Why are doubled vocals usually panned left and right?

Question

Why don't people usually have the doubled vocal go down the middle (i.e. two vocal takes in mono) - and then pan that combination left and right using a doubler plugin.

I heard someone say its to avoid phasing - i dont understand how...

Answer 1

I'd question the premise "don't usually pan centre". I'd also question the real efficacy of using ADT rather than just singing it one more time. It's not like most vocal takes are done in a single pass & the singer got too tired to do a second ;)

One thing about panning hard L/R with a lead vocal (or anything meant to be heavily featured) is you don't know what pan law is going to be used if your track is ever broadcast (in the broadest sense of 'being listened to') in mono. That could mean the vocals could be 6dB down in a mono mix & you'd never know if you weren't listening to that broadcast.

Learning to track yourself accurately is a skill to be learned just like any other, but its one well worth learning.
There are tricks you can employ to make the double track sit better - don't sing the heavy consonants t's & p's etc, so you never hear them flam. Definitely don't sing closing consonants on a line, leave the last vowel to over-ring. Sing long end notes slightly longer, then trim them back to the lead vox manually afterwards.
But above all - practise, practise & when you think you got it right, practise some more.

If you're doing 6 or 8 takes on a lead vocal anyway, see what they all sound like played back as a block. See what they're like in pairs. Practise a couple more specifically as a double-track, without consonants.

You will find this learning curve gets more & more important if you ever try to do a full block BV with 30 or even 60 tracks… you really don't want every consonant to sound t-tt-t-tt-terrible.

Another trick - which I would compare* to Monty Pythons "three christs" or T'Pau's China in Your Hands is do three vocals. Main centre, then two subordinate tracks panned slightly L/R. If you want it to sound good & crisp, don't put anything other than compression on the panned tracks, leave all your reverb/echo/whatever only on the centre, main, channel. This trick also gives you far better mono compatibility.

*"The two skinny ones balance out the fat one" and "The two sharp vocals balance out the flat one" Sorry, I was born cynical, but just have a listen to the LVox & DT on that T'Pau track if you can bear it ;) Best heard (with earplugs) on the big money note into each chorus.

Just for fun - and remember this was 1967, stereo was still quite a new thing & people smoked funny cigarettes & ate blotting paper & sugar cubes a lot;) This is my favourite ever usage of hard-panned double-tracked lead vox. They start in unison, but then split out into two-part. I love it.
Of course, it's nowhere near as tight as you'd be expected to do for a modern record - but hey, this was the 60s.

The Herd - From the Underworld

Writing this prompted me to see if I could find an old cover of this I did way back in the 80's. Quickly rescued from an old cassette, home demo only, never recorded for release…

…and, yes, I know this one isn't perfectly doubled either ;) It was just a quick demo, done in a day.

Late Edit
I added this one for why you leave out the consonants when you're tracking up a long way. 30 vocal tracks, 9-part harmony. No effects at all except compression. No cheating on the tracking. All sung one at a time, for real. It's only a 5-second radio sting, so blink & you'll miss it.

Answer 2

This question could have tons of answers, but most of them would be based on psychoacustics (even if they are not aware about that).

It's mostly about "size" and perception.
Think about listening to two sounds.

What would you listen "better" (or, what you could better differentiate)?

• listening to both sounds coming from the same source
• listening to those sounds coming from different positions

The simple answer is clearly the second, but that depends on a sum of factors.

Different sounds (most importantly, similar sounds, including two voices, no matter how different they could be) often have common frequencies that often create "ghost" notes in our heads (look up for "Auditory illusions"). Having [at least] two different sounds coming from two distinct sources (our two ears), makes us easier to make a better distinction between them and allows us to better create the correlation between those sounds.

It's not that different from what the binocular vision allows us (I know, it's not the same, but it's just for the example): imagine seeing an apple on the proverbial table.

When you see the same image from your left eye and from your right eye, you'd have no depth perception. Yes, it's almost the same image, but that's just it, a perception of an apple on a table.
Having different (and, obviously, correlatable) images from both eyes gives you a hugely different perception of what you're seeing.

You can see the size of the apple, related to that of the table; you can "picture" the size of the apple; if you were to touch the apple, you would certainly be able to on the first attempt.

The same happens in the "panning" of similar (but differently-voiced) vocals.

It gives them space.

And a three-dimensional object is usually much more interesting than a bidimensional one.

Having two similar sounds (as in doubled vocals, which share lots of common frequencies), creates more space, and makes it much more interesting.

It is not the same as having a guitar on the left and a drum on the right, as there's much less correlation between them.
Having doubled vocals on different positions makes them "bigger", which, on our minds, also makes them more important.

Answer 3

Doubling vocals (or guitars) panned left/right creates an organic stereo chorus effect in and of itself. If we think that the doubling technique was an effect plugin, it would work like this:

• signal -> organic stereo chorus

Doubling vocals without panning them left/right creates an organic mono chorus effect. So:

• signal -> organic mono chorus

Your suggestion of mono doubled vocals with a stereo doubler effect is the same as an organic mono chorus into a synthetic stereo doubler. It's just a different effect combination.

• signal -> organic mono chorus -> synthetic stereo doubler

Organic stereo tends to be more interesting than synthetic stereo. The amount of phasing depends on many things and you're supposed to detect and fix it, if it's a problem.

Answer 4

Seems that you need to clearly distinguish electronically doubled versus two actual, separate performances. The former will be literal duplicates and the latter will have small variations.

Electronic doubling basically makes a bigger sound. Panning electronically doubled tracks just helps make the sound bigger, fuller. But it doesn't create a sense of two separate performers. Doubling with a harmonic shift seems to make the point even clearer. Doubling with something like an octave effect versus two separate performances/performers in octaves do not sound the same.

Panning left/right two separate performances help spatially separate two performers as well as make a full stereo sound. Two performers can't physically be in the same places. Separating them in the stereo spectrum is a realistic representation.

Why don't people usually have the doubled vocal go down the middle (i.e. two vocal takes in mono) - and then pan that combination left and right using a doubler plugin.

That would put performer 1 and 2 in both right/left spaces and undermine a sense of spatial separation.