This blog began in the fall of 2008 as a forum for the discussion of issues relating to classical music composition. Although my primary objective has always been pedagogical – posts are generally aimed at students and others wishing to improve their composition skills – feedback from all interested readers, irrespective of their background, is welcome via the "post a comment" link at the end of each post.
I used to get my composition students at Memorial University's School of Music to write their own weekly blog posts reflecting on things that went well and things that were challenging in the process of creating their own compositions; links to these student blogs can be found at the bottom of the right-hand column, grouped by course number (Mu3100 is "Introduction to Composition," and Mu4100 is the follow-up to that course).
Enjoy your visit, and please leave a comment on anything that interests you, even if you are not taking my courses! I receive automatic notifications of all comments, irrespective of how old the original post is, so if you leave a comment, I will see it.
I am a composer of music for soloists, chamber groups, orchestra, voice, and electronics, as well as a guitarist and professor of Composition and Theory (since 1992) at Memorial University of Newfoundland, Canada.
I teach Composition, 2nd-year Music Theory (chromatic harmony and large form classical music), Renaissance Counterpoint, Baroque Counterpoint, and Music Technology. I have also taught Electronic Music, Orchestration, 20th-Century Techniques, Music History, Ear-Training, and Classical Guitar.
I welcome feedback from all non-spam bots! If you have any suggestions of topics you'd like me to cover in future posts, please let me know!
Having finished the chapter meant I could focus my energies on composing music once again, and the first project I gleefully tackled was a solo piano piece inspired by Domenico Scarlatti for my friend and colleague, Kristina Szutor.
I wasn't quite sure of how to go about writing a contemporary piece based on Scarlatti's music at first, but, as with so many initially-perplexing ideas in life, once I began to delve more deeply into it I discovered there was much to work with. I soon became quite excited by the project and ended up enjoying the experience very much (which doesn't always happen when I compose).
So much so, in fact, that when I was finished I decided to write another. Part of my rationale in composing a second work was a familiarity with musicologist/harpsichordist Ralph Kirkpatrick's claim that the great majority (almost 400) of Scarlatti's 555 sonatas were written in pairs1 (and they are often performed this way), so it seemed somewhat anomalous to be composing an azygous Scarlatti-influenced sonata. ;)
As I subsequently learned, however, more recent scholarship (notably by W. Dean Sutcliffe, in The Keyboard Sonatas of Domenico Scarlatti and Eighteenth-Century Musical Style, Cambridge University Press, 2003) has challenged Kirkpatrick's claim,2 so it really came down to my feeling that having completed one, it would be fun to write another.
My introduction to Scarlatti's music came through learning guitar transcriptions of his sonatas, and one of the most popular of these amongst guitarists is K. 208 (L. 238) in A major (listen to a performance by Leo Brouwer here). Two of the most pervasive characteristics of that sonata are the steady, repeated, quarter notes in the left hand, and the flowing melody in the right hand with frequent syncopations; these ideas formed the basis of Domenico 1°. The working title for this piece was originally "Domenico Daybreak," and perhaps this will make sense if you hear it (click the "play" arrow in the audio player below; if it doesn't work, click here instead).
The great majority of Scarlatti's professional life was spent on the Iberian peninsula, most of it in Spain, where he had five children, composed the majority of his single-movement harpsichord sonatas, and became familiar with flamenco music, the influence of which can be heard in some of his sonatas.
I had therefore planned Domenico 2° as a kind of fantasy based on flamenco-like scales (for example, phrygian mode with the possibility of raised third and seventh degrees), but I decided to make it an even-more overt homage to Scarlatti by quoting four bars of his Sonata in B minor (K. 27, which I transposed to A minor) that use a chord progression known as a "fandango," much associated with the music of Spain: Am - G - F - E, in 3/4 time. This quoted passage is also remarkable for the use of hand-crossings (left hand crossing above the right), an uncommon technique for the time it was written, and I based several other sections of my composition on Scarlatti's fandango material as well.
1. Ralph Kirkpatrick, Domenico Scarlatti, (Princeton University Press, 1983), 143: "385 sonatas [were] indubitably arranged in pairs… not intended to be performed separately… The real meaning of many a Scarlatti sonata becomes much clearer once it is reassociated with its mate.… The relationship between the sonatas of a pair is either one of contrast or of complement. The sonatas that bear a complementary relationship to each other may share a certain overall unity of style or of instrumental character or they may be composed in the same harmonic color. In the contrasting pairs, a slow movement may be followed by a fast; a simple movement, generally slow, may serve as an introduction to a more elaborate; or an elaborate and concentrated movement may be followed by a simpler and lighter movement, for example a Minuet, which serves as a kind of Nachtanz.↩ 2. "Not a single detailed commentary exists in support of any particular pair. Instead we find gestures towards opening thematic connections or an outlining of the sort of broad relationships defined by Kirkpatrick." (p. 368)↩ 3. Sectional repeats are observed in this recording of Domenico 2° (but not in Domenico 1°).
Below is an index of most of the 2008-09 blogs posted (and is an update to an earlier blog index). I omitted entries that I didn't think would be very interesting or relevant, mostly because they contained composition class business, such as reminders of deadlines, concert congratulations, order of class presentation, etc.
Feel free to browse these — you may find some that give you ideas about composition techniques, or that contain useful things to think about when composing. They are loosely organized by topic.
1. Originality and Quality of Initial Musical Ideas 1.1. The quality of ideas may not matter very much in determining the quality of the complete composition that emerges from them; and 1.2. The degree to which these ideas are original may not matter very much.
→ Here is an "octatonic" scale, also known as a "diminished" scale because a diminished seventh chord is formed from every second note. The intervalic pattern consists of alternating half- and whole-steps (or vice-versa):
Here is a short waltz based on this octatonic scale (click on it to enlarge):
Octatonic Scale Waltz:
→ Here is a Hirajoshi scale:
Here is the first phrase of the waltz, this time based on the Hirajoshi scale:
Hirajoshi Scale Waltz:
→ Here is one form of a blues scale:
… and here is the first phrase of the waltz based on that blues scale:
Blues Scale Waltz:
Discussion:
We tend to limit ourselves to the use major and minor scales if composing tonal music, but there are thousands of other scale possibilities that have unique nuances and harmonic implications. If you have fun (i.e., experiment) with even a few of them, you may discover that every different scale gives your compositions a slightly (or even radically) different feel.
Of these, there are a number of commonly-used alternatives to major and minor scales, such as anhemitonic (which just means "no semitones") pentatonic scales (5-note scales whose pattern can be found by playing only the black notes on a piano, any of which can be the tonic), the blues scale (there are different permutations, but all are derived from the form of the black-note pentatonic scale beginning on Eb, or La-Do-Re-Me-So-La), the Hirajoshi scale (another pentatonic scale, from Japan, but unlike the previous penatonic scale this one has two semitones (which means it is hemitonic): La-Ti-Do-Mi-Fa-La), or the octatonic scale (used in some Russian folk melodies and by some Russian composers such as Stravinsky and Scriabin, as well as by Bartok, and also used in jazz).
You can make up your own scales and modes; Messiaen created scales with repeated patterns that he called "Modes of Limited Transposition," such as:
Tone-Semitone-Semitone-Tone-Semitone-Semitone-Tone-Semitone-Semitone (which he called his third mode):
or Semitone-Semitone-Minor Third-Semitone-Semitone-Semitone-Minor Third-Semitone (which he called his fourth mode):
Messiaen's Modes of Limited Transposition are all based on repeating patterns within equal subdivisions of the octave, but in making up your own modes or scales, you do not need to be limited in this way. You could, for example, create a scale with a repeating pattern that spans a major sixth. After four such pattern repetitions, you would have spanned three octaves and the overall, three-octave, pattern would then repeat. But the pitch patterns in each of the three octaves would be slightly different. Such as this, for example:
The following scale is a mirror around the pitch F#, but you could also create a scale with few or no pattern repetitions in it:
Another approach, suggested by my friend and former colleague Dr. Scott Godin, is to construct a few (2-3) atonal chords that you wish to use as the basis for a composition, then construct a scale containing all or most of those notes. You can then use that scale to create additional harmonies if you wish.
Once you choose or create some scales with which you want to work (play), you could make charts of the triads and "seventh" chords formed on each scale degree. However, you need not build these chords in the same way as is done with major and minor scales (in thirds); you could form chords based on unusual patterns, such as chord I comprising the 1st, 2nd, 4th, and 5th scale degrees; chord II comprising the 2nd, 3rd, 5th, and 6th scale degrees, etc. Remember that there are no rules here, so you can form chords in whatever way you like.
All of these approaches create harmonic and melodic sound worlds that are distinct from major/minor tonality, but which can yield some fresh and attractive results.
Consider this: "Tonality" refers to far more than mere scales and the chords formed from those scales; it mostly refers to the relationships between the notes and chords in the scales. When using alternative scales and modes, you may notice some relationships between notes and chords that seem "natural" to you, and you are of course free to use them, but bear in mind that often what seems "natural" in these scales and modes are the aspects that are most closely tied to tonality, such as dominant-tonic relationships or leading tones. It can be fruitful to explore note and chord relationships that are not similar to the more familiar aspects of tonality.
Incidentally, the objective, when using something like a blues or Hirajoshi scale, is not necessarily to create blues music or Japanese music (although it obviously can be if you wish); it is to write compositions that may sound to the listener as though they are related the kinds of music from which the scales originated, but with your own unique spin on them. For example, my blues-scale waltz fragment above does not sound particularly bluesy, because the F#-F-F#-G in the 3rd bar is not characteristic of blues music. More bluesy in that bar would have been F-Eb-F-G, or even Gb-F-Eb-G, because the F#/Gb in that scale is usually treated as an inflection of F or G.
And finally, don't forget that the concepts of "non-harmonic tones" and "modulation" can be borrowed from tonality and applied to any music you compose using these alternative scales.
Any chord that sounds like it might have some distant relationship to tonal music (like Vb9 with an added 13th in an unusual inversion) should not function as it would in tonal music (i.e. the chord above should not “resolve” to anything resembling a I or vi chord).
The chords should sound connected in some sense; avoid giving the impression of a random series of unrelated sonorities. On the other hand, the sense of connection need not be obvious.
There should be a gradual increase in harmonic tension to a specific point, roughly 61.8% of the way through the progression(!), followed by a corresponding decrease in tension to the end. This proportion (61.8 : 31.2, which is the same as 1.618 : 1.0 or 1.0 : 0.618) is called the golden mean, or golden ratio, or phi, and is related to the Fibonaci Series (click links to read more about these).
No chords can be re-used, although they may be re-voiced. Note that the same collection of notes can have greater or lesser tension depending on the voicing.
Vary the number of notes in each chord to suit the desired tension level; avoid using four-voice chorale-style texture exclusively.
Explore the possibility of using different registers on the piano, either simultaneously in the same chord (hands widely spaced apart), or as a means of contrast from one sonority to the next, or as a means of contrast for subsequent variations.
Use a very slow tempo, in order to allow the listener’s ears to take in the uniqueness of each sonority before moving on to the next one.
Do not introduce dynamics yet; the increase in tension should be principally effected through harmonic means, not through dynamic control. Perform your chord progression without dynamic inflections.
Your primary composition tools should be your ears and instincts; when comparing chords x and y, which one feels like it has greater tension? However, if you study each chord you write to determine its intervallic content (do a Best Normal Order analysis and a Vector analysis, if you like), this information may be used when constructing new chords, or when altering existing chords for future variations. Each chord should sound “right” to your ears. Atonal music does not have to sound unpleasant, but almost all music is based on principles of tension and resolution, and your challenge is to do this within an atonal idiom.
→ When you play your chord progression for the class, we will each assign a "perceived tension level" score to each chord, notated on a sheet of paper, where 0 = no tension, and 10 = highest possible tension. This will be followed by a short discussion in which class members will be asked to identify the chords of greatest and least tension, and discuss any general trends with regards to tension in the chord progression (for example, you may find that the tension level increased substantially from chords 1 to 4, then took a dip for chord 5, then stayed the same for chord 6, then spiked (increased substantially) in chord 7, etc.).
Week 1 will be spent creating the chord progression.
Week 2 will be spent fine tuning it (based on feedback received in class), introducing rhythmic values to each chord (not all whole notes; try to create a sense of “timelessness” or unpredictability through notated rubato), considering the possibility of repeating a chord for rhythmic purposes or of re-using a chord, adding dynamics and articulations, marking in phrasing, and creating a melody.
Weeks 3, and 4, will be spent creating a new variation each week.
Week 5 will be spent creating the final copy of the score to be handed in, as well as a recording. The recording is normally made during the class recital, date TBA.
Note that the work you do each week does not have to be handed in (until the composition is complete), but it does have to be played in class.