Introduction to Free Counterpoint. ( or Bach Style Counterpoint ) by Glen Halls All rights reserved.

Size: px

Start display at page:

Edmund York
5 years ago
Views:

1 Introduction to Free Counterpoint. ( or Bach Style Counterpoint ) by Glen Halls All rights reserved. The First and most important distinction between strict and free counterpoint is the point of departure. The point of departure in strict counterpoint is the given line, or cantus firmus. The point of departure in Free Counterpoint is the harmonic progression, driven by harmonic function. Let us go way back to the beginning, and look at one of our first cantus firmus melodies. In strict counterpoint our first impulse is to find an independent countermelody using consonances. We are not thinking at this time about dividing or parsing up the line into smaller segments. Rather, we are looking at longer range melodic shapes. This linear continuity is desired.( Note, we are not thinking about harmony, just about consonances, BUT in fact we ARE tacitly thinking about harmony. That is, if one is only permitted perfect and imperfect consonances, AND one avoids parallel fifths, the range of acceptable 'chords' (though we will only have two of three triad members. ) Are I I6 IV, IV6, IV64, ii6, V V6 V64, ii6, and ii64, and the occasional vi vi6 and iii ) So, in free style counterpoint we would look at this cantus firmus and come up with a harmonic progression which could support it. For example:

2 So, the lower line ( not intended to represent counterpoint at this time ) shows the likely roots of the following implied chord progression, one of several possibilities. I IV I V V7 I V I In Free-Style counterpoint, Harmony is prior. That is, ones first impulse is to convey - through melody and/or arpeggiation- the prevailing harmony in a given measure ( or whatever the harmonic rhythm happens to be ). Before we continue it should be noted that the basic principles, if not strict rules, of strict counterpoint are still and always true and in effect. But we now must add a disclaimer,.... one should follow the rules "UNLESS THERE IS A COMPELLING REASON NOT TO FOLLOW THE RULES." There are two such compelling reasons. 1) for expressive accent, and 2) to convey the harmony and harmonic progression. In fairness, it takes a few years of working with harmony to get a 'feel' for 'what is enough' 'what is too much', what is best. This cannot be rushed and one should not expect to understand harmony, understand it to the degree that one feels comfortable in using harmony for creative, expressive ends, until one has worked with it for several years. So, does anything go? Are there no rules? What are we supposed to do? Well, the early 17th century contrapuntalists had an answer for this. Generally speaking, free counterpoint evolved as accompaniment to well known melodies such as chorales, and also, composers often IMPOSED A THEME OF SOME KIND, a short theme which was repeated many times in the course of a work, transposed once or twice, given to different voices. Sometimes these themes were referred to as 'subjects', sometimes as 'Fugue subjects', but this is a third and important element of free counterpoint. 3) Part of the craft and expressive intent is to find new and interesting harmonic contexts in which to place a repeating 'main theme'. In other words, the rules might be relaxed in order to 'fit in' the theme

3 Let us review some of the general rules of strict counterpoint and consider the extent to which they will still hold in free counterpoint. 1. Avoid parallel 5ths and 8ves. Yes, this is always a good idea, but sometimes in order to make a certain chord progression and to fit accompaniment to a set them, you may have to use a parallel fifth or 8ve. If this is the case, we still usually avoid parallel 5ths and octaves between the outervoices. Anyway, breaking this rule will be a rare occurrence. 2. Dissonances on Weak beats only. This is a hard one for the novice to grasp. Remember why we avoided placing dissonances on weak beats in the first place. We tried to avoid accent. The most 'natural' and least intrusive place to put a dissonance is on the weak beat. This conveys a sense of flow and ease- in short in gives an impression of a long, effortless phrase. We were adding counterpoint to chant- we didn't want to break things up into little one bar units. The weak beat dissonance rule help us achieve this effect. So, if we want to convey ease, and long phrases, we still tend to follow the strict rules of dissonance treatment. Let me express this another way. In the given cantus firmus above, if I wanted to preserve the sense of a long, unbroken melody, and basically no real harmonic change ( i.e.. We want to convey melody over a drone, that's all ), then I would stick with strict dissonance treatment. BUT, we are now considering 'breaking up ' the line into suggested chord progressions, sometimes chords are 1/2 measure long, sometimes basically a whole measure long. If we are focussing on harmony, - if we don't mind breaking things up into bar-to-bar units, then go ahead, put dissonances on strong beats. BUT REMEMBER, by measures end we must be sure that the harmony is clear. Also, this will create tension, expression. We may want to 'save up' our expressive accents for particular places, perhaps climactic measures, or really intense text we are trying to set. 2a) Accented passing tones are OK * ( but harmony must still be clear) read on. 2b) accented neighbor tones are OK *(but harmony must be clear) 3) No leaping into a dissonance.? In free counterpoint you may leap into a dissonance. This is termed an appogiatura. Don't do it too often, and try to resolve in stepwise contrary motion to a 'chord tone'.

4 ( Note, I am carefully shifting language from 'consonances and dissonances' to 'chord tones' and 'non-chord tones.'. In jazz theory we often talk about this distinction in terms of 'lower partials' and 'upper partials', but that is a specific context. ) 4) No leaping over barlines? Well in general this is a good idea. Certainly we don't ever want all voices leaping over barlines. Among other issues, any leap - in choral and string writing- presents the danger of poor tuning or a missed note, so we tend to minimize it in these contexts. Remember also, that leaps across a barline will always create a big downbeat accent. Now if harmony is prior, this is not a problem. 4a) But something happens in free counterpoint. We begin to key into root motion in the bass. At cadences, for example, V- I root motion is totally acceptable ( as opposed to the strict stepwise contrary motion to the tonic in 16th century counterpoint ) Leaps over barlines in the bass is OK, especially if harmony dictates that a clear root and clear harmony is required. Leaps in the soprano are also OK, but less frequent. ( Often it is used to convey a kind of 'out of phase' effect.) Nonetheless, leaps over barlines are generally less frequent and steps, and a good balance of the two is desired. Remember, you are now free to fill in leaps with stepwise passing tones, really at any rhythmic subdivision. 4b) Sometimes Free Counterpoint, or Bach-style counterpoint, is also called 'instrumental counterpoint' or keyboard counterpoint. This is an important concern. On a piano or harpsichord or any other fixed pitch instrument,( especially with the advent of equal temperament at about the same time ) leaps will not result in poor tuning or missed notes. For example, leaps abound in keyboard preludes and fugues. But when Bach writes for voices ( not soloists, but for choirs) he stays pretty close to the strict counterpoint rules, and minimizes leaps. ( Except... when following a theme..., sometimes bach 'translates' instrumental style for voices.. But it's very hard to sing. )

5 5. What about suspensions? Suspensions were pretty carefully controlled in 16th century counterpoint, but in free counterpoint we open it up. Suspensions are very very expressive, and this is what they wanted- to express emotions. ( 'Doctrine of the affections' ) The rules in the old style basically were: a) suspensions must be prepared by a consonance ( prepared meaning 'the same note' tied. )b) suspensions were resolved by stepwise descent only. c) suspensions were only ( mostly ) used at the prevailing pulse level ( half notes generally ) d) suspensions were described in terms of interval, not chord, and there were basically only 3 kinds; the so-called 7-6 and 4-3 for the soprano voice and the so-called 2-3 in the bass voice. Let's look at these now. a) " suspensions must be prepared by a consonance "? This is generally still the case in free-counterpoint, but, so long as harmony is clear, and perhaps you feel this to be a 'really intense' phrase, you may tie from a dissonance. This is comparatively rare, though. b)" suspensions were resolved by stepwise descent only."? No, this is no longer the case. Mostly stepwise descent prevails, but ascent is also very common. NOTE, we now shift our way of thinking from 'dissonance to consonance' to 'non-chord tone to chord tone. A suspension is a non-chord tone resulting from a tie in a previous beat or measure. It will resolve, most of the time, to a chord tone, - basically root, third, and fifth. It could be up or down, whatever seems the most natural- the strongest tendency. ( Note, sometimes suspensions are deliberately resolved to a chord, different than the expected one, but rarely. This is a kind of deceptive effect, perhaps dictated by theme, or some other melodic impulse. ) c)" suspensions were only ( mostly ) used at the prevailing pulse level ( half notes generally )"? This one goes out the window. Harmonic rhythm (the rate at which chords change) is no longer just the bar or 1/2 bar level of

6 strict counterpoint. Harmonic rhythm may be anything from two measures down to the 8th note, but mostly it is still at the bar or 1/2 bar level. In any event, suspensions will correspond to harmonic rhythm. Harmony is prior. d) " suspensions were described in terms of interval, not chord, and there were basically only 3 kinds; the s o-called 7-6 and 4-3 for the soprano voice and the so-called 2-3 in the bass voice. "? This terminology is no longer used, and in retrospect might even be considered counterproductive and confusing. Suspensions are heldover non-chord tones which resolve to chord tones, and are determined by the current root of the current harmony. Any chord in a key ( well the viio is a toss-up )will have a root, a third, and a fifth. To the root we may apply either a 2-1 or 7-8 suspension. To the third we may apply a 2-3 or 4-3 suspension. To the fifth we may apply a 6-5 or maybe even a #4-5 suspension. Here the numbers refer to degrees of a major scale above a given root, and not to interval ( though they are often the same ) When I say a 6-5 suspension, say in a bar of C, or tonic, I really mean and A resolving to a G, and nothing else. So, what is meant by 'harmony must be clear'? Well, it is generally agreed that the character of a chord/triad may be conveyed by just the root and the third. The fifth, unless it is altered, is considered optional. What this means for free counterpoint is that 80% of the time, the bass voice will involve motion between a root and third. For the duration of a given chord, in order to be 'clear' we must have at least the root and third sounded between the two voices, and if there is time we bring in the fifth and/or double the root. Now here's the thing- if you are using accented dissonances ( non-chord tones) in a line, and especially if you use accented dissonances in both voices, it is very important that the listener hear roots and thirds in the remainder of the measure. It really is a kind of 'feel'- if you upset the balance of clear harmony by using accented dissonances, you then will likely want to compensate with your other pitch choices- to be even more conservative- to ensure that the harmony is clear. Well, as an introduction that is enough information to think about for now. The next step will be to look at some real music from the period. Intro to Free Counterpoint. Part Two. by Glen Halls

7 This lecture and the techniques of free counterpoint are based on a simple analysis of Bach's Menuet from Partita #1. Have a look at this first phrase: What the heck is going on? Doesn't look like any Cantus Firmus here- and yet it is clearly in two parts. Also seems to be quite a bit of leaping around. Melodic direction changes are also more freely handled. Maybe- but it is a question of focus as well. 1. Harmonic Progression and Harmonic Rhythm. I... I6 I... I6 V6...I V6... V I..I6 IV...IV6. vii. I V...V7 Pretty terrible HTML rendering here. Sorry. But really we see a progression using I, V, IV, and vii. The harmonic rhythm, or the speed at which chords change, is basically two measures, sometimes one measure,

8 rarely beat to beat. ( This will change in every piece ) Basically we see TWO types of Bass motion, which results in PROLONGATION of the suggested harmony. Arpeggiation between triad tones, and passing motion between Root Position and the first inversion. Bb: I(I6) I (I6 ) In this second example, V6 (V) V6 V, motion from V6 back to V is embellished by the accented neighbor tone, and the accented passing tone. Two more examples of V: We note that 8ve leaps, (mostly root to root) are OK. The fifth is left off in the first bar. In the second bar, the third is left out of the bass arpeggiation, which suggests it would be strongly stated in the upper part.

9 This looks like vi in Bb. Here we have the root and fifth arpeggiation, but can we start with the fifth? We don't do this too often. There should be a reason, perhaps better voice leading, or more likely, it was supposed to sound like a I6 ( shares the same bass note) but then moves to a different tonal area. - kind of like a pivot or preparation. This last examples is from an Ami7 chord, probably headed to Dmi or something. ( I cant give roman numerals at this time- it looks like part of a sequence). But we note a NEW ARPEGGIATION, one which involves the 7th. Sevenths should basically be prepared either by a common tone or by passing motion, and should resolve so-called 'directly' down to the F in the next measure. Seventh chords are much less common than triads in the free counterpoint style, outside of sequences. I wanted to briefly talk about the barlines. In most instances we try for step motion across a barline. However, our main concern is being clear about the harmonic change, if any. If we have to get to a root, we get there, even if voice leading is poor. a) In this Bach example we note that arpeggiation is freely continued across barlines, ( i.e., if the harmonic

10 rhythm is more than a bar- the barline really isn't there. b) We see descending fifth, ascending 4th, motion across barlines. This is always strong. ( Note, not just V - I, any 5th bass motion is strong.. It definitely accents the downbeat. This suggests we want to notice the harmonic change. c) We also see peculiar leaps across barlines which do not seem to Conform to any model. Such as this: Which really means: So, we do not always 'resolve' directly. The Strong Bb root 'prolongation' is still in effect. Yes, there is the literal 4th leap over the bar, but the ear also hears the stepwise neighbor motion Bb to A, strongbeat to strongbeat. A Further Discussion of an Introduction to Free Counterpoint. ( Part Three) Leslie-Anne Ducette. All Rights Reserved. No Duplication Without Permission

11 Our introduction to free counterpoint was modeled after the opening few measures of a Bach Menuet. We noted that in terms of underlying harmonic progression the phrase basically moves through I, IV, V, and maybe vi. We also noted that most barlines could be explained as a prolongation of the basic harmony by moving from Root position to First Inversion, and on occasion arriving at 1st inversion through a seventh in 3rd inversion., Or simple arpeggiation of the entire triad. Basically, we could say the left hand was harmony converted into simple melody. Now, in terms of applying a RH counterpoint to this basic lower voice outline, we balance several concerns. 1) Harmony and harmonic progression must be clear. In practical terms this means we desired to spell out the prevailing harmony as fully as possible, all other concerns being equal. You might say simply, if the bass features the root and third, the fifth ( and possibly the seventh ) should be prominent, or at least present, in the upper voice. 2) The Thematic character of the piece should be preserved. The thematic character of this piece in general terms is 8th notes in RH, quarter notes in Left Hand. It also ( RH )features a kind of two bar alternating cycle; descending arpeggio in one measure, ascending line in the next. This does not hold throughout, but it is true that there will usually be a certain thematic character, a 'hook', or pattern which defines this piece as different from all others. A third thematic signature of this piece is the offbeat pedaltone, discussed below. ( For example, the fact that there are Bbs on the 'and' of 1, 2, and 3 in the first two measures. This is a hook, a theme. ) Now what happens with the coincidence of RH and LH notes in this example? This does not always have to the be case, but 80-90% of the time whenever notes are 'stuck' together, they are consonant so to speak. In other words, they are bound by relations of Old First Species counterpoint.

So, this is why, outside of more complex developing variation form like a FUGUE or something where you really do have to use the so-called 'subject' as the point of departure, I suggest we begin with

12 So, this is why, outside of more complex developing variation form like a FUGUE or something where you really do have to use the so-called 'subject' as the point of departure, I suggest we begin with the Bass. Bass prolongation, via motion from root to first inversion ( except for the V42 chord) ( V V V42 I6 - I ) So most of the time we will fill in the motion from I to I6 with a passing tone. THEN, we have some options, maybe even obligations, in terms of the RH note to go against each bass note, including the bass passing tone. In most cases we will use a consonance. Shortly we will look at ALLOWABLE DISSONANCES ( which is one thing which is really different in this period) but first we will attempt the most conservative, most predictable and definable solutions for some of these basslines. ( What my former teacher Ian McDougall used to call the "Kaspar Milquetoast" solution. ) So, in converting normal or stationary, homophonic harmony to free counterpoint ( harmony expressed through line) we first note the basic, underlying, sometimes called 'middleground' harmonic progressions- not worried about specific things like inversions at this point- just I V IV III etc., and whereabouts they occur. If given a simple harmonic progression such as I IIV I V I, we might consider expressing them as motion from root to first inversion, and vice versa. Another approach might be to employ one or more '42' type of chords. With the exception of V42, this means using an applied harmony, that is, a chord from another key, so really for two beats you are in another key. Typically the '7th' is approached by step from above, i.e. from a rearticulated root. Like this.

13 In the second example we see the applied V42 as an approach to, expressed in the home key, a IV6 chord. Generally speaking one doesn't use more than two applied harmonies in a phrase unless it that phrase is transitional in nature, i.e., in which we want the sense of home key to be loosened a bit. What is left for the bass to do? Two things really; a) staying on the same note via neighbor motion ( or indeed as a true, dissonant, pedal) or arpeggiation. Neighbor motion is easily understood, and there is nothing wrong with it. For example, to prolong a bar of I we create a C B C bassline ( probably, but not necessarily I -V6 -I or I - vii - I ) But remember, part of our interest here is to create melodic interest- to create a 'counterpoint or second independent line to the upper part, so neighbor motion is not preferred. Arpeggiation. Arpeggiation is common in situations where the same bass note, displaced by 8ve, lands on the downbeat of one and the next bar. Sometimes this is termed 'register transfer' ( Schenker ) Note, this 'same' bass note will

14 most often, but not always, be part of the same chord. I to I, is normal I6 to I6 is normal, but it is often 'hip' (Bach is Hip- Chopin is Hip- Schumann is Hip) to change the context of this same note, for example, how about root of biii (in minor) landing on bvi6, Or perhaps Root in one measure become a 42 in the next. NOW FOR THE RH Now it is very very important to keep in mind that the notes you are looking at are acting on different levels., Not all the bass notes will require a right hand counterpart. Also, depending on speed and rhythmic placement of the bass, some notes will be understood as prolongational or 'filler' only, and not require a counterpart. Generally speaking, weak beat bass notes will tend not to require support by first species counterpoint. (Weak beats at whatever pulse level or levels you feel to be active at a given time. ) Generally speaking, our bread and butter is root and third. For root in the bass, look for third in the upper. For third in the bass, look for root in the upper. For seventh in the bass, - a bit of a toss up. Most of the time the third against the seventh is preferred, though root against the seventh is also common. NOTE. How can you say that a 3rd and 7th tritone ( in most cases) is allowed? It is not consonant?!! Well here we go. In free or harmonic counterpoint, yes we speak of vertical consonance, but also of 'chord tones'. If the seventh is felt to be a chord tone then it may be struck with the third. It is not considered dissonant when it is part of an implied V7 or viio7. Now, one should try to resolve these tones properly by step, and also try not to leap into this seventh. ( Actually, a root-seventh leap is not uncommon, nor is an approach by arpeggiation, up through 3,5, and 7, but generally we try to use it as part of a descending stepwise line. ) Likewise, I seem to be indicating that a root- 7th, second cluster is also consonant or at least acceptable. Well use this one with great care. It is not uncommon, but one would typically find the third either sustained in another voice, or if in two part writing, stated just prior or just after the root. The third should be in the ear somehow. The Fifth, if supported, will either take the third, we might consider it as root of a completely new chord, like V, and give it the third of that chord. In effect we are making an 'interpolated' or mini progression, a small internal cadence, also called 'local motion'.

15 So, and THIS IS A VERY IMPORTANT POINT, your main resource for structuring free counterpoint lines IS.. LIMITED ARPEGGIATION- which really means A SINGLE MOTION BETWEEN TWO, maybe three ) CHORD TONES. In the bass, we call this motion to and between inversions. WE TEND TO CONNECT This simple, simple motion between chord tones, with direct or ornamented stepwise motion. ( Sometimes with abrupt leap to the third chord tone- implying a kind of compound melody ( soprano and alto 'lines' implied by one part ) but this is not required. ) To be even more blunt about it, 80% of the time counterpoint may be reduced and described simply as an embellished VOICE EXCHANGE.

MELODIC AND RHYTHMIC EMBELLISHMENT IN TWO VOICE COMPOSITION. Chapter 10

MELODIC AND RHYTHMIC EMBELLISHMENT IN TWO VOICE COMPOSITION Chapter 10 MELODIC EMBELLISHMENT IN 2 ND SPECIES COUNTERPOINT For each note of the CF, there are 2 notes in the counterpoint In strict style