Executive Summary of Barbershop, Part 2: the Overtone

In my last post, about my talk on barbershop for Scunthorpe Choral Society, we got to the point where someone asked a really good question, and then it all got too long to answer in one blog post. So we are resuming here, refreshed, and having had a bit more thinking time to consider the question: can you generate the characteristic audible overtones of barbershop expanded sound/lock and ring when making a multi-track recording with yourself?

I’m always a bit slow when thinking about the physics of sound, not least because when given the choice at university, I opted to study Italian for a year instead of acoustics, thinking it would be more useful for a singer. But I’ve learned some stuff since, and my understanding of timbre, vowel perception, and the harmonic series makes me think that in theory, yes, you should be able to do this. The overtones fall well within the range of audible sound picked up by microphones, so the frequencies to be reinforced are clearly present in the sound. Moreover, you’d think that one person singing all the parts has a head start on getting the sound well-matched.

However, I wasn’t hearing that screaming overtone at the upper octave in the last chord of Tony’s tracks for ‘It’s the Music That Brings Us Together’ that would be there if you had a quartet of Tonys singing live. Or even a quartet with one Tony and three other people, as he generally chooses to appear in live performances. I’d heard it from a quintet of Telfordaires the previous Saturday, harmonising outside under England’s current rules for live singing, so I know that chord rings readily. If we could get it ringing, there’s no way it wouldn’t be there on Tony’s voice if it could be: even if you can’t directly hear reinforced overtones in the recording, you can hear that his voice is very overtone rich, ripe to make them happen.

So, I was mulling overnight on this, thinking about both the acoustics end of this and the performance dimension. If the frequencies needed to make lock and ring happen were available, maybe what was missing was the ensemble context. You don’t generate expanded sound in an ensemble just by adjusting your vocal mechanisms, you lead with your collective ears, and once you’ve all locked onto the sound you’re after it becomes much easier to stay with it and develop it. It’s like scratching a cat behind the ear: you go to the right place through your knowledge and experience of feline anatomy and preferences, but the cat cooperates in getting the nuance of placement and pressure exactly right for the moment, and then you are rewarded with the purr.

If you’re singing lines one at a time, to an extent you’re flying blind. You can cooperate with your past self and your future self will in turn work with you, but you can’t all adjust to each other at the same time.

There’s also the question, in the context of recording, of how much you are processing the sound. We’ve all listened to a lot of virtual choirs over the last year, and quite often you hear a very clean sound that has become quite generic – the process of autotuning and other adjustments to clean up the recording has changed the envelope of sound in a way that strips out quite a lot of the individuality and character. I’m not sure if this kind of processed uniformity would mitigate for or against the production of audible overtones, but my guess would be against, if only because automated microtuning is likely to be towards equal temperament rather than just intonation.

In any case, this is something of a moot point for the recording that sparked the question. Any processing of Tony Da Rosa’s voice here has been done with a very light touch: the richness of tone and variety of nuance of his recording are the kind of qualities that would get stripped out in the cleaning-up process. I find I appreciate this a lot more now than when I first heard his recording last May!

And then as I woke up I realised that I had only ever listened to this recording through headphones or the speakers of small devices. The last thing I did before sitting down to write this post was to put it through a really good pair of speakers, and you know what? It’s not a screaming overtone over that last chord, but there is a clearly perceptible ghost of one haunting the ether above that last chord.

So this rather long and rambly exploration can actually come to a conclusion. Yes, it is theoretically possible to create audible overtones while multitracking yourself, but the circumstances of not being able to adjust intuitively in real time to the whole sound mitigates against it in practice. Get a really good singer and some really good speakers, and don’t mess with the recording too much in production and you might get lucky.

But of course the reason that barbershop fetishises this sound so much is because it is a fleeting magic born of the moment. You can capture it in live recordings, but even that sound will only really speak to the soul of someone who has experienced in person the collaborative sonic bonding of coordinating your sense of being with others to create something that none of you could do alone. The overtone is both symbol and embodiment of a deeply social experience, and it is wonderful to be inching our way back to being able to experience it in person once again.