Writing vocal harmonies

This article was originally published in 2004

I. Introduction

It’s happened to you before; you’re working on a song, laying down some great vocal lines, and suddenly it strikes someone that this track would be great with some harmonies. Unless you happen to be (or have) one of those amazing vocalists who seem to be able to pull harmony out of thin air by instinct, this may have been the point at which things got ugly.

Harmonizing a vocal doesn’t have to be painful, my friend. With a bit of theory and the help of a nearby keyboard (or whatever instrument you want to reference pitch with), you can work out some great harmonies; and while many vocalists can come up with a working harmony off the cuff, with a little time and planning you can come up with something a cut above the norm.

This article assumes you know a bit about basic theory, such names for intervals, scales, and some things about triads and seventh chords.

II. Getting started

When you take a theoretical approach to vocal harmonizing, the two most important pieces of information you can have are 1. the notes in the melody line and 2. the chord changes (this is where having a keyboard around can be a huge help). It’s important to look at the melody in the context of the chord changes, because where the melody is in relation to the current chord, and where it is in relation to the current key will both have an impact on which notes you choose for a harmony. Try playing the melody on a keyboard while playing or thinking about the chord that goes with it. This will help you to get familiar with the peculiarities of the melody you are about to harmonize.

Now that you’ve got a sense of what’s going on with the music, let’s look at three basic approaches to harmonizing a melody line: tight harmony, chordal harmony, and contrapuntal harmony (a.k.a. counterpoint).

III. Three basic approaches

1. Tight harmony

Tight harmony is one of the most common forms of vocal harmony in use in popular music today. Most singers who can instinctively harmonize tend to do tight harmony by default.

If you know much about music theory, you might know that Western harmony is built on the “stack of thirds”; in other words, any chord is simply a succession of major or minor thirds stacked one on the other. Similarly, tight harmony is typically done by following the melody line a third above or below. The third is major or minor depending on what is dictated by the key (or the current chord, if there are chords from outside the key).

Another variation of tight harmony uses sixths, since the sixth is the inverse interval of the third (in other words, if you went up a third, then dropped an octave, you’d be a sixth down from the original note. If you went down a third, then up an octave, you’d be a sixth up from the original note). Once again, you would follow using either major or minor sixths as dictated by the key or the current chord.

If you stopped here and didn’t read another word, you could probably do an adequate job of harmonizing most melodies with just tight harmony. But there are situations where sticking to thirds and sixths just doesn’t work out. We’ll look at some of those situations and talk about what to do with them later.

2. Chordal harmony

In sharp contrast to the tight harmony approach, the chordal approach virtually ignores the melody and simply picks chord tones from the current chord to harmonize with the vocals. The note of the harmony line stays the same until the chord changes, and it’s always on a chord tone.

This approach can be troublesome; it works great for building vocal pads, or when you want to add a third part when you’ve already got tight harmony going, but it can get bad when you are trying to harmonize one or two parts with the melody. Melodies move, and often they hit passing notes that don’t blend too well with the actual chord tones. This usually isn’t a problem when it’s an instrument playing the chord tones, or a choir of voices, but if you’ve got just two or three parts (especially of the same vocalist) you can wind up with some nasty dissonances if you aren’t careful.

3. Contrapuntal harmony

This is by far the most sophisticated approach to harmony, and takes the most work to get right. In counterpoint, you are creating a counter-melody whose rhythm and movement is not necessarily the same as the main melody. This can be trickier than it sounds, because to make a really good counterpoint line, you’ve got to keep in mind not only where you’re at with regard to the melody and chord changes, but you’ve also got to keep in mind the rules of good voice leading and melody construction to create something that is both complimentary to the original melody and pleasing in its own right.

Of course, you probably won’t find yourself writing full-blown counterpoint harmony to beef up a rock or pop vocal; the important this to grasp is the concept that we can move the harmony voices in different directions, different intervals, and at different times than the melody line. All of these approaches are extremes; in practice, we can borrow a bit from each to construct simple but effective harmony lines without agonizing for hours over them. Many arrangers start out with one approach in mind, and modify it with the other approaches to get out of trouble spots. We’ll look at some common trouble spots and how to deal with them, but first, we’ll talk about good voice leading.

IV. Voice leading

Voice leading is an art that dates back to the Baroque era, and many a freshman music theory student has agonized over its complex rules re-harmonizing Bach chorales. Because these rules are often taught in the context of 17th and 18th century music, and because rock musicians generally have a problem with any kind of rules, many musicians mistakenly assume that voice leading is irrelevant to modern popular music. But if you take the time to learn and follow these rules, you’ll find that they are still very relevant and will help you construct rock-solid harmony every time. I won’t go into a full-blown description of all the traditional rules, but I will touch on some ideas that I feel are most important and most likely to come up in popular music.

1. Avoid parallel fifths, fourths, and octaves

Parallel fifths are one of the biggest killers of a good harmony line. A parallel fifth is a situation where the melody and harmony create and interval of a fifth, and both move in parallel so that they again are a fifth apart. This gives a kind of Gregorian Chant effect and has a weak sound to it (although some styles of music have used parallel fifths specifically for the ominous effect of it, notably “Grunge” or “Nu-metal”). As the fourth is simply the inverse of the fifth, parallel fourths are also to be avoided.

The parallel octave is considered a weak movement, though I should explain that we’re not talking about a situation where the melody is doubled at an octave above or below, which is a common arrangement technique in pop and rock music. There’s nothing wrong with that. But in the context of creating a harmony part, you want to avoid hitting a parallel octave. In other words, if we’re moving along using fifths, sixths, and so forth, and suddenly use two octaves in a row, you will find it a weak and unsatisfying harmony.

In fact, pretty much the only intervals that can safely move in parallel are the third and the sixth. The only intervals we haven’t mentioned yet are the second and the seventh, and I have a hard time imagining a situation where you’d even want to attempt to move them in parallel.

2. Resolve tritone intervals and leading tones properly

The tritone (a.k.a. diminished fifth or augmented fourth) is the dissonant interval that drives Western harmony. If you know much about the resolution of V7 to I, you know that there is a note called the leading tone, which is a half-step below the root of the key, and that a tritone away from the leading tone is the 4th of the key, which is the 7 in the V7 or dominant chord.

If this is Greek to you, simply understand that if you have a dominant 7 chord, the third of the chord needs to resolve up a half step, and the 7 of the chord needs to resolve down either a half step (if it’s resolving to a major chord) or whole step (if it’s resolving to a minor chord). This means that whenever you encounter a tritone interval between your melody and harmony (or between two harmony parts), it’s going to either resolve inward to a third, or outward to a sixth, depending on what chord it’s resolving too. If you get into jazz harmony, you may find exceptions to this rule, but that’s beyond the scope of what we’re dealing with here.

3. Use contrary motion when possible

Contrary motion is when two of your voices move either towards or away from each other rather than moving parallel. This is a very desirable effect when harmonizing, though it has to be planned carefully. You can quickly wind up at a very ugly interval if you don’t turn it around at the right point.

One of the most common uses of contrary motion in writing vocal harmonies is as a transition between tight harmony in thirds and tight harmony in sixths, or vice versa. This can be a powerful technique to overcome the limitations of simple tight harmony.

4. Don’t get too far apart

With all this contrary motion going on, you have to be careful not to let your voices get too far apart. You should avoid letting two adjacent parts get more than an octave between them, and there really isn’t much call to even get beyond a sixth apart. If you have three part harmony or more, of course, it’s not a problem for the outer voices to get further apart than this, but no two adjacent voices should.

5. Avoid crossing voices

Crossing voices means letting a harmony below the melody cross over the melody and go above it, or vice versa. Sometimes it’s unavoidable (if you have a particularly soaring melody, for instance, with several intervalic leaps in succession), and on rare occasions it can actually be quite nice. For the most part, it’s best avoided, especially in pop productions where the same singer is likely to sing all the harmony and melody parts.

6. Be careful with seconds and sevenths

By themselves, seconds and sevenths tend to be blunt, ugly intervals. In the context of a chord, they can be powerful and haunting. For example, a major second, with a fourth on the top, becomes a nice suspended 2 chord (or suspended 4 chord in inversion). Just use your ears and be mindful of the context of these intervals in the chord.

V. Examples of common trouble spots in tight or chordal harmony, and how to work them out.

Here are just a few odd examples of “tight spots” I’ve run into when writing simple vocal harmonies. While you may not run into these exact situations, hopefully reading this will give you an idea of the kind of thought process you go through when writing a solid harmony, and maybe some ideas of different things to try that might prove more musically pleasing than the quickest and simplest solution.

1. Melody on the fifth of the chord using tight harmony above

Supposing you’ve got a song in G Major, the first chord is G, and the first note of the melody is D. You want a tight harmony above the melody. Your first harmony note, then, would be F#. Except, that would make your first chord Gmaj7. Nothing wrong with Major 7th chords, but they have a distinctive sound that isn’t always appropriate, especially on the I chord.

The simplest thing to do is start your harmony on G instead, harmonizing at a fourth. The trouble is that most harmony singers instinctively want to sing with the melody, so you just have to stay on G or make a leap off of it to make sure you don’t end up with parallel fourths.

Another option might be taking the harmony up to B and using parallel sixths (if the vocalist’s range can handle it), or even hitting the A (making the chord an add9; for whatever reason, I find that the ninth has less effect on the character of a chord than the 7th). This might be a good option if the melody moves up from the first note, because you can use parallel motion and get back to thirds in a pretty nice way. If the melody moves down, you’ll have to jump to avoid parallel 5ths or moving too far apart.

2. Melody resolves to the root with a lower tight harmony

This is a problem I hear all the time, particularly with improvised harmony. You’re doing tight harmony in thirds below the melody. The melody resolves to the root of the key. Let’s say you’re in C. When the melody hits the C, your harmony is going to be on A. Oops; that’s not resolving. What do you do?

The first and most obvious solution is to sing a G there instead of an A. If the melody is walking up to the C, then the second to last note of the harmony is a G already. Not the most exciting resolution, but it will work. If the melody walks down to the C, the second to last harmony note was a B. Here’s a problem. That B wants to resolve to a C. Resolving it down a third to the G will work OK, and if you’ve got a band going behind you, it’ll probably pass. But it’s a weak resolution, and a bit awkward.

I would think the best solution in this situation, provided it keeps the harmony in a good range for the vocalist, is to back up a bit and look for an opportunity to use some contrary motion to get the harmony to a sixth below instead of a third. Or, if the melody walks up to the C, get that second-to-last harmony note on an F, and resolve it down to E. Much more powerful and effective resolution than ending up on the G.

3. Tight harmony where the melody is on a non chord-tone

Melodies don’t always start or stay on chord tones, and songs don’t always stick to one scale or mode, even over the course of a phrase. I was working out a harmony once on a song where the first line of the melody was all on a D being sung over an Am chord. It got a bit tricky to harmonize, since we wanted an upper harmony on it. Complicating the matter more was the fact that the song changed freely from natural minor to Dorian mode (i.e. — it had both a D major chord and an F major chord in the progression), so it was unclear whether F or F# was appropriate to use on the rest of the progression. Clearly, an F natural wouldn’t work over the Am regardless, and the F# would give it the very distinctive minor 6 sound that isn’t always appropriate, even for a song in Dorian mode.

The best thing to do here is to decide how high you want to go with this harmony, and how far you want to extend the chord beyond the triad. Raising the harmony to a G creates an Am7 chord (of course the melody is adding the 11). Raising it up to an A leaves your harmony in the triad. In either of those cases, you’ll need to exercise caution to avoid parallel 4ths or 5ths.

Another solution is to go a sixth above the melody to the B, adding a nice 9 to the chord. Of course, if your melody goes up from the D, you’ll either need a vocalist with a great upper range, or a little contrary motion (i.e. descending, in this case) to get you back to thirds.

4. Chordal harmony where the melody hits clashing notes

As I mentioned before, the problem with chordal harmony is that very often the melody moves into a place where it clashes with the chord tones a bit. This doesn’t seem to be an issue when the voice is against instruments, but against other voices, the dissonance seems more noticeable.

A simple example: we’re singing a melody over D major. You’re harmonizing it with F# and A. The melody walks up from D to A, with the A hitting on the next chord change. When we hit the E, there’ll be a little dissonance with the F#, but nothing terrible. When we hit the G the dissonance will be a bit harsher. So we want to do something about the F#. You could move it to G when the melody does, but we’d rather not have our melody and harmony in unison. A better solution is to move the harmony down to an E when the melody goes to G. Of course, in this case the F# will still be unison, and the voices will be crossing. With chordal harmony, you’re almost guaranteed to have voices crossing, so we can let that slide for now.

Another solution is to have the harmony hop down to D when the melody hits F#. Again we’re crossing voices, but that’s almost unavoidable in this case. We’ll also have a little dissonance when the melody hits E, but that’s doable. But when the melody goes from G to A, we’re going to run into a problem with the upper voice. It’s going to have to get off the A before the melody gets there, but where to? Suppose, to make it interesting, that the next chord is G, making the melody note (A) a 9th. What to do with the harmony now? If we stick to the chordal approach, we’ll go to B. Not only will this leave us with a major second between the melody and harmony, but we’ll have come from a major second, giving us (yuck!) parallel major seconds. Not good. Our only real option, if we want chordal harmony, is to launch it up to D; rather a clumsy harmony line…

Because you can easily end up in situations like this, I don’t generally recommend using the chordal approach unless you can keep the harmony out of the way of the melody, either by putting it in a different octave, or by mixing it in such a way that it’s perceived quite separate from the melody line. It’s better to take a chordal approach for short spaces here and there where tight harmony might fail you, or by combining it with tight harmony in multiple voices (for example, if you look at the four part harmony used in many traditional hymns, the soprano and alto voices are in tight harmony, while the bass and tenor take a chordal approach), or by using it with a vocal “pad” (where the voices are singing vowel tones instead of lyrics, thus separating them from the melody). We’ll discuss pads below.

5. Resolving a diminished block triad.

As we discussed under voice leading, tritones always resolve a certain way — either inward or outward in contrary motion. If you’ve got a 3 part harmony, you can run into trouble because of this fact if you’re harmonizing either a diminished triad, or the top three notes of a dominant 7 chord (which is simply a diminished triad). Say you’re in G, and you’ve got the vii° triad F#-A-C as your harmony. The F# wants to resolve to G, and the C wants to resolve to B. So where does the voice with the A go to? Either way you go, you’ll be in unison with one of your other parts. While this is one option, it’s unfortunate that you can’t make good use of all those harmony voices. You could cross voices, but we try to avoid this. Perhaps the best thing to do is re-voice the harmony.

Classical harmony often used the diminished chord in inversion, with the third in the bass, which in this case would make our chord A-F#-C. We can see why now. The triad would resolve to G-G-B (with the two G’s an octave apart), a much stronger resolution than the others we’ve proposed. The only problems you face doing this would be that the A might be your melody line, so moving it might be out of the question; or there’s the fact that pop music generally doesn’t use the bass voice, so your lower line might get too low to be acceptable for pop or rock. (of course, you could move the F# up an octave too, though this would mean having two upper harmonies for the whole phrase, so you don’t cross voices. If you have an upper and lower harmony on a song, you’ll want to keep it that way. Chances are you’ll run into this sort of problem a lot, though).

VI. Some easy-to-apply contrapuntal techniques

While a complete study of the art of Baroque counterpoint would probably be a waste for most pop and rock musicians, we can certainly draw on the basic idea of counterpoint to take our background vocals to the next level. We’ve already discussed the use of techniques like contrary motion and changing the intervals in tight harmony; here are a few more simple ideas to get a more contrapuntal feel.

1. Simplifying the rhythm

When you’ve got a melody that’s syncopated or that moves fast, you can create a nice background part by simplifying or slowing down the rhythm and singing it against the melody in chordal harmony. You may have to drop words or phrases in the lyrics to make it all work out together, and sometimes this can work out nicely from a poetic standpoint (kind of like the background vocals are giving you the summarized version while the lead vocal explains the details).

2. Complicating the rhythm

Alternately, if you have a slower, smoother melody, you can create a faster rhythm for your harmony part to sing. You may have to add some lyrics here. Chordal harmony works great here as well. Notice how much better chordal harmony works when you aren’t singing tight with the melody rhythmically.

3. Echo and hold

This is a very old technique used in everything from barbershop quartet to doo-wop to Queen to modern vocal pop. It works well when you have a melody with a significant break at the end of a phrase (like 3 or 4 beats) and with at least 2 harmony parts. Basically, just like vocal pad, you hold out “oohs”, “aahs”, or even the first word of the phrase in a chordal harmony until the break; but then you echo the last line or last few words of the phrase in tight harmony (letting one of the harmony parts repeat the melody; alternately, you can echo just with harmony lines, omitting the melody line in the echo). When the melody comes back in, you let the harmonies “hold” again until the next break.

VII. Advanced Vocal “pads”

The basic vocal pad is just 2 or 3 vocal parts holding out “aahs” or “oohs” in chordal harmony. Of course, you can do this with just about any synthesizer, but the results when doing it with a real vocalist can be much more satisfying, especially for more organic styles of music. And there’s nothing that says all the voices have to change at the same time or stick to block chords. If you’ve got a songs that has that “chord-chord-chord” feeling, this can be a great way to smooth it out. Take advantage of passing tones, suspensions, common tones, contrary motion, and other voice-leading techniques to diminish the “blocky” sound that straight chord changes can often give. Here are some examples:

Passing tones: Say you’ve got a progression going from C to G with two-voice harmony. You are harmonizing the C with C and E. You could walk the upper voice to a G (to harmonize the G chord) by going E-F-G. The F is called a “passing tone”. The C could move down to B in this case, giving you G and B over the G chord. (Note– this is also an example of contrary motion).

Suspensions: This time you’re going from C to F. You’ve got 3 voice harmony, singing C-E-G over the C chord. You could move the upper voices up to F and A to harmonize the A chord (taking advantage of the common tone in the bass), but instead let’s just move the E to F and leave the G until halfway through the next measure, when we’ll move it to A. This creates a suspended 2nd, and breaks up the movement of the chord a bit.

Common tones: We saw in the preceding example the use of a common tone in the bass. If you wanted to smooth out that chord change even more, move the C to the upper voice (so the harmony would start E-G-C, the C being up an octave from the last example). Thus, between the two chords, the upper voice stays on C, drawing attention away from the change and smoothing it out. Common tones are even easier to use when you employ chord extensions– 7ths, 9ths, 11ths, and 13ths — in addition to the triad notes, because very likely at least one tone in a chord will be an extension tone in the next chord. In the C-F example, you could just hold out the C triad over the F (provided you had other instruments establishing the change to F, and the vocal harmonies were high enough to be extension tones and not clash with the triad tones), making it an Fmaj9 chord.

Contrary motion: As I’ve mentioned before, contrary motion is when two parts move away from or toward each other rather than parallel to one another. It’s a very strong movement and usually very pleasing to hear. A common place to encounter contrary motion is resolving a tritone, as we discussed under voice leading. Tritones resolve in contrary motion (when they’re resolved properly, anyway), either in or out depending on how the chord is voiced. For example, in a D7 chord harmonized D-F#-C, the tritone would resolve inward to make a G chord. If it were voiced, D-C-F#, the tritone would resolve outward.

Using these voice leading techniques and others, you can get a better control of how the harmony moves in your music, whether you want to smooth out a blocky progression or liven up a static phrase. These techniques work not only for vocal pads, but also for arranging string or brass lines, synthesizer parts, guitar harmony, etc.

VIII. Parting words

Great vocals and harmonies are a wonderful way to spice up an arrangement. If you want to learn more, check out a good music theory book and read up on chorale writing. Remember to always keep in mind the range of your vocalists, and what’s appropriate for the style of music. Happy harmonizing!

 

If you’re interested in recording, songwriting, and composing, you might enjoy my “One weekend recordings” series on Youtube.  Thanks for checking it out!

39 Thoughts on “Writing vocal harmonies

  1. Ephraim Tabackman says:

    Thank you so much for posting this. I had been searching the web unsuccessfully for quite a while for this kind of no-nonsense detailed info about vocal harmonies.

    This is perfect for my level (i.e. just a bit beyond my comfort zone), and I am looking forward to working through paragraph by paragraph and trying out each example to understand it.

    MUCH appreciated!

    A question: Is it possible to generalize about certain styles of music or artists, and say they usually/typically used this type of harmony to get their sound? Some examples would be the Beatles, Crosby Stills and Nash, Gospel, Country…

    Ephraim T.
    Efrat, Israel

    1. Alan says:

      I’m glad this was helpful to you, Ephraim.

      Regarding your question, I think the answer is “yes and no”. There are certainly styles that have very distinctive vocal harmony idioms (barbershop quartet, “surf music”, do-wop, etc), but probably in most cases it’s just harmony doing its thing within the melody and chord structure of the song (which itself might be particular to the genre).

      The rules don’t really change (though they stretch a bit when you get into jazz or gospel, e.g.), but what probably distinguishes say, the Beatles from Crosby Stills and Nash would be things like:

      – How many parts (2/3/4 part harmony)
      – Whether the parts are above or below the melody
      – The spread of the parts (thirds, sixths, etc)
      – Whether the harmonies are tight to the melody or contrapuntal.

      You just have to listen and analyze a bit. Keep in mind, some distinctive vocal sounds are just a factor of people having distinctive voices. Simon & Garfunkle weren’t doing anything amazing from a theory standpoint, they just had voices that blended well and a whole lot of well-crafted songs to sing.

  2. Nick Renna says:

    Thanks Alan! I’ve already put what I’ve learned here to good work. It’s given me some nice ideas as well as enlightened me as to what’s going on vocally in the pop and rock songs I cover. It has always been those lower and middle harmony parts that gave me fits. Now I have some frame of reference and ideas on where to start looking to figured the parts out. It is definitely not something something that comes naturally to me. This will save me a lot of time.

    Best regards,
    Nick

  3. Parag says:

    Awesome stuff!! thanks a lot!!

  4. JF says:

    Great post! Just at my technical level too, how lucky. Thanks!

  5. Toun says:

    Realy helpful! Thank you so much!

  6. Emma says:

    Quick question! say the chord is Cmajor and the lead vocalist’s note begins with an E (the 2nd in the Cmajor chord) Now if the the harmony part is to sing a perfect fifth (inverted) so inverted fifth, would the harmony still work with the chord since now the harmony would be singing something that is not the typical 1-3-5 for the chord?

    1. Alan says:

      Well, that would add a B to the chord (fifth from E), so it would alter the character of the chord to a Cmaj7. Why would the harmony need to be an inverted fifth?

  7. Emma says:

    WHOOPS! I meant the 3rd in the c major chord!!! Not the 2nd!

  8. Greg Filian says:

    The female vocalist in my band and I are having a spirited discussion about the harmonies in Lady Antebellum’s Bartender. She says parts of it are a no no according to your article.
    Can you comment?
    Thanks, Greg

    1. Alan says:

      I’m afraid I’m not familiar with the song in question; can you give me an idea of what the possible “no no’s” are?

      Your question, if I may draw some conclusions, hints at an age old battle between musicans who adhere to theory and musicians who reject it: If good/popular/respected piece of music X violates theory rule Y, is theory debunked or is music X not really so good?

      The answer is neither.

      Music theory doesn’t define good music from bad, nor right music from wrong; rather, it outlines the expectations of the listener. When you play Dm -> G7, most listeners expect a resolution to C major. This is theory: V7 resolves to I.

      But as a musician or composer, I can use my understanding of the expectation and play off the expectation to give the listener something unexpected, such as resolving the chords to an AbMaj7.

      So yes, you’ll hear music that has “no-nos” in it. Some of it out of ignorance or disregard of theory, some because the writer is playing with your expectations or employing a certain musical effect.

      Bottom line: let your music be informed by theory, but not ruled by it.

    2. Mike Norris says:

      To Greg Filian, regarding Lady Antebellum’s “Bartender”:
      The harmonies in this song do contain a lot of parallel fifths. Lady Antebellum’s harmonies are often based on bluegrass vocals (although the song doesn’t sound like a bluegrass song), which mostly use “stacked thirds”, which could also be called “parallel triads.” Parallel fifths are built in — doesn’t make it bad or wrong, just a different tradition from choral music.

  9. alex says:

    Hello Alan,
    I think your article above is fantastic. I’d like to include this in reading for my students who study vocals here at Nexus-ica in the UK. Would this be ok for me to include this article and point them to your site?

    We are a faith based organisation and looking at harmony is an area we do practically and theoretically and I think your article here is an excellent resource.
    alex

    1. Alan says:

      Thank you Alex, I appreciate the vote of confidence.

      I don’t have a problem with this being included in classroom materials as long as (1) it’s properly attributed (my name & this website should suffice) and (2) it’s not part of a text that’s being sold.

      I suppose that’s properly a CC-BY-NC license (see https://creativecommons.org/licenses/by-nc/3.0/us/), so we’ll go with that.

  10. Roy says:

    I just wanted to say “Thank You” for such an informative article written in such an understandable way. I am just starting up a small choir in our congregation and really needed some help – I now have it. I was finding it very difficult to transpose the piano music accompaniment to two or three part vocal harmony but can now see my way forward. Once again a big “Thank You”.

    kind regards

  11. Zach says:

    Great info on vocal harmonies! Much appreciated

  12. Brad says:

    Fantastic! I was just listening to “Our Prayer” by The Beach Boys, and it reminded me how much I adore vocal harmonies. Thanks for the informative article!

  13. Lola says:

    Thanks for this article! I’ve been also looking incesantly around the web for some clarification. I recently came across this video https://www.youtube.com/watch?v=EEnkAwNnz1o
    What do you think?

  14. This was so helpful. I’ve been writing songs for a long time, but only recently getting into the studio to record and flush out these songs to their fullest. Of course, that means writing harmonies, and I do find I naturally lurk around the thirds and sixths, but was looking for advice to branch out a bit, especially for my song with slightly uncommon chord progressions. Thank you, thank you!

  15. Nielen says:

    A very cool piece of general guidance for those of us who try to make the vocals in a band interesting and powerful – thanks!

  16. Matt says:

    really, really helpful. thank you!

  17. James Ferguson says:

    Very nice post! Thank you. So… The third below harmony I always thought of as the lower third. Meaning E below C. A minor 6th interval. Are you suggesting to sing a third interval below the melody note? If the melody note on a C chord was a D for example, I would sing a B? That would make a Cmaj7 with a lower 7th. Yuck. I always figured 4th were safer since most chords have perfect fourths and fifths it was safer and supportive to sing the lower fifth. I gotta hit the shed.

  18. Peter Ryan says:

    Fantastic article! Do you have any general tip for situations in a rock group with 4 vocalists. What do you do with the 4th part? Low bass parts tend to muddy things up and adding another part above is often too high, so we usually end up doubling one part. But which part?

  19. rosan says:

    Hello, do you give classes? could we have a skype lesson?

  20. ChrisP says:

    This is exactly what i was looking for, thank you !!
    If you ever write a book on this subject, please let me know 🙂
    Greetings from germany,
    Chris

  21. martin says:

    Are there more than the 4 basic tones in doo-wop harmonies when 5-7 singers are singing or are they all just singing ones threes and fives etc.?

  22. Michael W says:

    Alan, thank you so much. So very helpful!

  23. isaac says:

    thank you alen for the great piece of info
    alen please help me out with a video versions of this article. am a visual person
    and i understand better when i see it done. am very grateful

    1. Alan says:

      I appreciate your response, but, sorry, no video version will be forthcoming.

  24. PickerDad says:

    I’d like to throw in an 11th reason I love guitar, but it is only applicable to acoustic guitar. I see that you’re an electric guy, and probably most of your following is electric too, so some of this will be out in left field to some of you, but I’m going to make the case for acoustic guitar. I might occasional use my electric for a particular sound appropriate to the song, but to me it always starts with my luthier-made acoustic instrument. It needs additionsl skills in making the sound of each note, which can be frustrating, but once I master the full song- and note-making skills available, the feeling of making music directly from my hands to the listener is all mine. It cannot be duplicated or shared.

    The biggest reason I stick to acoustic guitar is that it’s sensual. Now don’t go off on some sexual trip. When I’m playing anything, I am one with the guitar, in a much deeper way than on an electric. That big 4-5″ body is an acoustic resonator; it is the one and only thing that shapes the music. Every sound I produce into the air is also vibrating in the same way through my arms, hands, chest, diaphragm, vocal cords, ears, and skull in the same instant. With an electric, I’m using this terrific tool to make the music, but with an acoustic, I AM the music. There are thousands of ways I can shape or refine the music from within. The music is directly vibrating in my brain, in real time, and I have all of these thousands of possibilities within my fingers to use instantly, in real time, and it all comes back to my body in real time, and I can adjust it further with my fingers within the same note. When I create a particular note with a particular sound, I’m not setting it up beforehand, or pushing buttons to shape it, I’m doing it with my own hands, feeling it in my body, and hearing it directly from guitar to ears. I’m playing with all its subtleties directly in my brain, even before it is cast from strings to soundbox to air to eardrum to brain.

    With electric guitars, the sound the music does not have to be in the air at all, it is an after-the fact combination of what I’m playing plus the electronic modification and the characteristics of the speakers (which one could suggest that’s a bigger accomplishment than controlling just one guitar). It’s not better or worse, it’s different. Vive la difference.

    Reason 12 I love (acoustic) guitar is that it is created as a combination of the shape and structure of the instrument and the effect of natural materials like wood, glue, stone or ivory, the spacing of frets by the luthier, and the way the note is held in my left hand and struck by my right hand, the pick I use or the nails of my right hand. It may be recorded electronically, but the recording is designed to reproduce exactly what was in the air, without any coloration. There are times the recording is deliberately changed by the recording process, but they are openly and obviously declared, and can be reproduced by equally skillful players and recording engineers.

    With electric, there is no music until it’s been modified by the electronics, the electronic settings, and then released to the air with the colouring of the particular speakers used at the time. For acoustic guitar, I am the only person making the music. I directly control the sound and the way it is coloured by the instrument. With an electric guitar, the final result is shaped only after it leaves my hands, by soulless electronics and speakers. I am only one participant in shaping the music. Granted, I may be the one who sets the electronics, but that is a separate action, at a separate time, without note-by-note adjustment, and can be copied exactly by anyone else.

    I’m not suggesting there is anything wrong with electric guitar. They open up entire dimensions of sound that are not available from an acoustic guitar. I am only saying that I love the sensation of the final music being under my control, and to feel it in real time in my fingers, my chest, and my brain.

  25. Sara says:

    Is there a work book that you know of that can help revise all of this?
    Also I am wondering if there is another book that you can recommend
    to compliment this because some of the concepts that you gave are easier to understand if I see examples of it in musical note form on the staff.
    Thank you.

  26. Aly says:

    Hi Alan. Thanks for the info. This was very helpful. My music teacher came up with a finale project where we have to arrange our 7th and 8th songs for our school’s annual spring concert on our own. I was placed in the arrangers/musicians group so I needed a little help on where to start. Thank you again!!!

  27. Cyrus says:

    Chords C Em Am F. 1 bar each.
    Rock melody bar 1
    C C C D E
    Bar 2
    B B A G
    First bar I can harmonize above in thirds but next chord melody is fifth above B over Em so I thought to stay on G harmony from the last chord which is the sixth?
    Second question: with rock vocals do you usually go parallel thirds most of the time?

  28. dave coulter says:

    I’m frustrated trying to sing backing vocals in a band I play with (bass guitar). Say we’re doing a Bob Seger song, what happens is our keyboard player/female vocalist (pretty good on both) insists that we other players sing harmonies based on the notes in the chord but, when we get her interpretation right it sounds like a “show tune” and not a rock song at all. When we tell her this and say we prefer our natural feel of the harmonies. She keeps saying but you guys are singing unison and you have to have the other chord notes in there. When we do what we feel(or sounds to us is right) the song sounds a lot more like the recorded song. Anyone have thoughts on this?

  29. Moa says:

    Can you give an example of a grunge or nu-metal song that uses a fifth harmony? interested to hear what it sounds like. Thank you!

  30. David says:

    Wow. It’s a great article. That is really fascinating. It’s nice to see it. Nice written and come with almost all important information for writing vocal harmonies. I love this. I would like to see more posts like this . Thanks for sharing this information.

  31. gregory macmillan says:

    very sound article, with proper information clearly presented, and humanist as well, with little hints of musical and historical context thrown in. Excellent! Many thanks.

  32. F says:

    Kaspersky anti virus is the whole time that i am on your website busy blocking a malicious link https://cdn101.st… , Why ?

    1. Alan says:

      My apologies, the theme I have been using for several years decided to go rogue and put a mining link in the footer. I have removed it and created a custom theme that I fully control. I appreciate you letting me know.

Leave a Reply

Your email address will not be published. Required fields are marked *