Music: Just what exactly is 'key'?
April 10, 2004 7:16 PM   Subscribe

Music: Just what exactly is 'key'? [more inside]

I've read the definitions in some textbooks. I know the formulae for the scale constructions. A Major key is supposedly upbeat, a minor is more sad/dissonant/whatever. But what exactly is the 'key'?

There was this person I knew who could listen to half a minute of the music and proclaim: "That's in E Major". Is that by just recognising the tonic, or what?

Also, what does it mean to compose something in say, C Major (C D E F G A B)? You can't use other notes (accidentals)? Or you can't use accidentals frequently? You have to end the piece on the tonic? Or your chords must be assembled from only the 'key' scale? What if you have no chords in the piece (a violin solo)? Well, as you can guess, I'm pretty curious and bewildered. Hopefully, someone can attune me to the answer.
posted by Gyan to Media & Arts (21 answers total) 19 users marked this as a favorite
 
There are really no hard and fast rules, but for most pieces written before, say, 1900: yes, the piece will end on the tonic, or some inversion of a tonic chord. Accidentals are certainly not prohibited; more often than not, a piece will modulate into another key for a while, but instead of explicitly writing out a change in key signature, accidentals will just be thrown in. For example, if a Classical or Romantic composer wrote a sonata in C Major and wanted to make things interesing, he'd modulate to G Major (using some interesting chord progression) and restate the main theme; but to make it sound right he'd have to throw in F-sharps instead of F-naturals. Similarly, within a section that you'd otherwise think of as being in a given key, you can get chord progressions that require accidentals.

I'm sure that a real musicologist will be along shortly to explain things better.
posted by Johnny Assay at 8:12 PM on April 10, 2004


Let me see what I can remember from my piano lessons of yore...

"Key" just refers to the note that the scale you're using starts on, known as the tonic. Typically it's the note you come back to throughout the tune and especially at the end. If a tune uses notes from the C major (diatonic) scale, it's in the key of C major.

Things get confusing when you throw in the idea of modes, though. If the tonic of your scale is A but it uses the same sharps and flats as the G major scale, it's in the Dorian mode (IIRC). That's the same thing as playing a G major scale but starting and ending on A rather than G. In that case I believe you'd still use the key signature for A and use accientals to "naturalize" C and G, rather than just using the key signature for G, since A is still your tonic, but I'm a bit rusty on all the modal details.
posted by kindall at 12:06 AM on April 11, 2004


I'm no musicocolgist, but I've written a couple of songs, so I can sort of explain this.

The "key" is often chord that the song is using when it's not addressing some other issue. Take, for example, Leonard Cohen's "Suzanne".

It starts in E, and then veers into F sharp minors and G sharps when he's explaining all the issues you have to take into account when you're dealing with the bewlidering and difficult Suzanne. While the song can be said to happen in "E", it still has other chords in them: five or six of them.

My dad's a physicist, and he uses this term: "Main Note". He uses it to describe non-physics things the way he would say that a physics paper has a thesis. His term is naive, but also kind of clever. He'll say things like:

"Laziness is the 'main note' of teenagers".

By this he's intuiting something about music, and he does it - consciously, I think - to reference music.

I don't know at all how to figure out what key a piece of music is in (aside from how difficult it is to play on the trumpet, since certain keys which are easy to play on the trumpet are hard to play on the saxophone), but for most folk-music, punk and rock, the key is almost always the very first chord that's played.
posted by interrobang at 12:08 AM on April 11, 2004 [1 favorite]


Best answer: Here is the basic idea in a nutshell. Certain patterns and combinations of notes sound nice together, or make certain effects -- we call these scales and chords. Certain combinations of scales and certain combinations of chords sound better than others -- we call this harmony. For instance, combine certain chords together, and you get music that sounds sad -- we'll call this minor harmony. And if you combine certain other chords together, you get music that sounds kinda happy -- we'll call this major harmony.

Search a bit more, and you'll find that you can make a number of different harmonies that all sound 'minor' in a similar way, it's just that you play them higher or lower on they keyboard. So you give them different names in order to distinguish them, based upon where in the keyboard they lie. This one up here I'll call 'B minor', and this one down here I'll call 'D minor'.

I know that spiel comes across as incredibly condescending, but that really is all there is to it. At the same time, I realise that in isolation, without examples, the concepts I've described above can be quite difficult to get your head around and relate to 'real music', and it doesn't really answer all of your questions. So now you will all watch in horror as I ramble on for five hours about music, to try and give a feel for what 'key' means to a musician.

First of all, you have to realise that if you're writing in a certain key, there are no rules written down someplace that say, "no less than 80% of your notes must be taken from the home key", and "you may use no other chords than those constructed from your key", etc.. Certainly, if you're trying to write in a certain style then you may find that your music does follow rules like this. But that is because the style/period of music dictates conventions about what harmonies are acceptable and which are not, what chords are Godly and which are lewd, and so on. The concept of a 'key' arises from the complex interplay of notes and chords and harmony, not the other way around.

Let's say I'm writing a really simple blues song. If you know your twelve-bar blues, you'll know it goes something like

C C C C F F C C G F C C

(where each letter corresponds to a measure of music, and 'C' means 'C major chord', etc.). Now, I can take that exact piece of music, and 'transpose' it into a different key, say, A:

A A A A D D A A E D A A.

What makes the first piece of music 'in the key of C major', and the second piece of music 'in the key of A major'? How would you tell the difference between them if you were listening to them?

Well, the first thing to notice is that in both pieces, there is one chord that I spend most of my time on. In the first piece, it's C-major, and in the second piece it's A-major. In a very real sense, this is the 'main' chord of the piece. Sometimes this alone can be a good indicator to what key a piece is in -- particularly for forms such as blues, country, and pop.

Another, more subtle thing to notice, is that there is a very specific arrangement of chords sitting around the home chord. In the first case, we start on C, and we go up to F, and up to G. If we call C number 'one', and count up the keyboard, you can see that we use chords 'one', 'four', and 'five'.

Now do the same with the second piece. We start on A, and go up to D, and up to E. If we now call A number 'one', then D turns out to be number 'four', and E is number 'five'. It's the exact same relationship.

In fact, in the vast majority of Western music, these chords, I, IV and V are the most common chords used, simply because they are the ones that sound good when they're played one after another. When your brain hears that relationship of chords, even if you don't consciously realise it, it will be able to pick out which chord is number I, the 'home' key.

It's like if I drew an arrow on a piece of paper, and asked you to identify the point of the arrowhead. No matter which way round I turn the piece of paper, you'll be able to find the point, because your brain understands the relationship between the lines on the paper, even if I turn it upside down, stick it on the ceiling, or whatever. It's just like that with harmony and different keys.

So to answer your question about how your friend can identify a piece of music as being in 'E major', he's actually using two tricks. The first is that, like most of us, he can identify the 'home note'. Secondly, your friend is gifted with 'perfect pitch', which is the ability to hear a note, and identify what letter (C, B-flat, E) that note corresponds to. That truly is a gift, and most people can't do it, but the essential part is that he's identifying the 'home note' of the piece, and that's an intuitive thing that most people can manage quite well.

Perhaps you think you can't do this. Well, put on a piece of simple music, pause it halfway through, and just tell yourself to "hum the main note". You may be surprised at how easily your brain picks out a note to hum. At the very least, you can usually tell whether the piece sounds 'complete' and it could stop there (even if it carries on in real life), or if it sounds 'interrupted' and needs to carry on to go somewhere else.

Try these hastily-edited examples (MP3). In each case, decide whether the last chord/note you hear is the home chord/note, or whether it is some other chord/note. Answers at the bottom of the post.

1. Jupiter
2. Cello
3. Fortuna
4. Thrill

Okay, so how does all this touchy-feely "hearing the home note" stuff apply to real, hard music theory?

Pick up a piano sonata by Mozart that is 'in F major'. If you look at the start of the music, you'll see that the 'key signature' consists of a single B-flat. That B-flat is a signal that the music is 'in the key of F major'. But what does that mean?

Well, really, key signatures are just a notational device. When Mozart sits down to write a piece 'in F major', he'll find that almost all of the B's that he writes down will be B-flats, not B-naturals (or B-sharps!). The key signature is just a way to avoid having to write down all of those 'flat' symbols. Indeed, I could re-write the piece in an entirely different key signature that I've just made up (let's say, one G-sharp and one D-flat), and although I'd have to put accidentals all over the place, and it would be incredibly awkward to read, the actual music -- the notes that are played -- won't have changed. Play it back, and it'll still sound 'in F major'.

So, the key signature can give us a clue about what the 'key' of a piece is, but it is certainly not the end of the story. However, it does give us a useful hint about how we can puzzle out the 'key' of a piece. It seems to be something to do with how often you use certain notes.

For instance, in a simple piece in F major, the notes (F G A B-flat C D E) will be used far more often than any other notes. And likewise for any other key -- if you see a lot of a certain scale in a piece, then that piece is probably written in that scale!

This is why even a solo violin or cello line can have a strong sense of key, and even a strong sense of harmony and chord progressions. Bach especially was a master at creating musical lines that suggest very particular chords, even if those chords aren't specifically being played by any one instrument. Listen to this sample from the double violin concerto in D minor, and notice how, even though all the instruments are playing their own, independent, very complex tunes, there is an incredibly strong feeling of harmony and direction and chord changes.

Bach manages this because his style of music sticks quite rigidly to the 'rules' about what notes are allowed in a certain key, and what chords are allowed to follow other chords. (He doesn't use other accidentals unless he has specifically modulated into a different key for a section of the piece, in which case he uses accidentals to notate the new key that he's in. He practically always ends his pieces on the tonic. His chords are only assembled from notes in the key's scale.) Because of this, your brain can deduce very clearly what key and chords Bach is intending, even though the notes themselves are fleeting.

As you move forward into the Romantic era, with composers like Brahms and Chopin, musicians were beginning to explore the possibilities of dissonant sounds, and complex, sliding, chromatic harmonies. (They will use random accidentals to create the effect of fleetingly slipping into another key for a brief time. They will not necessarily stay in the same base key for the whole piece, and will sometimes start and end a piece in different keys.) This sample from a Brahms violin sonata, for instance, slides through a few keys in a short space of time, and contains several surprising chord progressions. But the concept of a 'home key' is still very much apparent, even if the current 'home key' changes every few seconds! Listen especially to the end of the sample, where some very clear, well-chosen chords bring us back round to the original home key.

It seems that the concept of a home key is crucial to the ear being able to understand a piece of music as distinct from a sequence of random notes. This is why it has endured, and why the ideas of scales, and home notes are found in most indigenous musics. Despite the efforts of experimental 20th century composers, who have tried at various times to write music with no dependence on a particular key or structure of chords ('serialism' or '12-tone music'), have not swayed us from needing a home key to rely on. Indeed, some composers still use these techniques, but usually only as an artistic effect, creating sounds that are dissonant and confusing (for example, the insanely difficult but compellingly exciting Coloana infinita by Ligeti).

I certainly have given a definitive answer to what a key is. I'm not sure there really is one that is applicable to all circumstances. But I hope by jabbering on for a while I've given enough examples of the use (and misuse!) of key to give you a better understanding of what it means to have a key.

Or I've just confused you even more.

Answers.

1. The melody ends on the home note. (Taken from 'Jupiter' from 'The Planets' by Holst. Played by the Orchestre Symphonique de Montréal under Charles Dutoit.)
2. The cello finishes on the augmented fourth note of the home scale! (Taken from the Suite for solo cello No. 1 in G major by Bach. Played by a nameless cellist on the Neon Genesis Evangelion soundtrack.)
3. Although the music finishes on a triumphant chord, and there is a big rest after it, the final chord is actually chord V, giving a feeling of wanting to fall forwards onto the next segment of music. (Taken from 'O Fortuna' from 'Carmina Burana' by Carl Orff. Performed by the Chorus and Orchestra Salzburg Mozarteum under Kurt Prestel.)
4. The last chord you hear is the start of the next 12-bar cycle, and it is indeed the home chord. (Taken from 'The Thrill is Gone', performed by B.B. King.)

posted by chrismear at 3:23 AM on April 11, 2004 [46 favorites]


Here's my take on it, from a person who has written many songs, but has had absolutely no formal training, so the following will be written in a very naive style! Don't take the following as gospel as it's my own unorthodox way of writing music.

I make a "chord" :- in my naive style, I call a "chord" more than one note sounding....but mostly it's 3-5 notes played together. If it sounds good, I call it a "chord". I then try to make a sequence of these chords. I use my own ears to judge whether this sequence "works"....but I understand there are actual rules to chord sequences (what works best).

With the chord sequence complete, I now lay melodies on top of this. Now I notice that certain notes sound really bad and some sound good (in harmony with the underlying chord). I'm guessing these bad notes are "accidentals". However, sometimes I find the accidentals give something extra to a melody if sparsely used. The same rules apply when I lay a bassline on top of the chord sequence - some notes "work", some don't. Also I find myself coming back to the original chord every 4 bars or 8 bars.

I'm guessing I'm following the rules to a degree by eliminating all that sounds "wrong" when writing music. Just don't ask me what key a song is in.

Hmmm, don't think I've really answered your question here!
posted by SpaceCadet at 3:27 AM on April 11, 2004


Excellent post, chrismear!!
posted by SpaceCadet at 3:34 AM on April 11, 2004


Wow, chrismear.

Beautiful post, thank-you. For the first time in my life I almost understand this.
posted by cedar at 6:29 AM on April 11, 2004


Here's an article by Toby Lester that gives more background into the emotional relationship with musical notes. Here's a excerpt:
NO other artistic medium moves us the way sound waves do, and in that regard music's meaning is emotional, in the word's original sense. The languages of music and emotion are remarkably similar; indeed, the link between musical mode and emotional mood has been the subject of philosophical inquiry and censorious dogma for centuries. Certain modes of music were to be kept out of Plato's ideal State because they evoked sorrowful or ungraceful or indolent feelings (Socrates: "When modes of music change, the fundamental laws of the State always change with them"). Saint Augustine feared the power of music to overwhelm the spiritual message of the hymns it accompanied, and this fear led the Church to pronounce certain musical modes -- and even certain melodic intervals -- dissonant and unlawful. Soviet censors, too, were notorious: they tried, for example, to keep Shostakovich's lugubrious dissonances (and their political overtones) in check.

The musical mode considered least dissonant (and thus most standard) in our Western tradition is the major mode, which is often equated with positive emotions, such as happiness, triumph, and love. The scale upon which the mode is based -- do, re, mi, fa, sol, la, ti, do -- is a sequence of intervals (tones and semitones) that sounds very natural to Western ears as it moves familiarly up and down toward resolution on do. Stop in mid-scale, though -- say, at fa or la -- and you feel the need to continue. What's happening is that the musical intervals do-fa and do-la represent suspended motion; they demand resolution without supplying it. Resolve this tension in the expected manner, and the result is a confident and happy feeling of coming home.
posted by john at 8:14 AM on April 11, 2004 [1 favorite]


Chrismear just summarized years of music theory classes, right there. Very thorough, very accurate, wish you had've been around when I was learning all that.

*bookmarks this thread*
posted by precocious at 8:46 AM on April 11, 2004


The musical mode considered least dissonant (and thus most standard) in our Western tradition is the major mode, which is often equated with positive emotions, such as happiness, triumph, and love.

"In our western tradition" being a key phrase. During the first years of Japan opening itself to the rest of the world, a couple centuries ago, the emperor invited a european orchestra to play Mozart for his court; the nobles sat politely through it, then discussed among themselves the agony of having to sit through those hours of horrific, unremiting dissonence. For an example going the other way, check out the Bulgarian Women's Choir.

As another side note: in the 1940s, jazz players (lead by Charlie Parker and Dizzy Gillespie) began to change their playing and hearing to accomodate ultra-rapid, ultra-frequent modulations from key to key. (It sounds dissonent if you're not used to it, but it does have its own logic and coolness.) Eventually, combined with modalism, the expressed and implied key changes went by so fast that they could appear not to be moving much at all, like spokes of a fast-turning wheel; Coltrane's A Love Supreme exemplifies this.
posted by Tlogmer at 10:17 AM on April 11, 2004


Response by poster: Fantastic post, chrismear. The Brahms violin sonata sample was especially instructive. I remember reading about certain Indian classical forms of music, where in addition to a line of melody, the 'home note' is played continuously in the background, making the progress of the melody apparent.

HAving read your answer, I realised that I had already known the concept of 'key' !! I just didn't know that was it . ;-)

Thanks to the rest of y'all too. Lots of insight here.
posted by Gyan at 11:16 AM on April 11, 2004


Amazing response, chrismear. As a little added value, I'll mention that the Ligeti piece is named for this Brancusi sculpture, whose name means 'infinite column' in Romanian.

(I once spent an entire afternoon trying to explain the concepts 'key' and 'tonality' to my brother so he could understand Charles Rosen's The Classical Style—a book I heartily recommend—and got nowhere; I finally realized what a huge difference it made that I'd taken piano lessons and he hadn't. I wish I'd had you there to explain for me.)
posted by languagehat at 11:37 AM on April 11, 2004


Reading through it again, I notice that I made a typo in the penultimate paragraph. It should have said:

I certainly haven't given a definitive answer to what a key is.

Whoops!
posted by chrismear at 12:10 PM on April 11, 2004


I might add that it is fairly easy to deduce the key in which most western popular music pieces (especially guitar-oriented) are played even without perfect pitch. Instrumentation gives away the answer.

My guess is that 80% of all guitar pop-tunes are in either E, A, D or G minor/major. That's because the guitar is tuned so that it is easier to play in those keys. And you can "guess" the key correctly most of the time by listening to the way the chords are played.

In a similar way most brass music is in F, B-flat or E-flat. The trumpet for example is tuned in B flat. (You get a B-flat natural scale out of a trumpet if you don't press the valves.)
posted by hoskala at 1:52 PM on April 11, 2004


Tlogmer: Wasn't pre-Westernized Japanese (and Asian in general) music almost all pentatonic?
posted by abcde at 8:56 PM on April 11, 2004


Not to mention the traditional folk music of many Eastern European countries, Russia, etc., who used their own very unique scales and tunings, just like Western classical music did back before the advent of equal temperament.

These were all but wiped out a couple of hundred years ago when Western European engineers figured out how to mechanise the process of making instruments like accordians. These were constructed to play on an equal temperament tuning, and basically never went out of tune. And they were cheap. As people detuned their native instruments to fit in with these imports from the west, it essentially sounded the death knoll for widespread use of traditional, native tunings.
posted by chrismear at 3:29 AM on April 12, 2004


More here (self link).
posted by victors at 3:37 PM on April 12, 2004


That's true, abcde. (Chrismear may have answered this just now but I couldn't be sure from his wording. Amazing post, by the way.) You can hear it slightly in in J-pop, though there's a lot of stuff disguising it (especially key changes). Also, I think the christmas music they play in chinese restaurants is more predominantly pentatonic than most western music.
posted by Tlogmer at 6:58 PM on April 12, 2004


*Applauds*

I have long experience and familiarity with the vernacular usage of 'key' (used by folk and rock players to communicate which chord the song relies most on, and frequently begins with) but never had any understanding of how that ties to how it's used in formally-notated playing sitiuations.

FWIW, as a vernacular player, when I use the term I do mean something slightly different - the accidentals and so forth in the scale will get worked out by ear, and so the verbal statement of key might not carry that information explicity. Yet, since most of the playing vocabulary for rock, country, blugrass, the blues, celtic and that complex does rely on first chord majors, I guess it's reasonably accurate.

I wonder, does the formal use of the term stem from a longer vernacular usage, or is the vernacular usage a descendant of the formal terminology?
posted by mwhybark at 8:43 AM on April 13, 2004


Response by poster: For those who really want to get grounded...

MIT Prof Bamberger's book and software should help.

Project puts students in touch with their musical sides

Course OpenCourseWare page.

Book and CD with software at Amazon.
posted by Gyan at 1:39 AM on April 16, 2004 [1 favorite]


you dont need a weatherman to know which way the wind blows
posted by Satapher at 9:29 AM on May 1, 2004


« Older How do I get rid of viruses Norton finds but...   |   Bad links on major news sites Newer »
This thread is closed to new comments.