The three chord trick

Subject: RE: The three chord trick
From: MikeL2
Date: 24 Dec 09 - 05:21 AM

Hi

A very interesting thread and one that has no definitive answer.

I agree with most of you has been said.

I am a singer guitarist ( not a guitarist singer !!!!). So I find that once I have found the key I want to sing in, I play around with the "standard Three chords".....until it sounds about right. I think that singing the melody does provide a guide to measure what sounds in tune or not.

When I have a feel for this " rough draft" I then try the relative chords to try to get a more melodic backing.

This way I have a "tried and tested" simple backing and a more complicated one....if I find that I have any problems with the " difficult " one I just go back to my simple one.

cheers
Seasons greetings to all.

Mike

Subject: RE: The three chord trick
From: SPB-Cooperator
Date: 24 Dec 09 - 01:30 AM

With the number of chord changes I always have a few deciding factors.

(1) How many I can remember
(2) How quickly my brain can send a message to my fingers that they need to move.
(3) How quickly my fingers can actually move.

Subject: RE: The three chord trick
From: olddude
Date: 23 Dec 09 - 06:38 PM

As Dave Van Ronk would say
"close enough for folk music"

Subject: RE: The three chord trick
From: PHJim
Date: 23 Dec 09 - 06:12 PM

There are more than one "correct" set of chords for almost any song. Some songs can be accompanied by as few as one (or no) chords.
One of the first songs I learned to accompany was Burl Ives' version of a song I called Jock Of Diamonds. I used two chords for this song. Soon after I heard Jackie Washington play the same song and he used two chords per measure.
Most people would use 2 or 3 chords to accompany Oh Susannah, but James Taylor's version uses about two changes per measure and many more than two or three chords. GUEST,number6's mention of James Taylor reminded me of this.

Subject: RE: The three chord trick
From: GUEST,number 6
Date: 23 Dec 09 - 01:17 PM

"If you play more than two chords, you're showing off."
... Woody Guthrie

"I don't read music. I don't write it. So I wander around on the guitar until something starts to present itself."
   ... James Taylor

biLL

Subject: RE: The three chord trick
From: M.Ted
Date: 23 Dec 09 - 01:05 PM

One thing that is often overlooked is that the "chords" that we generally play, whether on guitar, melodeon, or whatever, often don't really harmonize with the melody--they are accompaniments--that is, independent rhythmic/melodic phrases that are really one or another species of counterpoint--

This is important to understand because if you sit down and write chord triads on the melody, you'll often come up with chords that don't really work on your instrument.

Subject: RE: The three chord trick
From: Tim Leaning
Date: 22 Dec 09 - 06:22 PM

Blimey a music thread!
and brain bendingly interesting too
Cheers.

Subject: RE: The three chord trick
From: Jack Campin
Date: 22 Dec 09 - 04:13 PM

One nice example is Weill's own arrangement of "Mack the Knife". The tune is completely diatonic and the basic chord structure is very simple and blues-like, but as the song goes on he adds more and more dissonance - gives it an atmosphere of menace that you couldn't hope to replicate using simple folk-style harmonies.

Subject: RE: The three chord trick
From: PoppaGator
Date: 22 Dec 09 - 03:30 PM

Defining the "right" chords to accompany a given melody is a slippery and subjective business. Most folk songs and other simple melodies work just fine with the basic three chords ~ often even with just two of 'em.

Now, within a passage where the chord doesn't change, the melody generally includes some notes other than the three which define the accompanying chord. Those who like to make things complicated sometimes name a new chord for every single such note, which may be technically correct but is certainly unnecessary.

For example, consider "Fishin' Blues" in the key of C, as typically fingerpicked on the guitar. I generally think of myself as "staying" on the C major chord for the first two full measures while picking out the melody on the two highest strings, playing each of those two individual strings open and also fretted at the first, third, and/or (optionally, for those who can stretch) fourth frets. To my mind, this entire sequence is played against the backdrop of ONE chord, the basic C major triad, but I have seen trasncriptions where a new chord name (presumably a correct name for a four-note chord) is provided for every single note of the melody; each of these chords, presumably, consists of C + E + G plus the current melody note.

On the other hand, of course, there are more complex tunes which pretty much demand a more complex harmonic accompaniment. While one can, I suppose, cobble together a basic three-chord arrangement for any tune, some tunes require the use of a fuller set of chords to convey the appropriate movement through the melody.

Subject: RE: The three chord trick
From: Piers Plowman
Date: 22 Dec 09 - 10:10 AM

I wrote: "One could invent any number of scales with various combinations of intervals."

And people have. There's a famous book by Nicolas Slonimsky called "A Thesaurus of Scales and Musical Patterns". He worked out lots of different scales (and musical patterns) using different methods. I practiced them a bit, but got bored with it, because none of them sounded much better or worse than any of the others. This isn't a criticism of the book; it could be fascinating, if approached differently. I've just been practicing other things.

That's the way scales and chords are: a harmony by itself is nothing, it needs to be used in a musical situation. That usually means attached to a melody.

Subject: RE: The three chord trick
From: Piers Plowman
Date: 22 Dec 09 - 09:56 AM

Songs in minor keys usually have notes that don't "belong" in the key, strictly speaking. For example, A minor is the relative minor of C maj.:

A B C D E F G A

One can call this "A Aeolian", if one is feeling modal.

The chord on the fifth degree of Am would be Em7: E G B D
However, people often want E7 -> Am rather than Em7 --> Am. V7 ("V" meaning a chord on the fifth degree of the scale and with a major third) resolving to the tonic (a chord on the first degree of the scale, that is, the main chord for the key we're in) is what people seem to want. So, Em7 is often replaced with E7 when playing in Am. What does this mean? The only note that's changed is G --> G#. But G# is the leading tone of A, i.e., people seem to think that a G# "leads" to an A more than a G does. This isn't a coincidence. If we break down a simple progression from E7 --> Am it could look like this:

E --> E (It stays where it is)
G# --> A
B --> A or C
D --> E or C

This is really the secret of harmony (or maybe _a_ secret): voice leading, i.e., one thinks of each of the notes being played as a voice and moving from note to note. When people are thinking about writing harmonies, they are often striving for "smooth" voice leading.   

There are three common minor scales: the natural minor, the melodic minor and the harmonic minor. Many songs in minor keys will use both the major and minor sixth and the major and minor seventh, depending on whether the melody is ascending or descending. That is, one might use the major seventh when ascending so as to use the leading tone, but the minor seventh when descending. When ascending, one might use the major sixth in order to avoid the large gap between the minor sixth and the major seventh, but the minor sixth when descending, because one is playing the minor seventh, so there isn't a large gap between the sixth and the seventh.

The Dorian mode is fairly popular. It has the minor third, the major sixth and the minor seventh.

There's nothing magic about any of these scales; what really counts is the melody. One could invent any number of scales with various combinations of intervals. What I find really helps is to learn the sounds of the various intervals (and chords).

From my point of view, the real use of practicing this sort of thing is to train one's musical intuition.

Subject: RE: The three chord trick
From: beeliner
Date: 22 Dec 09 - 09:40 AM

'Status Quo are renowned for using as few chords as possible - they called one of their albums In search of the fourth chord!'

"Pictures of Matchstick Men" requires six.

Subject: RE: The three chord trick
From: treewind
Date: 22 Dec 09 - 09:03 AM

John Kirkpatricks "Gravedigger's Song" is in the Locrian mode and very effective for that. Hildegard of Bingen wrote stuff in all modes including Lydian which must be the other one that's rarely used.

Whether Hildegard lived in the "real word" could be a matter of conjecture...

Anahata

Subject: RE: The three chord trick
From: s&r
Date: 22 Dec 09 - 07:45 AM

I was taught years ago that a 'discord' is a chord of unrest; a concord is a chord of rest, and they're all chords.

Stu

Subject: RE: The three chord trick
From: Piers Plowman
Date: 22 Dec 09 - 07:36 AM

Subject: RE: The three chord trick
From: Pip Radish - PM
Date: 22 Dec 09 - 07:03 AM

'Presumably there are some combinations of notes which sound so "different" that you'd never actually want to use them'

Well, just give it a try! For what it's worth, what really sounds discordant is two notes 1/2 step apart played simultaneously. However, even this is used fairly frequently, when playing major seventh chords. Harmony is all about resolving discords and the balance between consonance and dissonance (a bit like life, really).

" - just as (I seem to remember) there are a couple of modes which are more or less theoretical & have no real-world use"

Actually, I think it's just the locrian mode, which one can use when playing over a half-diminished seventh chord, although there are "spicier" variants (i.e., ones with more dissonance).

For what it's worth, what I've found to work best for me with respect to scales and modes is to learn all the major scales inside and out (as best one can) and otherwise to play by ear. However, I have practiced other scales (not modes so much, except for the natural minor scale) and it's fun and worthwhile, but learning to play by ear is more important, in my opinion.

Subject: RE: The three chord trick
From: MGM·Lion
Date: 22 Dec 09 - 07:17 AM

A linguistic or semantic oddity BTW — there is actually no such thing as a 'dischord'.

For some reason, tho a harmonious combination of sounds is called a 'chord', an unharmonious one is called a 'discord', without the 'h'.

Why? Who knows! — that's just English for you; go down to the 'except/accept' thread below the line for many more such examples. But I have carefully checked two dictionaries [Chambers & COED] which both confirm what I say.

Subject: RE: The three chord trick
From: Phil Edwards
Date: 22 Dec 09 - 07:03 AM

Status Quo are renowned for using as few chords as possible - they called one of their albums In search of the fourth chord!

That "Wind and Rain" is in C (Johnny C.'s version was in D (and sung an octave or two lower)).

Answering my own question (with a little help from my friends!), I think I've been working on the assumption that chords with accidentals in would sound "wrong". Actually they just sound "different" - just as a scale in different modes sounds different. Presumably there are some combinations of notes which sound so "different" that you'd never actually want to use them - just as (I seem to remember) there are a couple of modes which are more or less theoretical & have no real-world use - but I guess you'd have to bend a 1-3-5 triad quite a long way out of shape before you got an actual dischord.

Subject: RE: The three chord trick
From: Piers Plowman
Date: 22 Dec 09 - 06:58 AM

Subject: RE: The three chord trick
From: Pip Radish - PM
Date: 21 Dec 09 - 06:13 PM


"I guess the question is whether accidentals in chords are much more common than in melody lines"

No, I don't think so.

" - and whether you generally have more latitude, in 'chording' a melody, than the three-chord trick would imply."

Yes, definitely. However, there are some standard "tricks" that people use. Actually, "common idioms" would be a better way of putting it.

As far as sevenths are concerned, the flatted seventh is typical of blues and popular music developed from blues or influenced by it. On the other hand, people like to use the major seventh as "the leading tone", i.e., it "leads to" the tonic. For example, in C, it would be B --> C. So, it's fairly common for songs to include both flatted sevenths and major sevenths. Please note that a major triad with a flatted seventh is a dominant seventh chord, which resolves to a tonic whose root is a fifth below the root of the dom. 7 chord, i.e., G7 --> C7 or C7 --> F. So, if you're in C and play a C7, you can resolve to F and viola! you're in F (major or minor). You have modulated. You can stay there, or go back to C.

Subject: RE: The three chord trick
From: Piers Plowman
Date: 22 Dec 09 - 06:47 AM

Subject: RE: The three chord trick
From: Pip Radish - PM
Date: 22 Dec 09 - 05:21 AM

"Piers - that looked fascinating but I didn't get very much of it; you seem to have started with Lesson Four!"

Sorry, I'll try to explain better.

From your original posting:

"This leaves me with a puzzle: it suggests that the 1, 4 and 5 chords, plus their relative minors, are the only chords which are possible in any given key signature. I assume this isn't the case, or the whole of Western music would sound like Status Quo."

Well, actually it sort of is the case, except that there is the additional diminished triad, if we just take the triads.

Triads in C:

C E G --- C maj.
D F A --- D min
E G B --- E min
F A C --- F maj
G B D --- G maj.
A C E --- A min.
B D F --- B dim (without the seventh)

It gets more interesting when we take the seventh chords:

C E G B --- C maj. 7
D F A C --- D min 7
E G B D --- E min 7
F A C E --- F maj 7
G B D F --- G dom. 7
A C E G --- A min 7
B D F A --- B half-diminished seventh

These are indeed the only seventh chords possible if you remain strictly in the key of C. Lots of songs do remain in a single key for their entire duration, especially children's songs, Christmas carols and folksongs, although even this is an over-simplification.

As you've noticed, there is a relationship between the chords with the minor thirds and the ones with the major thirds. This becomes even more clear when you look at the seventh chords. The D min7 chord contains the F maj. triad, the E min 7 contains the G major triad and the A min 7 contains the C maj. triad. This is useful for substitutions, i.e., one can often play a Dm7 instead of an F maj. chord.

Many songs don't just stay in one key. One uses altered tones or alterations to make things more interesting and there are a number of standard ways of doing this. One common way is to "modulate" into another key, briefly or for a longer period of time. What I wrote in my last posting was an explanation of one way of modulating.

Subject: RE: The three chord trick
From: s&r
Date: 22 Dec 09 - 06:43 AM

A full and reasonably clear explanation of the formation of chords


Stu

Subject: RE: The three chord trick
From: treewind
Date: 22 Dec 09 - 06:27 AM

Interesting: I don't know "The Wind and the Rain", but from the evidence it seems to be in F - is that right? (or is it in C?)

As for Nick Drake's song, yes you would have accidentals all over the place, if you wrote a piano score based on those harmonies, because it uses chords whose notes are not in the scale.

(1) "how come so many accompaniments seem to use *more* than the three 'obvious' chords"

You implied a possible answered that in your original question, where you said:

(2) "This leaves me with a puzzle: it suggests that the 1, 4 and 5 chords, plus their relative minors, are the only chords which are possible in any given key signature. I assume this isn't the case, or the whole of Western music would sound like Status Quo.

So the answer to (1) might be "to stop it from all sounding like Status Quo" ??? (I'm not sure I understand the reference, actually. Are they a three chord band?)

As for question 2, I hope I've made it clear that, while those few chords are the only ones possible within the constraints of a given key signature, you don't have to restrict yourself that way if you have an instrument that can play other chords. It's not actually illegal to use accidentals!

Key signatures are simply a convenience to save you having to write sharps or flats all over the score. It's a misnomer anyway, as it doesn't really say what key something is in. One sharp could be G major, E minor, or what in the folk world is often called A minor but is really Dorian mode. And you get some unconventional patterns in Balkan music, with sharps and flats in the same key sig...

Anahata

Subject: RE: The three chord trick
From: MGM·Lion
Date: 22 Dec 09 - 05:35 AM

Ah, Pip — I note on re-examining the thread, that you WERE the OP. Well, my point still holds in a way - your expanding the thread rather than initiating, for your additional points, a new one, might well confuse others who have logged on to it motivated by the same curiosity which made you OP it in first place; so perhaps another thread for your additional reflections ... ???

Subject: RE: The three chord trick
From: MGM·Lion
Date: 22 Dec 09 - 05:30 AM

===Michael - actually I was asking something a bit different, which is how come so many accompaniments seem to use *more* than the three 'obvious' chords.===

Exactly Pip - you make my point for me. A very good question; but one which is obviously going to confuse, rather than enlighten, the OP, who asked a specific question, the simple answer to which your 'asking something a bit different' is going to distract him from. I think your interesting questions would therefore be better put on another thread dedicated to your 'different question' — as a kindness to the OP who wants to know something definite without distraction. Would you not agree?

Subject: RE: The three chord trick
From: Phil Edwards
Date: 22 Dec 09 - 05:21 AM

Michael - actually I was asking something a bit different, which is how come so many accompaniments seem to use *more* than the three 'obvious' chords. For a fairly extreme example, take Nick Drake's song "Time has Told Me". Simplified a bit, in the verse there's a C-F-G chord progression, but the first time through he gets from the C to the F via Bb ("A troubled cure"), and the second time through ("For someday our ocean") via Eb. Then the chorus alternates an Eb with an E7. Eb to E - three flats to four sharps - madness, I tell you!

I've never seen the sheet music for "THTM", what with copyright and everything, but the C-F-G suggests to me that the melody's in C. But then there's a Bb chord (Bb-D-F), an Eb (Eb-G-Bb) and an E7 (E-G#-B-D) (hmmm, see what he's done there...) - if you were writing that out in dots there'd be accidentals all over the show.

Piers - that looked fascinating but I didn't get very much of it; you seem to have started with Lesson Four!

He seemed to be saying that a C-E-G triad becomes different chords depending on the key signature.

No, 'he' was saying that if you take a piece of music paper (or a copy of BarFly) and write the notes C, E and G on a single staff, what you have written becomes different chords depending on the key signature. Perhaps it's not the most obvious way of looking at it, but I thought I'd made it reasonably clear what I was doing.

A bit more background, in reply to Don and Anahata - my cunning plan is to accompany myself on "The Wind and the Rain", based on the version sung by Johnny Collins. The chords only come in on the refrain; transposed to suit my voice, they go

Note:   A                        G
Chord: F                        C
         Oh, the wind and the rain
...
Note:   G               F                   E
Chord: C               Dmi                C
         Crying in the dreadful wind and rain

Not a lot to it - and it's fairly easy to hear which notes 'want' a chord behind them. I certainly never considered chording every note!

Subject: RE: The three chord trick
From: Will Fly
Date: 22 Dec 09 - 04:43 AM

Eighteen chord changes on a uke - mmm... (slavers)... yummy! :-)

Subject: RE: The three chord trick
From: MGM·Lion
Date: 22 Dec 09 - 04:43 AM

No, you don't need them, ChasFidjit, within the terms of this thread, which started off with a simple request as to how to do the simple 3-chord trick accompnmnt; which I & others endeavoured to answer simply. Now the thread is being hijacked, as I perceive it, by a whole lot of those who want to show what clever effects they can get by complex rather than simple progressions. They should start a new thread, called 'The augmented/diminished/modulate·till·blue·in·the·face chord tricks' or some such, & leave this one to the ones who started it off, wanting to know how to keep it simple but effective.

Subject: RE: The three chord trick
From: Fidjit
Date: 22 Dec 09 - 04:18 AM

"I'll see you in my dreams"

Joe Browns version as played at the Harrisson concert.
Has eighteen chord changes, but do you need them?
And on a Ukulele !
Chas

Subject: RE: The three chord trick
From: Piers Plowman
Date: 22 Dec 09 - 02:53 AM

Pip Radish,

I have to start work soon, so I don't have time to read all the messages here now. I just want to mention the term "cadence", which refers to something that might be helpful to you when trying out harmonies. A cadence establishes a key. A typical cadence in C maj. would be Dm7 - G7 - C.
Dm7 = D F A C
G7 = G B D F
C   = C E G

When people refer to a diminished chord, they usually mean the diminished seventh, e.g., C Eb Gb Bbb (Bbb is the same as A on a piano or other instrument using tempered tuning). Please observe what happens when you lower one of these notes by a half-step, e.g.,

C Eb Gb Bbb --> B Eb Gb Bbb

we can call this B D# F# A, which is B7, so C dim.7 is the same as B7 without the tonic and with a flatted ninth. This means we can substitute a Cdim7 for a B7 (or B dominant seventh) chord. Dominant seventh chords resolve to the major or minor chord whose root is a fifth below, i.e., G7 resolves to C (see the cadence above). B7 therefore resolves to Emaj or Emin. (You can use Em6, Em7, E maj. 7, Emaj. 6, etc.) You have now _modulated_ to E (maj. or minor).

This also works for all of the other variants, i.e., lowering Eb to D, Gb to F, Bbb to G#/Ab. This is because the intervals between the notes in a diminished seventh chord are all the same, i.e., a minor third. (Check if you don't believe me.)

I've got a couple of other tricks up my sleeve, if this is of interest to you. It's loads of fun, so I'd like to encourage you to experiment. To reiterate a point made above, your ears are the ultimate judge.

And now I've got to make my first liter of coffee of the day and start work.

Subject: RE: The three chord trick
From: beeliner
Date: 21 Dec 09 - 11:11 PM

That was my intended point to the original poster, but stated more-or-less in reverse. If a C-E-G triad appears on a D-signed staff, then it is no longer a C-E-G triad. The C may not have the sharp symbol beside it, but it is there at the beginning of the line.

He seemed to be saying that a C-E-G triad becomes different chords depending on the key signature.

But this all may be more semantics than music theory.

Subject: RE: The three chord trick
From: Don Firth
Date: 21 Dec 09 - 10:10 PM

beeliner, the C–E–G triad is always a C major chord. It becomes a "different" chord in different keys. It is the Tonic chord in the key of C, the Subdominant in the key of G, and the Dominant in the key of F. But it is always a C major chord.

Don Firth

Subject: RE: The three chord trick
From: beeliner
Date: 21 Dec 09 - 09:44 PM

Mooh: "A diminished chord on the 7th degree (in this case C#) often wants to resolve to the 1 chord (D). I'm sure you know this, but I'm struggling to understand something here. What am I missing?"

You are, of course, correct. I was trying, perhaps clumsily, to address the OP's observation that a C-E-G triad produces different chords depending on the key signature, which of course it doesn't, since in the examples s/he cited, it's no longer a C-E-G chord.

Subject: RE: The three chord trick
From: Don Firth
Date: 21 Dec 09 - 07:31 PM

Okay, here's a major clue for working out the correct chords to a melody.

First of all, the assumption, "With a bit of trial and error I discovered that the chord that sounded right would always have the relevant melody note in it somewhere .  .  ." is not guaranteed to be correct. It depends a great deal on the placement of the note in the measure (on the beat or off?) and the duration of the note (long, or just a passing tone?). If you try to put a chord to all the notes in a melody, you will a) be very busy and/or b) go quickly nutso!

A simple, straightforward approach is to sing the melody while playing, presumably, a guitar or other chordal instrument, starting with the tonic chord (C in the key of C) and sit right there until the melody and the accompanying instrument "clank!" Sound bad together, or at least just don't sound right. Then, try one of the other two chords. One of them should sound okay. Then keep going on that chord until you come to the next "clank." And proceed thus through the song until you have the chords worked out.

Once you do that, if you want to, and you think the song can take it, you might experiment with substituting an occasional relative minor chord to see how it sounds. Trial and error.

Even classically trained musicians (which I am) have to rely on the ear to tell them what works. I can look at a melody line in written music and get a pretty good idea of what chords to use—but—I can't really be sure until I try singing the song and experimenting with the chords, as I have described above.

Here's something to note:   Assume two experienced musicians may work out the same song. Most of their chords will be the same. But not necessarily all.

Example: Do you know the song "I Ride an Old Paint?"

At the end of each verse, most song books give the chords on the last line of each verse as (say it's in C):

For the [G7]fiery and the snuffy are a-[C]rearin' to go.

I do it this way:

For the [G7]fiery and the snuffy are a-[F]rearin' to [C]go

Both are perfectly "correct." But I think the F adds a nice harmonic touch to the end of the line.

Some songs are sufficiently simple and straightforward that using any but the three primary chords (or sometimes only two, such as a song like "Down in the Valley") is gilding the lily and doesn't really sound that great. Other songs offer lots of opportunities for alternative chords.

Experiment. Use your ear.

Don Firth

Subject: RE: The three chord trick
From: PHJim
Date: 21 Dec 09 - 07:16 PM

If you take every other note in the C diatonic scale and play three note chords:
C-E-G = CMa
D-F-A = Dmi
E-G-B = Emi
F-A-C = FMa
G-B-D = GMa
A-C-E = Ami
B-D-F = Bdim

If you take every other note and play 4 note chords:
C-E-G-B = CMa7
D-F-A-C = Dmi7
E-G-B-D = Emi7
F-A-C-E = FMa7
G-B-D-F = Gdom7 or G7
A-C-E-G = Ami7
B-D-F-A = Bhalf-dim or Bmi7b5

Subject: RE: The three chord trick
From: Will Fly
Date: 21 Dec 09 - 06:49 PM

Meant to include an example of chord ambiguity in my last post. One of the most complex tunes I've ever attempted, in terms of potential chord structures, is Jerome Kern's All The Things You Are. (The link is to a short improvisation around the tune). Many of the chords are improvised around the melody, and the melody line itself is embellished - all on the spur of the moment in a one-take effort.

Whether any of it is "correct" or not is a matter of taste...

Subject: RE: The three chord trick
From: Will Fly
Date: 21 Dec 09 - 06:40 PM

What's interesting, from a jazzer's perspective, is that a melody and an improvisation on that melody can flow through chords that don't appear to be the "right" chords, or even substitutes for those chords.

Basically, you can put whatever chords you want to any tune in any key. Whether they sound right, wrong or indifferent - or anywhere near to the composer's intentions is all in the ear of the arranger. The reality being, of course, that to go way off track produces dischords which might be completely incomprehensible.

An interesting example of chord complexity/simplicity arises when you put chords behind melodies such as Christmas carols. Most carols will function with a very simple chord structure - perhaps just two or three chords - but will also support incredibly complex and architecturally interesting harmonies (say in a choral setting). You can find similar examples of alternative/substituted chording in tunes in a range of music from jazz to folk tunes. The correctness or otherwise lies within the taste of the player and the listener.

Subject: RE: The three chord trick
From: treewind
Date: 21 Dec 09 - 06:29 PM

Looks like you've got it pretty well right. If it sounds right...
At the start of the thread I had no idea what you were playing. There are lots of good folk dance tunes and songs from the British traditions that don't need complicated chords. There are also more complicated alternative chords that you can use, but don't have to, and usually can't on a melodeon. They are popular with piano accordion players who like to use them just because they can...

Anahata

Subject: RE: The three chord trick
From: Ron Davies
Date: 21 Dec 09 - 06:28 PM

Thanks, Don. I wish I could claim it. But I'm just spreading the plague of bad puns. (Actually all puns are good puns, from my perspective. I can't get enough of them.)

Subject: RE: The three chord trick
From: Phil Edwards
Date: 21 Dec 09 - 06:13 PM

For everyone who thinks (like MtheGM) that I'm overcomplicating something simple, all I can say is that it looks different when you come at it from the outside. I can work out tunes by sol-fa, turn them into notes & transpose them between keys, but chords are a closed book to me. (I can play two chords on the guitar - one of them's G and the other one isn't.)

This all started when I wanted to put some chords to the melody of a song, with a view to putting together a very basic melodeon accompaniment. I knew which bit of the melody the chord should go under & even which note of the melody it should support, but what chord should it be? With a bit of trial and error I discovered that the chord that sounded right would always have the relevant melody note in it somewhere, but the 'right' 1-3-5 chord might have that note in any of the three positions. (In other words, if the melody note was an A and the tune was in C, the 'right' chord might be D-F-A (D minor), F-A-C (F major) or A-C-E (A minor).)

But what this suggested to me - after I'd done a bit more experimenting and made a lot more notes - was that the 'right' chord for any note, given a tune in the key of C, could only be C, F or G major, or else A, D or E minor. This made me feel I must have gone wrong somewhere, as I know I've seen tunes written out with a much broader range of chords.

So I'm really interested by Anahata's comment -

bear in mind that the home key of a song or tune is only the key it starts and finishes in (usually). On the journey from one end to the other, it may well wander off into other keys, even if only very temporarily.

although I suspect it raises more questions than it answers. I've only played 50-odd tunes, but out of those I can count the ones with accidentals on the fingers of one hand - and almost all of those are flattened sevenths, generally sharpened again the next time you come to them (a la Elsie Marley - those half-Mixolydian tunes remind me of a stair tread that sometimes creaks when you step on it and sometimes doesn't). Some of the chord sequences I've seen seem to range much more widely. I guess the question is whether accidentals in chords are much more common than in melody lines - and whether you generally have more latitude, in 'chording' a melody, than the three-chord trick would imply.

Subject: RE: The three chord trick
From: Fidjit
Date: 21 Dec 09 - 05:35 PM

Wow !
I understood most of all this, Mumbo Jumbo and I don't read music.

Never mind what it's name is.

If it sounds right then it fits in.

Chas

Subject: RE: The three chord trick
From: Don Firth
Date: 21 Dec 09 - 05:23 PM

Excellent, Ron! Thank you! I've just added it to my repertoire!

(Gawd, I love horrible puns!!)

Don Firth

Subject: RE: The three chord trick
From: Mo the caller
Date: 21 Dec 09 - 05:15 PM

It seems as though some people know instinctively about chords and some (me!) don't.
It caused no problems with a recorder but I'm now trying to learn the accordian. I suppose I could just use my left hand to pull and push, but maybe there is more.
How do I 'do' the 3 chord trick, and how do I know when it is appropriate?
A lot of the tunes I like are Playford dance tunes and such. Often modal and triple time but not a waltz. So Um-pah-pah doesn't seem right.
Learning piano was a matter of right hand play what's on the top stave, left hand plays the line/s at the bottom. So that's no help.

Subject: RE: The three chord trick
From: Ron Davies
Date: 21 Dec 09 - 05:04 PM

But please don't forget about this--can't resist--though I think it's been on here before.


After a session a C, Eb, and a G stopped in a pub.   The bartender was about to ask them what they'd like but then looked down and said:   "We don't serve minors". So the Eb, being a self-sacrificing sort, left, and the others shared an open fifth.   Eventually the G left and the C had a tonic.

Subject: RE: The three chord trick
From: Mooh
Date: 21 Dec 09 - 04:56 PM

Beeliner said, (quotes Mooh) " "Everything is something. C#, E, G is a C# diminished." (end quote)

Right, which is why I put 'nothing' in quotes.

Now, if a piece in D required a C major chord, which is certainly possible, then of course a natural symbol would be put in front of the C, negating the sharp."

Sorry. To clarify, I was responding to the OP.

Beeliner again, "So if a piece is signed D (two sharps), it's true that C#, E and G are 'nothing', and that's why you are not likely to encounter that combination in that key."

Oh, I don't know about that. All 3 notes are in the key. A diminished chord on the 7th degree (in this case C#) often wants to resolve to the 1 chord (D). I'm sure you know this, but I'm struggling to understand something here. What am I missing?

Peace, Mooh.

Subject: RE: The three chord trick
From: WalkaboutsVerse
Date: 21 Dec 09 - 04:54 PM

I'm glad that "English folk-music, for centuries, has entertained people, with telling and/or dancing, via, mostly, the repetition of TUNES" (here ).

Subject: RE: The three chord trick
From: Don Firth
Date: 21 Dec 09 - 04:29 PM

I'm not going to get into this can of worms. I've been here before, and I'm always amazed at how people try to make hard work of something that really isn't that complicated.

MtheGM at 21 Dec 09 - 03:55 p.m. is right. If it's in a major key, the three major chords (tonic, subdominant, and dominant—major chords built on 1, 4, and 5 of the scale) are all that is really necessary.   You can use the relative minor chords as "color" chords if you wish, but for many folk songs, they should be used judiciously. Same with songs in a minor key. Accompany with the three primary minor chords, with the relative major keys usable as color chords.

In certain modes, the flatted 7th chord (Bb in the key of C, F in the key of G, etc.) may be called for.

But--get the regular major and minor keys down first before you start messing with modes. They, too, are relatively simple, but it's astounding how people seem to like to make hard work of something that's not really that complex.

Cripe! I wasn't going to get into this!

Don Firth

Subject: RE: The three chord trick
From: Acorn4
Date: 21 Dec 09 - 04:04 PM

Then of course you get the complication with folk in that you get the modal chord - for D this would be the chord of C as described above, which isn't one of the triads you get by playing up the scale - in C this would be Bb D F -the chord of Bb.

Try looking at Lennon McCartney chord sequences -they explore all the posibilities.

Subject: RE: The three chord trick
From: treewind
Date: 21 Dec 09 - 03:59 PM

Well, I understand this to be the main point of the question:
"the 1, 4 and 5 chords, plus their relative minors, are the only chords which are possible in any given key signature"

Well yes - that's one reason why they are the popular "3-chord trick" chords, because they don't particularly suggest you are moving off into another key.

If you want go beyond 1 - 4- 5 and relative minor (6) there are still plenty of chords, but you have to use notes that aren't in the scale of your home key, and therefore aren't in the key signature. When writing music you put sharp and flat symbols in front to indicate this. If you put chord symbols under the music or lyrics those accidentals are implied, so if something is in C and you have a D chord in it, that D chord will have F# in it, even though that's not part of the home key.

If that seems strange, bear in mind that the home key of a song or tune is only the key it starts and finishes in (usually). On the journey from one end to the other, it may well wander off into other keys, even if only very temporarily. Apart form accidental appearing in the chords, even the melody of a tune doesn't always stick to the seven notes of the scale, and then the 3 chord trick really won't do.

Anahata

Subject: RE: The three chord trick
From: beeliner
Date: 21 Dec 09 - 03:58 PM

"Everything is something. C#, E, G is a C# diminished."

Right, which is why I put 'nothing' in quotes.

Now, if a piece in D required a C major chord, which is certainly possible, then of course a natural symbol would be put in front of the C, negating the sharp.

Subject: RE: The three chord trick
From: MGM·Lion
Date: 21 Dec 09 - 03:55 PM

I think you are over-complicating something quite simple. The 3-chord trick simply means that the chord of the first (or tonic), fourth (subdominant) and fifth (dominant) of each scale contain between them all the notes of that scale, so any tune played in that key can be effectively accompanied by just these three chords and no other [you can intro others if you wish, but it is not strictly necessary].

In the key of C, therefore, any tune can be accompd by the 3 chords af C [tonic] F [subdominant] G [dominant]. An added flattened 7th is often added to the 3 notes (1st, 3rd, 5th to form the major chord) of the dominant so the chord would be G7 - but this is not essential, tho it does give that 'anticipating return to the tonic' effect.

In G, it will be G C D; in F, F B♭ C; in D, D G A;   & so on —

Once you know this, it is easy to work out the first, fourth and fifth of any scale to tell you which chords will produce this '3-chord trick' accompt to a song in that key.

Subject: RE: The three chord trick
From: Mooh
Date: 21 Dec 09 - 03:45 PM

Everything is something. C#, E, G is a C# diminished.

Peace, Mooh.

Subject: RE: The three chord trick
From: DonMeixner
Date: 21 Dec 09 - 03:35 PM

Oh Jeepers, Back into Modes again!

Subject: RE: The three chord trick
From: catspaw49
Date: 21 Dec 09 - 03:35 PM

...........uh.........................Pip? I'm missing this one completely as I am equally thick as Will in this case. Perhaps if you took cranberries and stewed them like prunes they would taste much more like applesauce than rhubarb does..........That too is a possibility for you!

Spaw

Subject: RE: The three chord trick
From: beeliner
Date: 21 Dec 09 - 03:29 PM

At first I couldn't understand what you mean, now I do, but you're looking at it from the wrong direction.

The Ionian scale on C has no sharps or flats.

The same series of intervals, beginning with G, has one sharp, the F#.

The same series of intervals on D has two, the F# and the C#, etc.

The key signature is only a convenience, so that the sharps and flats don't have to be signed individually.

So if a piece is signed D (two sharps), it's true that C#, E and G are 'nothing', and that's why you are not likely to encounter that combination in that key.

Don't know if that's any help - I'm sure there'll be other replies.

Subject: RE: The three chord trick
From: Will Fly
Date: 21 Dec 09 - 03:21 PM

I'm not sure I understand your point, Pip (thickie that I am). I only know that various note combinations (at least 3) in any scale produce chords - and that those chords and their relationships are the same in any and every key.

Thus (in the scale of any key):

1+3+5 = major
1+b3+5 = minor
1+3+5+B7 = seventh
1+b3+5+b7 = minor seventh
1+3+5+6 = sixth
1+b3+5+6 = minor sixth
1+3+5+7 = major seventh

etc...

Subject: The three chord trick
From: Phil Edwards
Date: 21 Dec 09 - 03:10 PM

I've been thinking about chords. (I don't play any instrument that produces more than one note at once, but some recent messing about with a borrowed melodeon has made me think that perhaps I've been missing out.)

I've been thinking in particular about the relationship between chords and key signatures. So, say you take the notes C, E, G. If you play them in the key of C, that's C natural (hereinafter known as 'C'), E natural and G natural, and it's the C major chord.

In G it's C, E, G which is a C major chord
In D: C#, E, G which isn't anything
A: C#, E, G# => C# minor
E: C#, E, G# => C# minor
B: C#, E, G# => C# minor

Heading the other way

F: C, E, G => C major
Bb: C, Eb, G => C minor
Eb: C, Eb, G => C minor
Ab: C, Eb, G => C minor
Db: C, Eb, Gb => (nothing)

What's really interesting is what happens if you start from the triads rather than the key signature. So in the key of C, the possible do-mi-so combinations are

A, C, E => A minor
B, D, F => (nothing)
C, E, G => C major
D, F, A => D mi
E, G, B => E mi
F, A, C => F ma
G, B, D => G ma

In other words, there's one triad which is a dischord, three which are major chords and three minors. (This is the same whatever key you use - C just makes it easier to write out.) The majors are the 1, 4 and 5 chords, and the minors are the relative minors of those notes (C->Ami, F->Dmi, G->Emi).

This leaves me with a puzzle: it suggests that the 1, 4 and 5 chords, plus their relative minors, are the only chords which are possible in any given key signature. I assume this isn't the case, or the whole of Western music would sound like Status Quo.

More research needed!