Thoughts on Tuplets vs. Subdivision?

Question

This was a point of interest brought up in the comments to my answer on Help with analyzing sus and add chords in this progression.

In that question, the sheet music has a measure of 4/4 time written with a quarter note followed by a quadruplet that took up the last three beats (a 4:3 figure - four in the space of 3).

My answer (here it is for reference) took the position that this was a terrible way to represent the rhythm, that it should have been written as dotted 8th notes (or, equivalently, 16ths tied to 8ths so as to be beamed correctly), completely eliminating the need for any tuplet/polyrhythm notation.

My main argument is that this tuplet obscures the beat completely. Visually, the tuplet does not show where beats 3 and 4 fall in the measure, making it unnecessarily difficult to sight-read. Representing this rhythm on the 16th-note sublevel would, when beamed appropriately, make the locations of beats 2, 3, and 4 obvious. In the same regard, showing the midpoint of the measure is always a good thing, and the quadruplet fails to do so.

My secondary point is that the use of a tuplet subtly implies a temporary suspension of the ordinary meter, but in this case the tuplet is unnecessary because the rhythm should primarily be felt as syncopation in 4/4, not even notes in a faster tempo starting on beat 2 of a slower measure. This is a subtle "feel" difference that gets elaborated upon in the video I put in my answer.

I did face some opposition in the comments, though, the relevant bits of which I will quote here:

"I've actually been told by multiple sources to prefer tuplets over dotted-note notation (i.e. they recommended the opposite of what the video in your answer does)."

"[...] counting out the dotted notes [is] more painful than intuiting the tuplets, even at slow speeds. This even seem[s] to apply regardless of obscuring the middle of the measure."

Faced with differing viewpoints, I think some discussion is in order: What are some arguments in support of or against the tuplet notation shown above? Or, if there's another alternative, what is the best way to represent that rhythm?

---

If you need the full piece, this is what the original question linked to.

Answer 1

I agree with you. If the intention of the composer is the exact rhythm posted, don't use a tuplet. Write out the dotted eighths, so one can see how they line up against the beats. For additional comprehension, one could use both: I've definitely seen scores with the literal (in this case, dotted eighth) rhythm written on the staff, with tuplets stemmed and bracketed above the score to help see the rhythmic continuity.

I'd say the problem with using a tuplet is that many performers don't tend to play them exactly or at least see them as a potential invitation to a more rubato performance. Doesn't happen as much with simple ones like 3:2 or 4:3, but outside of percussionists, very few performers are going to bother getting an exact 5:4 or 5:3 tuplet precisely correct. Certainly not anything involving a 7 or greater number.

In this case, given that the tuplet is also spanning an odd part of a measure, it's not necessarily easy to just "feel" the subdivision intuitively. There is an argument in some other cases for tuplet notations where the resulting literal rhythm would be exceedingly complex. But durations like dotted eighths are not uncommon in 4/4 bars. And any performer who couldn't parse the rhythm you propose (with ties) and realize it's equal dotted eighths isn't likely to be the sort of performer who could "intuit" how 4 notes should feel divided up over 3 beats in an odd division of a bar.

Writing it without the tuplet gives a fighting chance to get the exact rhythm performed accurately. Writing it with a tuplet almost guarantees you're going to get some sort of rubato approximation from many performers here. Unless the composer wants the latter, I'd avoid the tuplet.

(And by the way, regarding the Neely video linked in your previous answer, I don't think most classical musicians would view 2:3 duplets in 6/8 as necessarily a reason for freedom in rhythm. The example he gives is "Clair de Lune," which is obviously a bit of a rubato piece to begin with. If you had 2:3 duplets show up in the middle of a 6/8 march, it would be played precisely the same as the dotted eighths. Overall: context matters, but in the appropriate context or where the tuplet is hard to parse precisely, it's more likely to be interpreted imprecisely than a literal non-tuplet rhythm.)

Answer 2

Looking at the whole piece, I'd say the the first 5 measures are a rather rhapsodic introduction to an otherwise fairly metrical main section and I think the slightly odd rhythmic notation helps to emphasise this. There isn't a 3rd or 4th beat in the measure, and I don't think that renotating in dotted eighth-notes would help much. Play the first two quarter-notes in time, and the next three just fast enough to fit. And if it doesn't, it's called rubato. There's already a rit in measure 4, and I assume a missing a tempo in 6.

Musical notation is never that exact, even when it looks like it. Stravinsky would disagree.

Answer 3

Any meaningful argument either way needs more context.

These two bars don't exist in isolation from the rest of the piece, which we haven't seen. For all we know, the "best" notation might be something completely different, like