Tags and Tessitura

‹-- PreviousNext --›

As you’ll know if you’ve either read my first book or hung out with barbershoppers for more than five minutes, barbershoppers do like tags.* They like to sing them all night as a social activity, and when they have to sit in an audience and keep quiet, they like the people on stage to sing them for them. Hearing a someone nail a good long post and some serious chord worship gives a particular style of vicarious pleasure that is amplified by all the hours spent in stairwells attempting it yourself.

This is, I suspect, the reason why the genre has developed the phenomenon of the ‘out of context tag’. The arrangement charts its way through the journey of the song, and just as it is heading into where it should culminate, it suddenly dives off into a screaming tag from nowhere. An outsider might think: why would you do that? But an insider knows they do it because the other insiders in the audience will respond with delight.

Now, back in my days in as a Music Category judge, there was some discussion about this, with the aim of triangulating between the quality of musical choice (did the tag fit the song?) and the quality of the performance produced (how well did the ensemble sing it?). Kevin Keller produced the following rather useful matrix to help judges decide how to inflect their score in such circumstances:


My recent listening has focused on the top right box: how often you can tell an out-of-context tag by the way the performers fall off it. A group will be going along perfectly happily, and the audience with them, but when they dive off into what is supposed to be the big finish, they don’t quite make it.

Typically, this is seen as a failing of the ensemble - that they’re good, but not quite good enough. But I am starting to recognise certain features in the arrangements that seem strongly correlated with difficulties in performance. I think the fault may not be with the performer, that is, but with the arrangers.

(Or whoever decides to use that tag, of course. It is one of the irritating facts of an arranger’s life that they spend time and effort crafting a tag that will finish a song effectively, then either the performers or their coaches decide this isn’t big enough for their egos, and rewrite it. And then botch it in performance, and the arranger sighs. Ah well. They botch the ones we write too, so let’s not get too up ourselves here, eh?)

So, the particular musical shape I have heard mishandled frequently enough to generalise about goes like this:

  • Coming to the end of the melody, the tenor goes up to the top tonic, and holds it as a post while the lower parts do some business
  • The business resolves with the bass on the tonic an octave below, which is held as a new post
  • The chord is restacked above this, voiced a 10th, with the tenor on the 10th, lead on the upper tonic, and the bari on the 5th. None of these sound particularly secure, but the 10th is the one which is conspicuously strained

Now, you would think that a final chord voiced at a 10th would be more secure than one voiced at a 12th (the 1-1-3-5 voicing, designed to give a mega overtone at octave above), but in fact, you hear this one botched less often. Or, at least, I have heard it sung more securely in the recent performances I am reflecting on here.

And I think issue lies not with the voicing itself, but with its tessitura relative to the rest of the song. (You knew I was going to say this from the title of this post, didn’t you?) The problems arise when the tag breaks out into a pitch space that’s higher than that which preceded it the song.

Of course, that is what the intention is - to move into a new, higher world for a big finish. But if the singers aren’t prepared for it vocally, and the music isn’t prepared for it harmonically, it doesn’t work. In particular, if there has been a clear cadence taking the tenor up to the top tonic as a way to finish the song, this creates a very strong pitch ceiling that contains the music. The subsequent move that breaks through into the next octave is thus not just vocally higher, but is musically struggling against the processes set up within the song.

So, how do we set up our big tags to succeed rather than to fail? Singers and audiences aren’t going to settle at the top tonic just because not everyone can break through into the next register safely. If anything, the fact that not everyone succeeds at these makes them more attractive: the thrill of the risk has always been a major pleasure in the tag.

I think the key thing is exactly that of setting up the tag. There needs to be some sense of need, desire, hunger pulling to up to the final voicing, a sense that the music isn’t finished till you get there.

(This is proving a time-consuming post to write, because I keep thinking of arrangements I’ve delivered in recent months that I want to check will work now I’ve figured this principle out!)

Specifically, I think the key thing is that, whichever register the final chord sits in, every part should have visited that part of the pitch space at some emotionally important moment during the song. That pitch space, and expressive capacity of the voice need to have been opened up and brought into use before you get to the big finish, so coming back up there at the end gives a feeling of completion.

Given the trend for humungous intros these days, it could be there - ending the intro in a register that you next visit in the tag. Or you could build a process of gradual growth, such that you work you way up through tessituras, getting nearly to your end chord but not quite as the last thing you do before the tag.

Even better: make the places where you visit that register earlier harmonically unstable. So, that when you get to the tag, there is the feeling that last time you were up in that emotional world (whether all the way back at the start, or on your journey through the song), things were unsettled, you couldn’t stop there. Then, when you get back there, you will have a sense of harmonic resolution as well as coherence of register.
What I find interesting in thinking this through is that the same music-psychological processes that help the singers succeed in getting to the end of the song safely are central to audience satisfaction. The music is what mediates between voice and emotion; it structures experience both physically and aesthetically.

A tacked-on screamer of a tag may be fun to sing (if you pull it off under pressure), and exciting to listen to (likewise, if the performers pull it off), but it’s a bit like empty calories - the pleasure’s all in the moment without much lasting effect. The same musical gestures crafted as integral to the song carry meaning and thus offer greater satisfaction.

*The tag is a song’s coda. I always think of it like casting-off in knitting: it’s there to make sure the song finishes properly and won’t unravel from the end backwards with repeated use...

Of course, not every tag has to be a 'screamer' either. I think arrangers also need to consider the age range of the people they are writing for. Just because you can sing up in the stratosphere, it doesn't necessarily mean that you should. I am reminded of "From the First Hello to the Last Goodbye" which has two possible endings. It would be good to see some arrangements with alternative tags - particularly, when an arranger has produced something with a specific performer in mind.

It's interesting, though, that if you try and deliver a tag that strays from certain standard gestures, the performers clamour for you to change them. I have at least trained most of the people I arrange for these days to talk to me about it rather than just changing them, so I can help them avoid the traps I've started to identify here.

Indeed, the published ending to Happy Together is, in my view, incongruous and stereotyped, but the editors insisted their customers would want it like that. The original ending was far more like The Turtles' approach. It seems to be popular though, so maybe they were right about their customers, but I still use the original version with Magenta :-)

Oh, you've just reminded me of a quiet tag I wrote for an arrangement that got extended by the performers or possibly their coach in the opposite direction and ended up with them gravelling about on a note a bit beyond the realistic range for the bass voices involved, and way out of tessitura for the rest of the chart. Ah well.

Anyway, it seems the tessitura implication-realisation thing may generalise beyond the specific cases I was thinking of while writing this. Thanks for helping me realise that (even though I know that's not what you thought your comment was about!)

I think this idea of tessitura and/or ranginess (as I describe it to some) is important. A lot of recent arrangements seem to be placing the baritone part higher, and expecting more of the lead too. I wonder if this has come from introducing younger singers. If so, fine.
On the other hand, some/most of the songs aught to be capable of being (re-) arranged for older voices too, though.
Personally, I feel that a men's quartet singing closely voiced arrangements has the best chance of "shooting for the overtones". It's just a matter of physiology and physics (get a quartet of baris and basses from LABBS to sing some stuff in the men's key)... on the other hand, well constructed arrangements aught to allow the singers to produce overtones, whether or not they are men or women, or a mix of both. Screaming tags may be impressive when sung in isolation, I don't think they may be so impressive if inappropriately 'tagged' onto a song with a completely different feeling. I also don't think they are appropriate if they are just giving the chance to 'show off'... that would be unmusical.

And relevant to your last point are my thoughts back in the summer about certain forms of showing off...

Archive by date

Syndicate content