brahms’ handel or handel’s brahms?

© Tilman Skowroneck 2011

In anticipation of Murray Perahia’s new CD with Brahms’ Handel Variations, which I ordered minutes ago, a few thoughts about the tangles of performance practice in this work are in order.

These magnificent variations are based on an aria from Handel’s first keyboard suite in B-flat Major. Although Brahms – as we read in the article I linked to above – drew his inspiration mainly from the bass, the theme, with all its added and omitted twiddles, is Handel’s own. Now, how does the pianist have to approach these eight bars of Early Music?

(I lifted this music example off Wikipedia for the sake of simplicity. I have no idea which edition this is, but it concurs with the Urtext I have at home).

A quick performance-practical walk-through will identify a few minor problems with this version. The end of bar four, for example, lacks a trill symbol on the fourth beat of the right hand (that is, the second A; the fourth thirty-second note before the end of the bar. The complete figure consists of a preparatory run f-g-a, a short upper-note trill on a, and a turn g-a at the end that either leads back to the beginning — in which case one normally would go straight to the b-flat on the downbeat of bar one, instead of playing two b-flats in succession — or on to the second half). And in bar eight, the notation of the trill with a preceding turn seems a little idiosyncratic, although its effect is pretty much standard of the time.

Going from there, one might have to discuss whether Handel, in this particular piece, had his Italianate cap on, and thus wanted common trills (tr.) to be started with the main note, or whether he was thinking Pan-European with a French lilt and wanted them to start with the upper note. Both would be faithful to identifiable original styles. As to the zigzag trill signs in bar four (the missing trill would likely have to be notated with this sign as well) and bar eight, the repeated notes in any case only work idiomatically correct and technically plausible if the (short) trill is started with the upper note in the second mentioned manner, and integrated in the flow of the flourish. So, a distinguishing feature in this local context between tr. and the trill sign might be its starting note, or it might just be its length; tr.: long, zigzag: short.

There are good arguments for starting all trills in this piece with the upper note, in fact. Some historically informed harpsichordists, as I found out, even do it sometimes this way, sometimes the other, and add some little notes here and there to boot (link here; the piece begins at 4:05. It is a bit of a free-form approach, but by all means a possibility). The question is, would Brahms have known about all this? Would he even have cared? This problem is a little trickier to address than the previous historically informed sketch suggests.

The Romantic trill confusion had already been generated around the time Brahms wrote his variations. Some observers maintained that, in spite of Hummel’s trill-start reform (in his Ausführliche theoretisch-practische Anweisung zum Piano-Forte-Spiel, 1828), older music should be played with upper-note trills. Others began ping-ponging a bunch of rather messy rules among each-other, and that’s how we still have it today.

One such rule is the approach-from-below-means-main-note-trills-rule, which to my knowledge is not to be found in any 18th-century treatise about embellishments, but rules solidly in modern texts about Classical performance. In our case, this rule would indicate that the tr. signs in bar one and three are to be played in a different manner than all the other ones, were it not for that other messy rule, the approach-from-the-same-note-rule, which has not been decided upon; some feel that the trill’s first note should repeat the preceding one, others like an upper-note beginning that avoids this repeat. Then, there is even an approach-from-above-rule, which in the penultimate of my links to youtube examples, further down in this post, induces the pianist to begin these trills with the main note and all the others with the upper auxiliary. Such a picky pianist might thus end up playing the tr. signs differently in different instances (I remember that Rudolf Serkin does this too, in his second, well-known studio recording, but with other results).

Most of the performers, however, fall apart into those who play only main note trills, and those who play only upper note trills (it turns out that Perahia plays the tr. signs as main-note trills, and the zigzag with the upper note. We are left to think that Brahms knew or cared little about how we today think that Handel perhaps played Handel, but we are made to see that there’s an end to all this when sheer practicability is concerned. I like this).

As seen in my first harpsichord example, all this is good and well, and both can be called authentic, if we argue well. Two problems remain, however. One is the question of the missing trill in bar four. Every pianist I am aware of just plays the notated mid-run-double-A-step-dance of the awkward lead-back of the right hand, but no one seems to be aware of the (for the Baroque practitioner obviously) missing trill sign. For those who know this work well, the double A has essentially become Handel according to Brahms; it is not merely Handel any more. Any historically plausible mollification (and a well-played trill would make that twiddle-run sound much smoother and mellower) would somehow take that impression away. I admit nevertheless that, if I would at all be able to play the beefier parts toward the end of the variations, I’d be tempted to play a retro-beginning à la Handel, with a short upper-note trill on the second A. But it could well be that Brahms thought the notated stutter to be attractive, and knew of no trill. In this case, a Handelization of the lead-over would be out of place. The problem is, we do not know the correct answer.

Another problem is, conversely, that some pianists seem to find the Baroque-y beginning simply too cute for words and start adding glitter, and messing with the meter. What the listener gets offered here is a postcard picture of Handel, with frills, decadent sweets in a dish beside his harpsichord, and a colorful parrot climbing in the curtain. Not only do we thus sometimes hear some upper-note trills that are, for no apparent reason, played noticeably before the beat (others aren’t. Some start with the main note. Keep listening), which strikes me as the musical equivalent of coffee spill on a white table cloth. Some pianists seem to find trills so much fun, that the listener ends up with almost nothing else, as we can hear here.

The problem of trill performance practice in this work cannot be solved for once and all. We don’t even know whether Brahms cared one way or another. There are too many layers present, and one would always have to choose, and argue for one’s choice. So, uncharacteristically for my stubborn mind, I have decided that I am fine with almost any type of trill in this theme, unless they’re played messily or spill out over the edges. It is better not to add even more layers to this Handel-Brahms sandwich than we already have; especially not sticky and sweet ones.

Tags: , , , , , , , ,

3 Responses to “brahms’ handel or handel’s brahms?”

  1. Siebe Henstra Says:

    Ha Tilman,

    Je schrijft goed en leuk!
    Het kan aan mij liggen maar ik zie niet wat voor triller aan het eind van maat 4 ontbreekt. Bedoel je dat het zigzag symbool niet goed is? Volgens mij zijn (bijna) alle niet heel specifiek Frans georiënteerde componisten totaal ongeïnteresseerd wat voor teken voor lange/korte triller of mordent gebruikt word, zo ongeveer als de + in de bovenstemmen. Overigens noteert zelfs Couperin niet apart de triller met lange bovensecunde wat D’Anglebert en Purcell (en Bach soms?) nog wel deden, waarom?


  2. skowroneck Says:

    Hoi Siebe,
    Er ontbreekt een zigzag op de vierde tweëndertigste a voor het eind van maat vier. Het is een triller met aanloop. f-g-a; dan triller op a; dan naslag g-a (ik heb dit nog wat duidelijker gemaakt in de text) Voor de overgang terug naar de herhaling zou waarschijlijk de b niet eens gespeeld moeten worden – als je doorgaat naar de tweede helft natuurlijk wél.

    Wat je zegt over het door elkaar gebruiken van de symbolen is natuurlijk geen vraag voor Handel met zijn rommelige manier van noteren. Maar toen Brahms dit model gebruikte was er een hele discussie gaande over wat wat betekende. En dat is hier nou juist het probleem: denken wij dat hij dacht dat het verschil van belang was, of niet?

  3. Siebe Henstra Says:

    oh ja, inderdaad zo logisch dat ik het niet eens zie! Je hebt gelijk.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: