In the last blog I looked at the first few bars of the Klinglers’ Op.127; I’m going to stick with the same recording in this one, and talk about the way they handle the whole opening section. I’m especially interested here in the relationship between the structures we see in the score, and the structures that we hear.
It’s a bit of an open question at the moment as to how good the brain is at perceiving large-scale structure in music. A listener obviously experiences music as a series of moment-to-moment progressions, but can also make connections between passages that have been heard already, and generate expectations of what’s going to happen next. Reading a score, though, is a rather different experience, because it removes the element of time, enabling us to see a piece much more as an object, or architectural structure. Of course, we know this isn’t how the music is actually experienced, but putting passages side by side, comparing them, and trying to explain their relationships can be a really useful tool for preparing a convincing performance.
Although it might seem as though this process is relatively concrete – it does use a written score, after all – what you tend to end up with is actually a pretty large selection of different, but equally interesting, conclusions. In other words, using analysis like this to understand a piece almost always raises more questions than it answers. Which is actually a very good thing, because what it really ought to be doing is stimulating curiosity, rather than just fitting square expressive pegs into square analytical holes.
One of the many reasons that early recordings are so interesting is that they frequently demonstrate approaches to structure that simply wouldn’t even occur to most performers now. It’s a complicated picture to untangle, though, because we can’t really talk about structure in isolation from expressive conventions. As I’ve said already, strictly speaking we don’t actually listen structurally. Perhaps a more useful way of thinking about this is therefore to say that performers’ understanding of structure has a big impact on the expressive gestures they make from moment-to-moment. Over time, both of these things – the structural readings, and the expressive conventions – change, and so do the relationships between them.
To illustrate what I mean, I’d like to go back to the Klingler Quartet and their Beethoven recording. Here’s the link again: http://sounds.bl.uk/Classical-music/Beethoven/026M-1CL0018510XX-0000V0. This time I’m interested in a slightly longer passage, still starting at the beginning but going up until the end of the first main section (until about 53 seconds).
(I’ve put the score here for reference; it’s just the first page: http://ks.imslp.info/files/imglnks/usimg/0/05/IMSLP97131-PMLP05045-op127_score_booklet.pdf)
On the page we have two separate tempo markings, and two very different musical textures. For the first six bars all four parts play at the same time, in the same rhythm; at the Allegro they start to interweave melodically, working with and against each other. Going by the tempo markings and character alone, we can be pretty clear in saying that bars 1-6 are an introduction, and bar 7 is where the main theme starts, and with it the movement proper.
Harmonically, though, things are a bit more interesting. The main theme at b.7 begins not on the tonic (E-flat), but on a chord of A-flat, held over from the last bar of the Maestoso. And although there is a tiny cadence into E-flat from b.13-14 (easiest to see in the cello part), most of the theme avoids the tonic, giving it a beautiful suspended quality. In fact, the main cadence isn’t until much later, in b.21-22, (one bar before the forte marking). On the basis of the harmony, then, you could argue that the movement only really gets going after this main cadence, and that the entire first twenty bars, rather than the first six, are a sort of introduction. In practice, most modern recordings make quite a big difference in tempo between the Maestoso and the Allegro, because that’s what’s indicated. When the Allegro is so much faster and freer than what came before it, this music absolutely does feel like the movement’s main theme; the lack of tonic harmony doesn’t seem to stop it sounding like a real theme. The Klinglers, however, embrace an approach to the structure that is much more driven by harmony than the tempo marking, and the way they make this audible is interesting and surprising.
I’ve already talked about how the Klingler Quartet play the first 6 bars, with a fast tempo and sense of gesture that shapes an entire phrase. This is even more interesting in the context of the Allegro that follows it, because their concept of the relationship between these two sections is pretty much the exact opposite of most modern performances. As I’ve said, the Maestoso is generally played in quite a slow tempo, with quite a few beats in a bar; sometimes it sounds in two, sometimes in four. When the Allegro begins, this tension dissipates, replaced by a lilting one-in-a-bar feel, easy-going yet tender. (This might not be the case for every single recording, but I think it’s a reasonably accurate generalisation.) It seems to be just what Beethoven is asking for.
One of the things that performers have a lot of control over – and, as a result, tend to spend a lot of time arguing about – is how many beats there are in a bar. A 3/4 Allegro like this, if it’s going to lilt, would normally be thought of in 1, so as to avoid lumpy phrasing, and to help get a sense of the whole 8-bar phrase. By contrast, one can achieve a heavy character in the Maestoso by making each beat in each bar quite important. So, by giving lots of beats to the opening, and far fewer to the Allegro, performers can make quite a big distinction between the feel of the two sections, and bring out their contrasting markings. It’s not just about tempo; the number of beats that are felt has a big impact on how you actually experience the meter, especially without a score to follow.
The Klinglers don’t do this at all – they go completely the other way. The Maestoso is hardly beaty at all – you can easily hear the passage in 1 if you want to. The Allegro, though, is most definitely in 3. It is extremely expressive, yet always unsure of itself, as the tempo varies from beat to beat. Using hardly any bow, they make a sound that is a long way from being free and easy-going, drawing out every single expressive interval. The small, restrained vibrato and occasional portamento only enhance this sense of fragility. It also means – I think – that it is possible to perceive the Allegro as being slower than the Maestoso. Certainly, the tempi seem a lot closer to each other than they do in more recent recordings.
By bringing the tempi so much closer together, they knit these two apparently disjointed sections together to form a much larger ‘introduction’. The thing that’s really surprising, and that we would almost never do now, is the pretty noticeable accelerando into the main E-flat cadence at b.21-22. This has the undeniable effect of saying: here we go, this is where it really gets going! It is clearly no accident; it’s a very well-thought through, harmonically justified structural conclusion, that the introduction comprises the entire section until the main perfect cadence. Expressing this depends on their use of a device – unmarked accelerando – that modern players are usually told to avoid because it implies a lack of control. (Ironically, it is also sometimes discouraged because it interferes with the musical structure!). In this case, though, the musicians actually create their own very sophisticated structure that is quite different from the obvious one that can be derived from the tempo markings.
Again, we can think of this as a sort of hierarchical model, in which certain features of scores are seen as governing, and others as subsidiary; this seems to have changed over time. For the Klinglers, it seems that harmonic structure was pretty high up the pecking order, and so tempo variation could be used for expressing their harmonic reading of the structure. (Interestingly, as in their take on the opening 6 bars, you can also hear exactly what the marked meter is - 3/4, not one in a bar…)
Structure, then, is not necessarily something that’s embedded within ‘the music itself’, although it clearly does have a lot to do with the score. It’s worth remembering that we can’t actually hear barlines or tempo markings when we listen to music; instead, we only perceive (indirectly) the structures that the players create by manipulating the moment-to-moment progression of the sound. Often these are the same ones that a quick reading of the score would imply – that’s the thing about listening to music played in a familiar performance style. Unfamiliar styles, on the other hand, sound strange to begin with partly because they conceptualise and express musical structure in different and surprising ways. Groups like the Klinglers seemed to have a rather more hands-on approach than many modern players do; perhaps nowadays we prefer to ‘let the music speak for itself’. But I hope that this example shows that for all the bad press that ‘rubato’ (and especially rushing!) gets, when used in sophisticated ways it can be extraordinarily effective, allowing us to see familiar scores in a new light. Evidence like this also gives us license not only to ‘look for’ structure, but actually to create it for ourselves.