Tessitura for (Barbershop) Tenors

Quite a lot of the things I get interested in have applicability across choral genres and beyond, but today’s subject is pretty specific in focus and will likely hold little interest beyond people arranging for barbershop ensembles, and rather fewer than a quarter of the people singing it.

I had an email recently from a singer who after a good many years’ experience as a first soprano has joined a female barbershop chorus and been placed as a tenor. She had considered herself hitherto pretty confident with high notes, but is really struggling with one particular arrangement they’re currently singing, and wondered, having read a post of mine about arranging barbershop for female voices, whether the problem was the male arranger (and male chorus director in turn) not understanding the female voice.

She sent me the arrangement to ask what I thought, and I’m not going to identify it to you, because it’s not fair to discuss another person’s work publicly (though she might feel able to share this post with the people involved, of whom it is clear she is fond). But I can talk about some of its features, in the hope that it helps other people in both their arranging choices when writing for upper-voice ensembles and their arrangement choices when shopping for music to sing.

The first and easiest thing to consider is range, which is the total compass of notes used by the part in the chart. In this arrangement the tenor has E4 – F#5. My instinct was that this was perfectly normal, maybe on the more challengingly high side, but as it’s a substantial piece clearly intended for contest purposes, being challenging is reasonable. A quick survey of the first 25 charts for female voices on my list confirms this hunch: a couple go higher (G5/A5 respectively) but those were written for ambitious ensembles where I knew the specific singers would be up for it. The most typical top note is E5, with a few F5’s and E flat5’s thrown in.

So it’s not the range that’s an issue particularly – people who sing 1st sop would see F#5’s as all in a day’s work. The question then becomes one of tessitura. I had some quite specific thoughts about how to think about this some years back, but didn’t have a way to automate them. That post elicited my first interaction with Alexander Koller (who many barbershoppers will now know as a fine coach, arranger and Musicality Category judge), with some interesting ideas of how one might do this if you could export an XML file from your notation program.

At the time I couldn’t, and as it happens, I am still using software that irritatingly doesn’t allow this, but if yours does, that post will be much more useful to you. In the meantime, I decided to do a manual check of a single page of the arrangement my correspondent was asking about. I picked page 5 of 6, and it took me about 10 minutes to check out the one part. So, something one can realistically do manually as a spot check, but too unwieldy to do for all of every part of very many songs unless you have something else that you are really desperate to procrastinate. The exercise produced the following results:

• Mean average pitch, i.e. halfway between highest and lowest is D/D#5
• Mode average pitch, i.e. the note that is used the most often by duration is E5
• Median average pitch, i.e the one in the middle if you line all the pitches up in order, also weighted by sung duration, is D5

My first surprise was how high even the mean average for this page was: in these 16 bars, the lowest note was B4, which is actually the mean of the whole song. Looking back at the earlier pages, the tenor part does get as low as G#4 quite often, but the moment with the E4 is a brief unison with lead and baritone, and so quite unrepresentative of the part as a whole.

The other two averages are more representative of tessitura per se as they quantify what proportion of your singing time is in the upper range. And these both looked, from a sucking-my-teeth-and-squinting-at-it perspective, higher than I would want to write for female tenors. But I thought I had better actually check that instinct against my own work, so did the same exercise on the tenor part for the penultimate page of my most recent contest chart:

• Mean average pitch is D5
• Mode average pitch is B
• Median average pitch is C#5

The mean average is thus pretty similar, and in fact my chart goes a semitone higher, reaching a G5 for a dotted crotchet as we head into the tag. Likewise the overall range for the song goes down to an E4, but at this stage of ramping up to the big finish the tenor line doesn’t go below a B4, just like the example sent to me. The median average is likewise only a semitone lower, so they’re looking quite similar at this stage. But the mode note – i.e. the one the part spends most time on – is in mine the lowest note of the page, as opposed to only a tone below the highest. That could make a significant difference experientially I should think.

(For what it’s worth, I consider my own tenor line analysed here pretty challenging. It was written for an ambitious group I have arranged for before and I’m confident they can do it; I wouldn’t recommend it for less skilled singers. It is not a tenor line I personally would be comfortable depping in on, although - at the risk of over-sharing - since the hormonal changes of middle-age, my comfort in upper ranges is no longer as useful an indicator of singability as it used to be, and I am having to rely more on knowledge and experience.)

The other thing I found useful in thinking about this line is to consider how it would lie on the voice of a male tenor if transposed a 4th or 5th down. And you know, it still looked pretty high, so I went another tone down, which gave a range for the page of D4 – A4, a mode of G4 and median F#, which looked pretty realistic for a song voiced for TTBB winding up towards its climax. The thing to be aware of here is that, as male tenors routinely sing in falsetto quality, they often find it more comfortable to hang about in a region clearly above the passaggio between the registers. If you spend too long around or just above middle C they can find it tiring.

So, this does look remarkably like a part conceived to be comfortable for male tenors transposed wholesale upwards, but the story comes with a twist in the tail. Having considered that the tenor line needed to be transposed down a 6th to be realistic for male voices, I looked at the bass and lead lines and was surprised at how high they lay too. You’d need a pretty confident male lead section to manage the line only a 5th down, and if you had some nice deep basses they’d be largely going to waste in that key.

In fact, whilst it would be better suited for female voices with some judicious switching of baritone and tenor notes to tighten up the voicings and make the tessitura more manageable, you could actually just sing the whole thing a tone down from what was sent to me and make everyone’s lives easier and the sound richer. The issue, that is, is not simply that the shape of the tenor line produces an exhaustingly high tessitura, but that the whole song has been shifted up by too much.

This has turned out to be a bit of a monster of a post, but it needed walking through the detail to make its points. I hope this gives some useful ideas both to people assessing repertoire for suitability for their groups and to my fellow arrangers about things we can do to bring out the best in singers. The big take-away for me has been to generate the hypothesis that:

When arranging for female tenors, you would expect the mode average to lie in the middle to lower part of the overall range, whereas for male tenors it is more likely to lie above the mean average.