Introduction to Harmony:

==========================

From "Harmony Simplified":

Harmony is the logical connection of chords. It takes root in melody, as it is
viewed as the melodic movement of several parts simultaneously. Melody is the
logical and pleasant movement of a part through different notes. Every note in a
melody owes its aesthetic effect to harmony. By a note's harmonic relation, we
determine its function. [Overture series, blah, blah, blah...] Our diatonic scale
is found from the series of overtones.
I. There are only two kinds of clangs: overclangs and underclangs; all
dissonant chords are modifications of overclangs and underclangs.
II. There are only three tonal functions: tonic, dominant, and subdominant.
The essence of modulation is the change of these functions.

From "Harmony (Schenker)":

Tonality is the key to music. Scale steps (harmony) should be separate from voice-
leading (counterpoint). Two basic concepts: "prolongation" and "compositional
unfolding" [of chords].
Voice leading is the means by which a chord may be unfolded and extended in time.
Much composition in the 18th century was voice-leading with embellishment. It's an
error to interpret a bass line harmonically that is composed contrapuntally.

From Schoenberg's "Theory of Harmony":

Harmony should be more than just how, but also what and why. Action will always
succeed where deliberation fails.

From "Structural Functions of Harmony":

Writing parts above a figured bass is too easy, harmonizing a given melody is too
hard. Both are wrong. A composer should invent harmony and melody simultaneously.

From Piston's "Harmony":

Music theory follows practice--it is a deduction of how music was composed, not how
to compose music. [Of course, for someone who wants to compose in the same manner
that music has been composed, this works just as well.]
Observations of common practice will be presented without any explanation of
aesthetics.

From "General Principles of Harmony":

[Not a textbook for traditional harmony, should already know tonal harmony and set
theory.] Harmonic relationships can be divided into groups by when they become
noticed.

Scales and intervals:

=======================
Blah, blah, blah. If you're reading a textbook on music theory, you should already
know the scales, and the difference between the melodic, natural, and harmonic
minor modes. (It's a raised seventh in the harmonic minor, and a raised sixth AND
seventh in the ascending melodic minor, btw. Creates a leading tone. Never use
the melodic minor when descending--use the natural minor.)

Triads:
=========
The combination of two or more intervals makes a chord (i.e. 3 or more pitches).
Common practice used intervals of a third most of the time (hence "tri-ad"). A
triad with the root lowest is in root position. A triad with the third lowest is
in first inversion. A triad with the fifth lowest is in second inversion.
The triads formed on a given scale differ in not only pitch, but also sonority.
[And different keys sound different because of the differnece in timbre of the
instrument affecting the sonority.] A major triad has a major third and perfect
fifth. A minor triad has a minor third and perfect fifth. An augmented triad has
a major third and an augmented fifth. A diminished triad has a minor third and a
diminished fifth. Augmented and diminished triads are dissonant, while major and
minor triads are consonant.
much music in 18th and 19th century was based on four-part writing (voice leading).

Triads in major and minor scales:

-----------------------------------
Major scale: I ii iii IV V vi viib
Harmonic minor: i iib III# iv V VI viib
Melodic minor: i iib III iv V* VI viib

*Without using the raised seventh degree of the harmonic minor scale, V would be v,
but V is always used in both the harmonic and melodic minor.

You almost never hear music without the influence of melody on the harmony. Chords
are usually spaced with wider intervals on the bottom (following the overtone
series). When the three upper voices are as close to each other as possible, the
spacing is called "close position." Otherwise, it's "open position." In close
position, the interval between tenor and soprano is less than an octave.

Harmonic progression:
=======================
Contains two aspects: which chord goes next (and type), and how to connect the two
chords. More important than an individual chord's sonority is the relation of the
roots, and their function on a given scale.

Common root progressions:

---------------------------
Chord: Is followed by: Sometimes: Rarely:
I IV, V VI II, III
II V VI I, II, IV
III VI IV II, V
IV V I, II III, VI
V I VI, IV III, II
VI II, V III, IV I
VII* III

*VII is rarely used in root position because it sounds like an incomplete dominant
seventh chord.

Moving the root down a fifth or up a fourth releases tension. Root by a step
changes harmonic color. Root moving by a third has a softer effect. While the VII
is rarely used as-is, the II of the minor scale (which contains the same notes as
the VII of the relative major) may be used in root position quite freely because
the II doesn't lead to the tonic like VII does.
The third triad of the harmonic minor contains an augmented fifth. Usually double
the third of that triat (the dominant of the scale, E in A minor). Each
progression possesses certain rhythmic values.
Now the question is of connecting the chords (i.e. voice leading); what is usually
sought is smoothness.

1. If the two triads have one or more common tones, keep the common tones in the
same voice. Move the other voices in small intervals. Exception: for II-V, move
the three upper voices down (contrary to the bass).
2. If the triads have no common tone, move the three upper voices contrary to the
bass. Exception: in a V-VI progression, the leading tone moves up to the tonic,
and the third is doubled.
Application of these rules will ensure smooth (though boring) progressions. Rule-
breaking can add interest and is most noticable in the soprano line, which is
instinctively heard as the melody.
When the root is repeated, it's a good idea to change the other voices. A change
from open to close or vice-versa is an effective way to change the soprano note.
You can sometimes omit the fifth (having three roots and one third). Omitting the
third leaves the chord sounding empty.
Doubling the third or fifth instead of the root can also make more notes available
to the soprano.
Dividing notes to add ornamentation, or holding repeated notes also adds variety
and melody.
A good melodic line uses mostly conjunct (stepwise) motion, with disjunct
(skipwise) motion only used for variety. To many skips causes angularity rather
than flow. Bass can have more skips than alto or tenor. Should also make use of
inversions to make the bass smoother.
Three types of relative motion: contrary (two voices move in opposite direction),
oblique (one voice stationary while other moves), and similar (two voices move in
same direction). Similar motion on the same interval is called parallel motion.
Two voices moving in parallel are less melodically independent, and may be
perceived as the same line. Parallel unison, fifths, and octaves should be avoided
at all costs (well, only when they're salient). Parallel thirds and sixths are
common. Parallel fourths should only be used when supported by a third below.
Parallel dissonant intervals occur every now and then because of nonharmonic tones
and irregular resolution of dissonant chords.
Relative motion should be planned to sreserve the voices' independence. Max
independence comes from contrary motion. Oblique motion also provides good
contrast. Similar motion shouldn't be used too consistently.
Be careful when approaching a perfect octave or fifth through similar motion,
especially with skips in both voices. Generally only move up to an octave when the
upper voice goes by half-step, acting as a leading tone.
The tritone was commonly viewed as an "evil" interval in Baroque and early
Classical music. It was commonly softened or avoided whenever possible. The
tritone was, for the most part, acceptable in a V-VI progression with the leading
tone (third of the V) in the alto or tenor voice.
When two voices more upward in similar motion, the lower is not supposed to move
higher than the position just left by the higher. Same vice-versa.

Tonality:
===========
Tonality is the organized relationship of tones in music. The relationship implies
a central tone around which all other tones gravitate and support. Modality is the
choice of tones within a scale.

Dorian mode: w-h-w-w-w-h-w

Phrygian: h-w-w-w-h-w-w
Lydian: w-w-w-h-w-w-h
Mixolydian: w-w-h-w-w-h-w
Aeolian: w-h-w-w-h-w-w (natural minor)
Major: w-w-h-w-w-w-h

Each scale degree has a function in tonality. Dominant and subdominant lend
balanced support to the tonic. Those and the tonic are also called the tonal
degrees. Mediant and submediant are called the modal degrees because they strongly
suggest mode, rather than tonality.
The supertonic is usually treated as the dominant of the dominant, but it usually
becomes absorbed into the harmony of the subdominant triad, and is sometimes used
as a substitute for the subdominant.
The leading tone strongly indicates the tonic melodically, but is usually absorbed
into the dominant chord.
Too much stress on the modal degrees gives the impression of a different key, being
interpreted as the tonal degrees of another scale.
The dominant has the strongest effect in tonality. The ambiguity of the function
of a single chord is useful for modulation; chords become far less abiguous when
heard in a progression.
Presence of the tonic chord isn't necessary to establish a key so long as the
dominant and subdominant are balanced (including the supertonic). [Write down as
many tonal progressions of three chords and describe their affective
characteristics for later consultation as needed.]
Differences in spacing, doubling, and choice of soprano tone are ready variations
of a chord progression. Change of mode is another easy variation that doesn't
change tonality. In fact, major and minor modes are almost always interchangable.
Chords of the other mode may even be used in progressions of a given mode (e.g. III
in a major progression instead of iii). The VI chord of the minor mode is
frequently heard in a major progression.
Ending on a major chord when the preceeding section was minor is called a Picardy
Third (i.e., instead of i, use I), or Tierce de Picard. Large-scale compositions
(such as symphonies) that have the first movement minor will usually have the last
movement in the parallel major.
The chromaticism resulting from the interchangeability of the modes is what led to
the development of the chromatic scale.

Harmonic Rhythm:
==================
The first beat of a measure should receive the rhythmic stress only when the music
calls for it.
Rhythm of music comes from two sources: melodic and harmonic. Harmonic rhythm can
be derived from melodic rhythm by picking out the roots of chords.
When everything is subordinate to one melodic line, there is melody and
accompaniment. Accompaniment frequently lacks rhythmic interest so as no to
detract from the melody.
Frequency of change and rhythmic quality of change are the two main features of
harmonic rhythm. When harmony changes rapidly, the music tends to be restless,
while widely-spaced changes result in breadth and relaxation.
Sometimes, however, there can be a complete lack of harmonic rhythm when, for
instance, the same chord may be used for an entire section.
Long and short are analogous to heavy and light or strong and weak. Imparting
rhythm to a chord with the same root is NOT harmonic rhythm. Agogic accent should
be used according to whether a progression is trong or weak.
A weak progression goes from weak or light to strong or heavy. In general,
progressions where the root moves by a fourth, fifth, or second are strong. Moving
the root up a third is a weak progression. Down a third is neither definitely weak
or strong.
A bar line should be placed between chords of a strong progression and not between
chords of a weak progression.
Dissonances and their resolution can be weak or strong. Harmonic rhythm is
determined by root movement and may be confirmed or contradicted by its polyphonic
superstructure.
Dynamics are not part of harmonic rhythm, but should be used to confirm or
contradict the naturally perceived accents of a rhythm.
Tempo should be taken into account when determining passing chords.
Pedal tones are usually heard as melody over a single root.

There are four ways that syncopation may be used:

1. Melody against harmonic rhythm
2. Can be heard only be analogy with preceeding measures
3. Both melodic and harmonic rhythm may go against previously established
rhythm
 4. Harmonic rhythm may go against established rhythm while melody continues
(opposite of No. 2)

An anacrucis may be strong to a weak down-beat, or it may be rhythmically weak.

The Structure of the Phrase:

==============================
A phrase of music doesn't usually contain many changes of harmony. The length of a
phrase can vary with tempo (avg. is 4-8 measures). Phrases can artificially be
made longer by sequence or deceptive cadence.
A phrase may be constructed on static harmony, but there should usually be a good
reason for it. The other extreme can happen, too, and sounds rushed in a fast
tempo, but sturdy in a slow tempo. Most phrases should show a balance.
Harmonic rhythm is very important: a regular pulse is common, but some sort of
rhythm is better.
Phrases don't have to start with the tonic. A phrase may start on either an upbeat
of the downbeat. It's customary to establish tonality in the first two or three
chords (i.e. with a circle progression up a fourth or down a fifth).
The end of a phrase is a cadence, usually either authentic (V-I) or half-cadence
(__-V).
If the final chord of a cadence is rhythmically stressed (i.e. agogic or metric
accent), the ending is masculine. If the rhythmic stress of a cadence is strong--
>weak, then the ending is feminine. [Complaints regarding sexism in the
terminology can be sent to /dev/null]
The harmonic progression of a phrase must balance unity and novelty according to
the purpose of the phrase. [Alan Belkin's books are all about balancing unity and
novelty.]

Harmonization of a given part:

================================
You'll almost never have to do this except as a technical exercise or to harmonize
the melody of a folk song. The basic process is to work backwards. Note that each
figure in a melody may suggest any number of possible chords using those notes.
It's all a matter of picking what seems like the best choice, as if composing the
segment of music yourself. [Of course, Schoenberg hated this entire business of
harmonizing a given melody, citing it as too difficult and fundamentally wrong.]
Determination of key depends on the cadence.
When a melody moves by a skip, it's usually best to use the same harmony for both
notes. However, harmonic rhythm must be taken into account, too: if the skip
occurs where a strong progression would go, a strong progression probably goes
there.
Sustained notes could indicate either that there should be a progression with the
sustained note common to both chords (should the harmonic rhythm require it), or
that the melody has moved to another line.
Then it's a matter of exhaustively picking out all possible chords for a given
chord tone. Finally, pick out which chord works best, starting with the cadence.

Modulation:
=============
Staying in the same kay for a piece of much length is boring. Modulation to a new
key relies on the ambiguity of single chords. There are three steps in modulation:
The composer must establish a tonal center, the composer must establish a new tonal
center, and the listener must be made aware of the change.
If the phrase containing the modulation is not the first of a piece, the cadence of
the previous phrase may establish the initial tonality.
Next, the composer should choose a pivot chord, which is common to both keys and is
preferably not the dominant of the second key. The cadence of the phrase typically
establishes the new key, thought there may be other strong progressions in the new
key.
Technically, all keys are related. it's just a question of how closely they're
related--and that depends on how many tones they have in common.
Changing modes only became populat in the 19th century. Usually, a major key and
its parallel minor are very distantly-related. Later, though, they were considered
two aspects of the same key. The relation of two keys should be thought of in the
same way as the tonal function of a chord in a given key.

First inversion:
==================
A triad in first inversion is also called a sixth chord due to the sixth interval
between the root (which is now sounded highest) and what's usually the third
degree.
Regarding doubling with a first-inversion triad, there are two rules:
a) If the bass is a tonal degree, double it
b) If the bass is not a tonal degree, double a tonal degree
A first-inversion triad is lighter and more melodic than its root-position
counterpart. A first inversion triad allows the bass to move stepwise where it
would usually move skipwise.
The I6 chord is extremely useful to provide relief from the finality of the tonic
triad in root position, especially where it must follow the dominant or preceed it.
The first-inversion supertonic triad with bass (subdominant) doubled is cadential
when following a root-position tonic. It's also preferred in the minor mode
instead of the diminished II.
III6 is almost never alone--it's usually only temporary and can sometimes
substitute for the dominant, which it's often formed from.
IV6 can be used freely as a substitute for IV when weight must be relieved.
Placing the leading tone in the bass as with V6 gives the bass strong melodic
significance.
VI6 is similar to III6. VII6 is usually used when passing from I to I6.

Nonharmonic tones:
====================
Nonharmonic tones are, of course, melodic in origin.

Passing tone:
---------------
A melodic skip may be filled in with the intervening tones, diatonic or chromatic.
They are unaccented normally. Passing tones used in multiple voices increases
testural complexity and may obscure the harmony when used too frequently. In the
minor mode, a passing tone of the sixth or seventh degree should be derived from
the melodic scale (i.e. chromatically raised one semitone when ascending, left
alone when descending).

Neighbor tone:
----------------
Goes a step (or half-step) away either direction and back to embellish a stationary
note. A change of harmony may occur, though the embellished note remains the same.
An upper neighbor tone and lower neighbor tone together makes a turn. Often, the
third note (the chord tone) is omitted, leaving just a double neighbor tone.

Appogiatura:
--------------
Skip, then step to resolve into a chord tone. Accented--do not couble the resolved
chord tone, esp. when in outer voice. A common lower appogiatura is leading tone
to tonic.

Suspension:
-------------
A note held over before resolving. Melodic rhythm often felt strong-weak while
harmonic rhythm may be weak-strong. If the first of the tied notes is shorter than
the second, it feels like an appogiatura that arrived too early. If the suspended
note is the leading tone or is chromatically altered, it will usually progress up.
Suspended chords are ovten used in cadences in slow movements.

Ornamented resolution:
------------------------
Before a nonharmonic tone resolves, a composer may add a neighbor tone, escape
tone, cambiata, or anticipation. There may be a chord tone before the ornament.

The second-inversion triad:

=============================
Called six-four chord because it contains the intervals of a sixth and a fourth--as
the fourth is a dissonant interval and the sixth is an unstable consonance, the
chord is unstable. It's most commonly seen as a grouping of nonharmonic tones, as
in the "cadential 6-4"--an I64 resolves to become a V. That progression is strong-
weak, btw. If, however, a strong chord goes before it, then its seen as a weak-
strong progression to the dominant. III and V chords don't introduce the cadential
6-4 very well.
The 6-4 chord gives a feeling of suspense and may be used for effect.

Cadences:
===========

Authentic cadence:
--------------------
V-I, may be extended to II-V-I or IV-V-I. Strongest cadence is II6-I64-V-I. The
cadence may be altered for varying degrees of finality.

Perfect and imperfect cadences:

---------------------------------
A "perfect" cadence is usually only one where both V and I are in root position,
with the tonic sounding highest (i.e. in the soprano line). Anything else is
imperfect, and you can really change almost anything to make it less final. When I
precedes V, it's usually inverted so as not to weaken the cadence.

Half cadence:
---------------
The cadential 6-4 also accentuates the half cadence, such as IV-I64-V, VI-I64-V, or
I-I64-V. Sometimes, the chord before the dominant will have a raised fourth
degree, creating a leading tone to the dominant.

Plagal cadence:
-----------------
IV-I. Often used after PAC to add extra finality to the end of a piece, but not
always. Using a minor IV at the end of a piece in a mojor mode creates a colorful
ending.

Deceptive cadence:
--------------------
PAC, but using some other chord instead of tonic. Very good indicator of tonality.
Most frequent example is V-VI. Using a minor VI in a phrase that has been major
adds great surprise--often indicates change in mood. Just about anything you can
think of can be used for deception. A deceptive cadence is often used as a joint
for overlapping phrases. The use of a deceptive cadence near the end of a piece
renews interest before the final PAC.
Cadences can also be extended either by repetition or by continuing the melodic
line.
The Dominant Seventh Chord:
=============================
Chords are made by moving voices--the seventh chord did not just come about by
stacking a third on a triad. The seventh on V7 is usually seen as purely melodic.
The seventh chord is dissonant. VII often acts like V7. The resolution of V7 (and
it must be resolved) has two aspects: harmonic and melodic.
Melodically, the tendency tones want to move to somewhere consonant. The
dimineshed fifth wants to contract to a third (major or minor), with both voices
moving inward. If only the leading tone moved, the interval would still be a P4.
The seventh wants to move down to a sixth. The dominant wants to move up to the
tonic.
V7 regularly resolves to I. V7-I is a very good indicator of tonality. The
seventh descends one degree, the third ascends to the tonic, and the fifth descends
to the tonic. Watch your doubling. The fifth of the chord is frequently omitted,
doubling the root, and providing a fifth for the tonic triad. The third is
occasionally omitted, resulting in less sonority. If that's the case, the leading
tone can descend to the dominant to provide a fifth for the tonic chord.
The V65 (first-inversion) works just like V7 and any other first-inversion triad.
V43 (second-inversion) is very rhythmically weak and is only used in passing. V42
(third-inversion) is a strong chord. The dissonance between the bass and the upper
voices gives the bass a strong tendency to move. The soprano voice often moves
from the supertonic up a fourth to the dominant, doubling the fifth in a tonic
triad.
Preparation of dissonances varies wildly throughout common practice. A dissonance
is prepared when:
a) it enters as a suspension
b) it enters as a repeated tone
c) it enters by step
d) it is not prepared when entered by skip
A melodic fragment contained within a tritone will usually have to be resolved.

Secondary dominants:
======================
Accidentals will be present on almost any given page of music, and can signify
alterations of nonharmonic tones, modulation, etc. Composers prefer the sound of
dominant harmony to non-dominant functions. The procedure gave color by adding new
tones and increased sense of direction and movement. When taken to extremes, it
formed progressions where each chord was the dominant of the next.
"Any degree of a scale may be preceded by its own dominant harmony without
weakening the fundamental tonality." If tonal chords are supported by their own
tonal chords, their effect is strengthened. However, the main purpose of the
secondary dominant was originally to add color by adding new tones.
Resolution of a secondary dominant is the same as for V-I cadence--observe proper
voice leading. The smoothest way to introduce a secondary dominant is often a
chord that could be in the key that the secondary dominant belongs to. If this
can't be done, the preceding chord may be chromatically altered.
Such cross-relation is common where the altered notes are treated as scale degrees
of the melodic and harmonic minor.

V of II:
----------
Not used when minor mode is prevalent because minor supertonic cannot act as tonic.
The tonic of the key is chromatically raised to become the leading tone of II.

V of III:
-----------
There are two different ones, each a semitone apart in pitch (for major and minor
third). When major, the second degree is raised to become leading tone to the
third, and the fourth degree is raised to form a perfect fifth with the seventh
degree root. Chromatically raised second and fourth resolve upward.
If minor, the secondary dominant will occur on the lowered seventh degree, and the
progression will sound like the V-I cadence of the relative major.

V of IV:
----------
Essentially an altered tonic triad. Very frequently used. Helps balance out
dominant progressions and modulations. Also used to establish the subdominant at
the beginning of a movement.

V of V:
---------
Used sometimes in half cadences.

V of VI:
----------
Like the mediant. Major is more common. The original dominant should be raised to
become leading tone. The original dominant is not altered when minor.

V of VII:
-----------
Not a good temporary tonic.

Irregular resolutions:
========================
All dissonances require resolution of some sort or another. The V7 usually
resolves to the tonic. Irregular resolution could be in root progression or voice
leading. How and why you choose to use an irregular resolution depends on your
goals.
The supertonic triad makes a weak resolution of the dominant seventh, because of
its similarity to a dominant ninth. The seventh is common to both chords and does
not resolve. Instead, V7 may resolve to V of V. The root of the V of V may be
eliminated to get rid of another common tone.
Chords on the third degree become less independent when associated with dominant
harmony. The major third sounds more like a nonharmonic tone than a resolution. A
more common resolution to III is when it acts as V of VI.
When resolving to VI, the leading tone should be in an inside voice and prolonged
(or repeated) in the IV. More often, though, the IV is in first inversion, and the
composer may double either the tonic of the root.
In V7-VI, each part may just proceed upward by a tone. Resolving a secondary
dominant irregularly can strengthen tonality if done right (e.g. V7ofVI-IV, because
of IV's tonal significance). V of II leads quite naturally to V7 by way of a
tendency tone. All similar to V7ofVI-IV are: V7ofV-III and V7ofIII-I. Secondary
dominants can also resolve to each other.

The Diminished Seventh Chord:

===============================
Another third on top of V7 makes V9. Dominant ninth chords are frequently found
without roots. The dominant ninth, most commonly seen as a rootless minor chord is
paradoxically the most ambiguous of chords, though it is made up almost entirely of
tendency tones. Because of its makeup, the ear can't tell which tone is which
until its resolution. Any note could be the leading tone, depending on how it's
written. (BTW, the minor ninth w/o root is also called the diminished seventh.)
The dominant diminished seventh often resolves to the tonic, with the dim fifth
contracting to a third. Just remember: avoid direct fifths.

The Incomplete Major Ninth:

=============================
The seventh chord on the leading tone using major intervals is totally different
from the diminished seventh. While the factors of the dim seventh are the same and
sound the same in inversions, the major third between the upper two factors gives
the inversions different qualities.
The inversion with the ninth in the bass is rarely found except as a dominant
seventh where the sixth is a suspension in the bass, resolving to the root.
In resolution, parallel fifths can occur, but are usually avoided by moving the
second up to the third or by skip of a fifth down to the dominant, but the root of
the tonic triad won't be doubled.
If the upper voices are arranged so that the fifth is inverted to a fourth, they
may proceed down in parallel, but the sonority of the chord is better with the
ninth in the top voice.
The six-five-three inversion must resolve to I6 to avoid parallel fifths.
(Parallel octaves because of doubling is not a defect.) The six-four-three
inversion can resolve to either first inversion or root position.
The dissonant sixth degree can't descend on the minor scale without prominence of
the cross-relation of the tritone between sixth and third degrees.
The incomplete major ninth can't really resolve anywhere else--resolution to the
supertonic is weak because they share all the same tones. When resolved to III, it
sounds more like VII7. Resolution to IV can be done, but the tonic in the bass
makes it into a regular resolution. Going to VI is basically like II and IV in one
package--if VI is in first inversion, it sounds like III. And if VI is in root
position, the progression sounds like iib-i.
It doesn't work well as a secondary dominant because it can't act as dominant of a
minor tonic.

The Complete Dominant Ninth:

==============================
Far less frequent than incomplete forms. Fifth is usually omitted.
The ninth may appear as any type of nonharmonic or melodic tone, resolving to the
fifth degree before the chord itself resolves. Often an appogiatura, which makes
it very pungent.
The ninth may be used as a chord tone, though it may be absent from the chord
itself.
Or, the ninth can be treated as a normal dissonant chord tone requiring resolution.
The major dominant ninth was used for color at the end of the common practice
period. Not used often then, and almost not at all until the 19th century.
On the plus side, the chord doesn't require any preparation. The ninth should
always be placed at least a ninth above the root, and the leading tone is preferred
under the ninth. Placing all the tones within a single octave creates a tone
cluster, rather than a ninth. However, that arrangement isn't uncommon. Keep
spacing in mind when doing inversions.
The major ninth as a secondary dominant should not be used to introduce a minor
key. Modulation using the ninth as a pivot chord is best done as a secondary
dominant.

The Sequence:
===============
The systematic transposition of a melodic, rhythmic, and harmonic pattern is an
important resource of development in music. It adds variety to the unity of
repetition.
The pattern chosen can be a single motive or a whole phrase. Most often contains
two chords. A long pattern might not be heard as a sequence (which, depending on
your goals, might be a good thing). Using an anacrucis in the harmonic rhythm of
the pattern can make the sequence more fluent, and may even be strong enough to
make the melodic rhythms syncopated. The rhythmic impression of pressing forward
is heightened if the anacrucis is long, followed by a short downbeat.
A single transposition isn't enough to make a sequence. However, composers rarely
let it continue past three appearances of the pattern without adding variation or
some sort of break. Exceptions include virtuoso cadenzas and technical pieces.
Determining what interval to transpose by depends on the desired harmonic
destination and the feasability of connection. A sequence may or may not modulate
at each transposition. When not modulating, each transposition sticks to degrees
of a single key. A nonmodulating sequence may also vary the interval transposed by
(e.g. first major third, then minor third).
Using secondary dominants in a nonmodulating sequence adds harmonic color from
foreign tones and contributes to rhythmic dissonance and unity. Sequences in which
each harmony is dominant of the next are of uncertain tonality. [Is this where
atonal music comes from?]
The most common modulating sequence embraces three keys. The modulation doesn't
take place in the pattern, but the last chord of the pattern is the pivot chord.
(Suspensions can help avoid cadence in passing modulations.) The pattern of a
modulating sequence must be tonal. The modulating sequence has more possibilities
for the degree of transposition.
Sequences are good keyboard practice. Best not to completely write out the
sequence, just one transposition.

Nondominant Seventh chords:

=============================
Dissonant nondominant chords were rare until the late 19th century. A dominant
seventh always has the leading tone a major third above the root and a seventh a
tritone above that.
The seventh resolves down one degree. The root usually resolves a fourth higher
(e.g. I7-IV), except for IV7 and VI7. The fifth or third is sometimes omitted,
especially when resolving to another seventh chord. Then the root is doubled.
Inversions resolve basically the same, but the root will usually stay put.

Tonic seventh:
----------------
Regular resolution to IV. In minor, lowered seventh descends to minor sixth.
Seventh occurs often as appogiatura to octave of root. Common irregular resolution
is to II7, with root becoming seventh.

Supertonic seventh:
---------------------
Regular resolution to V. Common in cadences to V or I64. Minor form used in all
modes, major form only in major surroundings. Irregular resolutions to I, III, VI,
and secondary dominants.

Mediant seventh:
------------------
Regular resolution to VI. Irregular resolutions to IV, II, and secondary
dominants.

Subdominant seventh:
----------------------
Regular resolution to II7-V. Seventh treated as appogiatura then. When fifth is
omitted, root is doubled and supertonic chord will be first inversion. Direct
resolution of IV7 to V requires care to avoid parallel fifths. Irregular
resolutions to I and some secondary dominants [which ones?].

Submediant seventh:
---------------------
Regular resolution to II. When the third is present (i.e. the tonic), it will be
kept to become the seventh of II. Voices may go through IV7 chord before resolving
to II.

Leading-tone seventh:
-----------------------
Treated as incomplete dominant ninth unless it resolves to III.

A continuous series of nondominant seventh chords may be used as a sequence. They

can also be used as pivot chords when an irregular resolution in one key becomes a
regular resolution in another.

Ninth, Eleventh, and Thirteenth chords:

=========================================
When a chord with five or more factors is played, one factor must be eliminated
unless two factors are played in turn by one voice. The omission of factors
weakens the structure of thirds and allows the ear to hear the upper factors as
melodic tones or at least ambiguous.
An appogiatura to the octave above the bass will create the effect of a ninth chord
if it's sufficiently salient. When the seventh is included and both are used in
the appogiatura, the fundamental harmony is first inversion. (BTW, all appogiature
can appear as suspensions.) The nondominant ninth is more independent if a change
in harmony occurs before the melodic tones are resolved. When leaving the
appogiatura unresolved, the resolution must still be implied and recognized as such
(whatever the hell that means!).
Consideration of eleventh and thirteenth chords brings up the subject of pedal
tones--tones which are persistent through either repetition or suspension in a
single voice over several changes in harmony. Using pedal tones can render
harmonic rhythm static, which should be offset by using chords that are dissonant
both by themselves and with the pedal (a good pedal tone will start and end as a
chord tone, but will be nonharmonic at sometime or another). Pedals are most
commonly on the tonic or dominant in the bass line. A pedal can be broken into a
rhythmic figure, and can even be decorated by other tones (e.g. left-hand arpeggios
in Chopin's Nocturne 1, Op. 9).
Because pedals strenthen tonality so much, accompanying harmony can go anywhere. A
double pedal using tonic and dominant makes the tonality even stronger. Pedals
(esp. double and even triple pedals) are the origin of polytonality.
Eleventh chords are brought about melodically (just like every other damn dissonant
chord). A tonic eleventh can be acheived by superimposing a V7 above a tonic
pedal. That's often arranged so that the dominant is heard as an appogiatura chord
resolving to the tonic.
A dominant eleventh is formed similarly--IV is played over a dominant pedal.
Eleventh chords of II and IV can be formed by appogiature.
The thirteenth chord is the highest you can go. It's almost always formed
contrapuntally.

The Raised Supertonic and Submediant:

=======================================
An altered chord is any that has one or more of its tones altered by an accidental
from any of three sources: change of mode (but not tonality), secondary dominants,
and truly altered chords.
A raised supertonic or submediant chord is a seventh chord on either of those two
degrees whose root and third are each raised. Their resolutions are to I and V7,
respectively.
If the harmonic rhythm of the chord and its resolution is strong-weak, the effect
is that of an appogiatura chord.
[Wait until you have more experience before trying to use these and the following
chords.]

The Neapolitan Sixth:

=======================
The supertonic triad with minor sixth degree and chromatically lowered root is
called the Neapolitan Sixth. It's called that because it was usually done in first
inversion, but is still called that even when not.
The chromatic alteration tends to give it a downward quality. Fells strongly
subdominant and most often resolves to V in some form. Best to double the bass.

[I just condensed a 300-page book into 32 pages of handwritten notes. Whew!]