hand playing sweeping ascending glissandi while the left hand settles on C# against D.
For a 2-minute composition, the piece could have just about ended here. See how it would have sounded by going now to bar 145, the start of the final section, or even bar 157.
7.
Section 5 (bars 98-144). Play and follow the excerpt from the score.
Bars 98-144.
Page 7, 8, 9, 10
1'.53"
- 2'.37"
Instead of ending here, at a point where the piece appears to have climaxed,
I decided to add another twist to the piece.
An emphatic quaver statement of the "riff" from bar 7 in octaves
and parallel 5ths (bars 98-101), followed by a contrary motion working
of the same idea (bars 102-103) plunges into a new section from bar
104. The quaver momentum continues (bar 104) in the staccato ostinato
bass C#'s, with a very gruff sustained G# (the interval of the open
5th again) as a drone above. The tempo has dropped, and from bar 108
the volume switches to piano, heralding a more introspective
section.
It is not always the case, but I find that it is often a good idea in such an energetic piece to build up the tension and energy, then bring it down near the end and have another build-up. It seems to increase the impact of the ending. I often don't do this consciously, but I can see it in other pieces I've composed, and in numerous examples by other composers. Scriptwriters in film and theatre use a similar technique.
Like
a volcano preparing to explode, the ostinato bubbles along with syncopated
staccato major 7th interjections from violin and right hand piano, in
unison. If played well (i.e. "tightly") these chords add to
the tension of the build-up.
Low open 5ths are again used in the metrically uneven section from bar
122 on, with the violin interjecting with snatches of bar 1 (e.g. bar
126) and bar 8 (e.g. bar 132). Increasingly angular and discordant chords
occupy the right hand as the bass ostinato rises in pitch almost imperceptibly
over many bars. The volume also increases, as do the jittery violin
interjections, all combining to reach another climax at bar 144.
Many unsuccessful pieces get stuck in ostinatos that don't go anywhere. By playing around with the metre, having other ideas come in against the ostinato in upper parts, and gradually changing the notes of the ostinato itself, one can avoid this stagnation.
8.
Section 6 (bars 145-end). Play and follow the excerpt from the score.
Bars 145-end.
Page 10.
2'.37"
- end
You could call this section a coda, and I could have come here straight from the end of section 4. Another emphatic re-statement of the bar 7 riff suggests a bravura finish, but after only two bars the creaky opening of the piece is recalled in very high register, the violin emphasising the deliberate clunkiness by playing pizzicato. The split personality of the piece as a whole is summed up in this coda. After two bars it switches back to the emphatic quaver figure, again for only two bars, before a very high and discordant version of the opening bars returns. A crashing piano chord cluster brings the Dance to an end.
Reflection:
What would I do differently now?
The short answer is: nothing. I had the advantage of tossing the main
ideas around in my head for quite a while and working on the piece with
a violinist as we prepared for a concert of other violin and piano music.
Thus I tried out things as we went along.
I don't tend to write many drafts. I basically think it through before I write, any modifications being minor. As you can hear on the recording, many of the 3/4 bars (e.g. bar 13) were originally 3/8, but proved difficult in performance, so I changed them back to an earlier version.
There is no real right or wrong way to write music, but there are common pitfalls to avoid, and I've tried to point these out. The important thing is to write a piece that "sits well" as an entity, not just as a string of half-baked ideas. This is true of any music, from Handel to Hendrix. I write out of what's in my head, which is what I'm playing and experiencing now, plus all the influences of my past, not trying to make something which fits into the musical fashions which come and go.
It
is interesting for me to look at the range of compositional techniques
I employed in the piece, considering I didn't deliberately set out to
use any in particular. I suppose this is the benefit of studying and
playing loads of music of different styles over the years: when you
go about creating something new, these techniques come out of your subconscious
as tools for producing your own work.