41. audacity

I recently came across Richard Egarr’s recordings of the Handel organ concertos—pieces I have known for decades, but that I learned first from the recordings of E. Power Biggs (with Adrian Boult) and Simon Preston (with Trevor Pinnock).  Much as I respect and enjoy Biggs and Preston, the Egarr recordings have a impetuous audacity that strives to emulate the (basically unwritten) example of the composer.

To take only the most striking example, at the end of the second movement of Op. 7 no. 4, where Handel’s instruction reads merely “Organo ad libitum,” Egarr uses that rarest of intervals, the doubly-augmented unison/octave, to effect the modulation from D to F major, for his interpolated slow improvisation.

SOURCE:  my transcription of this moment (the transition from Track 15 to 16) on disc I of Egarr’s recordings of Handel's Op. 7 (and more) with the Academy of Ancient Music; Harmonia Mundi HMU 807447.48 (2009).  The recording is streamed on Spotify here.

I spell it thus (D-sharp against D-flat) because that is how these notes are functioning:  chromatic side-steps up from D to E and down from D to C.  The resulting passing dissonance should not be spelled as a diminished third (which would indicate converging rather than expanding motion); a major second would be meaningless.  Audacious is the only word.  The interval is not part of Handel’s musical style, but it has the (I think) appropriate consequence of directing the spotlight onto the soloist.  What will he do next?

Handel himself seems to have relished the spotlight:  he used his organ concertos mainly in performances of his oratorios, where during the breaks between sections he could have all the attention to himself.  But he wasn’t above stealing the show from his highly-paid vocal soloists.  His first London opera, Rinaldo (1711) includes a remarkable moment at the end of Act II when Handel apparently wanted to divert the attention at least temporarily from the stage to the pit.  His autograph of the aria “Vo’ far guerra” has disappeared, but the early copies indicate that in the opening ritornello there was to be an extended extempore passage for harpsichord (Cembalo), for the composer himself to display his gifts:

SOURCE:  Opening of “Vo’ far guerra” (Rinaldo, Act II) in a copyist’s manuscript held by the British Library (f. 91r of R.M. 19.d.5); scanned from Graham Pont, “Handel versus Domenico Scarlatti:  music of an historic encounter” in Göttiger Händel-Beiträge IV (1991), p. 234.

Susan McClary famously characterized the texture of the first movement of Bach’s fifth “Brandenburg” concerto as one in which “the harpsichord, which first serves as continuo support then begins to compete with the other soloists for attention, and finally overthrows the other forces in a kind of hijacking of the piece” [p. 28].  How much more audacious was it for Handel to upstage the singer of an aria, left lingering on stage while the composer showed off?  The aria includes collaborative passages in which the soloist and the harpsichord run in parallel as a sort of duet, and where Signora Pilotti (for whom this aria was written) holds a note for several bars while a harpsichord obbligato is conspicuously busy underneath, but there is another totally free section—left to the discretion of the soloist—in the closing ritornello, while again the singer is left at loose ends on stage.  Moreover, as it is a da capo aria, all this happens twice (with, presumably, different extemporizations). 

We do not know what Handel played in these ad libitum episodes, but I expect they could not have been lasted very long, as otherwise there would surely be some press comment.  The Spectator lampooned the first production of Rinaldo, but no mention is made of excessive keyboard virtuosity in those reviews.  Perhaps later in the run the solos became more extended and showy; certainly there’s no reason to assume that he always played the same thing.

When Chrysander published Rinaldo in the old complete works (HG vol. 58 in 1874), he had the portions of the autograph preserved in the Royal Music Library, and copies like that above, and so his edition has the same “Cembalo” instruction with no indication of what to play.  After doing a little more legwork and tracking down more sources, he published Rinaldo again (HG vol. 58 [bis] in 1894), including both the 1711 and 1731 versions.  “Vo’ far guerra” was cut from the 1731 version, but for the 1711 version this time Chrysander added an appendix with a complete realization of the harpsichord solo:

SOURCE:  detail of revised edition of HG vol. 58 (1894), p. 117; from IMSLP #18974

If we look up the aria in the new HHA volume presenting the 1711 version of Rinaldo, an almost identical realization is given not in an appendix but in the main text—without even a footnote to indicate that its source is not easily authenticated.  One has to look elsewhere in these scholarly editions to find that “the Harpsichord piece perform’d by Mr Hendel” comes from a keyboard arrangement, Song’s in the Opera of Rinaldo; this was originally published by John Walsh in the weeks after the February premiere (and a scan of that first edition is available as IMSLP #71438), but this elaborate keyboard part for “Vo’ far guerra” materialized only in a later printing (with a new title page Arie dell’opera di Rinaldo, apparently June 1711).

What originally prompted me to look at all of this for this post was finding a seminar paper I had written in graduate school that was comparing these keyboard passages with those found in Handel’s organ concertos, essentially arguing that all of this could easily be cobbled together from the figuration of Handel’s other bravura works.  (I’ve scanned some of my examples for that paper here.)  Now the glaringly obvious problem with my thesis is that all of my Handel examples post-date these Rinaldo performances by at least two decades; I picked the wrong music for comparison.  Handel had written a concertante part for organ in his Il trionfo del Tempo e del Disinganno (Rome, c. 1707), but it’s not all that much like the published Rinaldo solos; the closest comparison would be his Sonata for Harpsichord with Double Keys (HWV 579), which Terence Best dates to c. 1707-08 [p. 125].  Best argues that BWV 579 “has no connection” with Walsh’s Rinaldo realizations.  In that the text is different, Best is correct; but the similar figuration at least shows that Handelian origins of the latter are plausible.

SOURCE:  cropped scan of the beginning of HWV 579 as given in HHA Ser. IV Bd. 6, p. 80

There is no reason to assume that Handel ever notated the Rinaldo cadenzas; particularly as he apparently had no professional relationship at this time with the publisher, John Walsh, it is much more likely that another hand supplied these keyboard passages.  The scribe seems to have been William Babell (c. 1690–1723), who would have heard Handel’s original performances at first hand, as he was a violinist in the King’s Theatre where Rinaldo was produced.  Babell was much more widely known in his short life as a keyboardist—indeed, this reputation was strong enough that Johann Mattheson would cite him in 1739 as possibly the greatest organist of the age.  Here is the relevant bit of Mattheson’s Der Vollkommene Capellmeister (as translated in Deutsch):

SOURCE:  cropped scan of Deutsch, Handel:  A Documentary Biography, p. 485, with a portion of Mattheson.

Even though Mattheson never heard Babell, this is remarkable praise.  It is known that he studied with Johann Christoph Pepusch in London; that he studied under Handel is not certain, but he was clearly in Handel’s circle.  Moreover, in 1717 Walsh published his Suits of the most Celebrated Lessons Collected and Fitted to the Harpsicord or Spinnet by Mr. Wm. Babell with Variety of Passages by the Author.  This was a tremendously successful publication, and remained in print (from one pirated edition or another) throughout the eighteenth century.  Note the last item in the table of contents:

SOURCE:  cropped from (incomplete) scan available at archive.org.  A complete monochrome scan is available at IMSLP #279417.

The “Vo’ far guerra” in the Suits of Celebrated Lessons is considerably more elaborate than that published in Arie dell’opera di Rinaldo a few years prior.  Graham Pont has probably devoted more attention than anyone else to the Handel/Babell connection, with several intriguing articles published over two decades.  From the first of these, he has contended that the substance of the harpsichord elaborations were Handel’s rather than Babell’s (and in subsequent articles he demonstrates that there were a number of copies of this made by people close to Handel—although he also shows that the text kept changing).

We thus have several different versions of the keyboard cadenzas (for lack of a better word) for "Vo’ far guerra," from the rather tame but still dramatically intrusive version that Walsh published in about 1715 (and which has become the main text of the HHA, tacitly presenting it as the work of Handel himself) to the wildly fantastic version Babell’s Suits that was surely never intended for operatic performance, but rather uses “Vo’ far guerra” as the medium for a solo keyboard showpiece.  Chrysander published it—or one version of it—in HG vol. 48, and it must be some of the most visually-stimulating pages in the whole of that monumental edition:

SOURCE cropped from IMSLP #18931 scan of HG vol. 48, p. 242 ; but there's much more where this came from.

Gotta love that beaming!

If the original aria wasn’t audacious enough, this blows it out of the water completely, with a variety of special effects.  (Peter Holman, in a fascinating article that posits that Babell rather than Handel should get the credit for the first English keyboard concerto, characterizes Babell's music as a “mixture of boldness and limited compositional technique”; from my limited exposure to it, I have to agree.)  There are remarkable moments, to be sure.  Consider this compelling crescendo–decrescendo effect, achieved by a thickening and subsiding of the texture:

SOURCE:  ibid., p. 239

Perhaps this conveys some element that originated with Handel.  (Perhaps.  I doubt it.  There is nothing else I know from his pen that is remotely like this.)  If so, I suspect that Babell’s audaciously over-extended cadenza strings together ideas that Handel might have used in different performances of Rinaldo, not ever intending them to go together, and connected by who-knows-what.  Another speculation occurs to me:  maybe Handel did not conduct all the performances, and Babell (who must have been the finest keyboard player in the orchestra) took over for the harpsichord solos, later reworking his ideas from those extemporizations into the work that was eventually published as a “lesson.”   (For a fine recording of Babell’s lesson, you cannot do better than Erin Helyard’s.)

For one more audacity, René Jacobs’s 2003 recording of Rinaldo (Harmonia Mundi HMC 901796.98) turns to the Babell lesson for inspiration for the harpsichord solos of “Vo’ far guerra” (starting about two minutes in).  I think this choice absolutely spoils the dramatic momentum that ends the act—or rather, it redirects all attention to the pit.  (Forty-five percent of this track is taken up just by the cadenza after the singer’s last cadence.) I think this decision is a serious mistake, but I concede that at least

1.  it makes the da capo different than the first time through, 
2.  maybe this shifting of attention does less damage on a recording than a live performance, and 
3.  it is audaciously well played.

32. Père et fils (or, “WTF-horn?!?”)

One day in a music history class in the autumn of 2016 (shortly after launching this blog), my students and I were huddled around a study score of Jacques Ibert’s Divertissement.  It is a piece I like very much—très amusant.  Ibert derived the suite from some incidental music he wrote for a 1929 production of Labiche’s hilarious Un chapeau de paille d’Italie (a stage production directed by René Clair, following hard upon Clair’s silent film adaptation of 1928).

There weren’t very many students in that class, so we all had a serviceable view of the score, despite its fairly small dimensions.  I had assigned them the second movement, “Cortège”—although it must be the most riotous cortège in the repertoire.  Compare, however, what we saw on the page

SOURCE:  scan of Ibert Divertissement (Durand, 1931, reprint n.d.), p. 17.

with what we heard from the CD recording I had chosen for them—Yan Pascal Tortelier’s 1992 Chandos recording with the Ulster Orchestra [Chandos 9023]:

[I am very grateful to Chandos Records Ltd. for permission to use this excerpt for this post.]

I had heard this recording many times before.  I am generally partial to Mr. Tortelier’s recordings.  (Among his very many fine accomplishments, I would recommend particularly his recording of Guilmant’s Symphony no. 1 for organ and orchestra (with organist Ian Tracey and the BBC Philharmonic) [Chandos 9271].  It is a work that seldom gets played, but it gets a splendid airing on that disc, and—as it exists in two rather different versions (organ solo and organ with orchestra)—it will probably emerge sooner or later as the topic of a post on this blog.))  [ADDENDUM:  it did.]

As I say, I had heard this recording many times, but apparently not with the score at hand.  That morning in class I exclaimed, “Where is that horn part coming from?”  As my students put it (in the vernacular), “WTF?”

Soon after class, I pulled a few other recordings off my shelves, but none of them had this extra horn line.  (This figure happens twice:  Reh. 6, and then again at Reh. 10 up a half-step; the horn part seemed the same in both places on the Tortelier recording.)  There was nothing in the Chandos liner notes to indicate that this recording featured a new edition of the score.  Anyway, I mentally filed it away to explore later.

Returning to it about two months ago, I was just as mystified as before.  I investigated getting the performing materials on perusal to see if anything useful was there; but, as the US distributor is Boosey & Hawkes, that effort proved prohibitively expensive.  Moreover, a more recent recording from Chandos—an excellent one with Neemi Järvi conducting the Orchestre de la Suisse Romande [CHAN 5168]—clearly manifests the text as printed in the Durand score.

 [Again my thanks to Chandos Records Ltd. for permission to use these recorded excerpts for this post.]

I have not laid my hands on an item catalogued in Worldcat as a “neue Auflage” apparently issued by Durand in 2012, but—as will be seen below—it wouldn’t have solved the mystery even if it is a new edition rather than just a reprint.  (It may well just be a re-setting using music notation software; no editor is listed.  I’ve complained about that sort of thing before on this blog, with the Glazunov Saxophone Concerto as my example.)

I decided it couldn’t hurt to try to contact Mr. Tortelier directly, but what exactly was I asking him?  I thought it best to transcribe what I thought I heard the horn playing, and I enlisted some other keen ears to give it a go.  Here is what I could convince myself I heard:

SOURCE:  my attempt to transcribe the rogue horn part at Reh. 6; I thank my colleagues Paul Rawlins and Michael Bratt (both University of Mary Washington) for their willingness to tackle the same problem.

I then sent what I had to Mr. Tortelier’s agent and wondered if I would get any reply.  A few weeks later it came, revealing that the source of this interpolation was his father, the esteemed Paul Tortelier (1914-1990), cellist, composer, and conductor.

Indeed the passage at Reh 6 for 8 bars as well as the one at Reh 10  for another 8 bars have an added and indeed optional horn part counterpoint which was added by my father Paul Tortelier and is none other than the main theme of rehearsal 2 ( trumpet pianissimo ) and tutti forte at 4 but in this instance it is funnily clashing with the same tune played  in augmented values on trumpet and flute.
To answer your question and sum it up, my father is to blame for that and I assume it should be heard on his own recording with the English Chamber Orchestra.   [email of 28 April 2018]

Once he mentioned the melody at Reh. 2, I saw why the second bar of my transcription had been naggingly familiar.  Here is Ibert's tune there (third staff, trumpet in C):

SOURCE:  detail of scan of Ibert Divertissement (Durand, 1931, reprint n.d.), p. 12.

Moreover, Mr. Tortelier was kind enough to provide a scan of his score, with the interpolated part neatly added in his father’s hand:

SOURCE:  Durand score p. 17 with Paul Tortelier's interpolated horn part in manuscript (by courtesy of Yan Pascal Tortelier, email 6 May 2018); for the parallel passage at Reh. 10, see this page. 

SOURCE:  Tortelier père et fils recording
Tchaikovsky in London in 1973, from EMI's 1981
Grand Echiquier reissue.

I haven’t actually been able to locate any recording of the Ibert conducted by Paul Tortelier, and it may be that he performed it thus without ever actually committing it to disc.  There is something charmingly audacious about this addition—in a way showing a loyalty to the impish style of the composer even while departing from fidelity to the text as such.  I am glad that Tortelier fils shared his father's inspiration with a wider audience.  It is a remarkable moment, and when I now listen to other recordings the original text seems... well, not bland, but at least a little lacking.

All this prompted me to wonder, though, what other contrapuntal “Easter eggs” (to borrow a gaming term) are lurking on recordings of standard literature—whether intentional interpolations by the conductor or as pranks by the players.  As a continuo player, I have in my realizations occasionally introduced a snippet from another work as a sort of countermelody.  (I have found that the phrase “Way down upon the Swanee river” works particularly well.)  I imagine others have amused themselves in similar ways.  I would be happy to add an addendum to this post if readers can point me to other examples.