21. Moving targets (Episode #3)

Two Settling Scores projects intersect in this post—the ongoing series of “moving targets” and the seventh installment of my

Even the exact boundaries around Bach’s oeuvre are a perpetually moving target, and the best illustration of these is the very notion of the “complete” organ works.  A review of the contents of the standard complete editions of Bach's organ works is a good introduction to the disputed borders of this repertoire.  Those editions, all widely in use today, are (in roughly chronological order)

Peters = the first attempt at a complete edition of the organ music, edited principally by Friedrich Conrad Griepenkerl (1782-1849); seven volumes were issued by C. F. Peters 1844-1847, with an eighth following in 1852, when the series was regarded as complete.  In 1881 the ninth volume appeared, and that gradually morphing ninth volume is my principal concern in this post.  Despite its age, the Peters edition is not to be discounted by any means, as some important manuscript sources available to Griepenkerl have subsequently disappeared. This edition had a splendidly ostentatious title page:

SOURCE:  cropped scan of title page of a copy of Peters Vol. XVIII (1852, but this copy must be printed between 1881 and 1904, as the first version of Volume IX is listed in an advertisement on the back cover)—cropped because of the huge tracts of nineteenth-century margins that would take up too much real estate on my blog.  Subsequent reprints significantly reduced the margin size.  The changing dimensions of different printings of a single edition would be an interesting topic, if one had but time.

BG = although Griepenkerl beat them to it, of course the first attempt at publishing the complete works eventually got around to the organ works.  These appeared in five volumes during the years 1853-1893; these became the text underlying a practical edition issued by Breitkopf & Härtel, but the original BG is still used today because of (for example) Dover’s reprints of much of it.  It also became the main source text for a number of other practical editions—particularly those issued by G. Schirmer (the Schweitzer edition), Novello & Co. (early volumes were based on Peters, and some volumes have subsequently been re-edited), and Bornemann (the Dupre edition).  

20th B&H = In the late 1930s, Bärenreiter had started an edition of the organ works, edited by Hermann Keller; this project as aborted because of the Second World War after only two volumes.  After the war, two new editions capitalized on the recent explosion of Bach textual scholarship.  Heinz Lohmann edited this ten-volume set for Breitkopf & Härtel, with the first volume appearing in 1968, but with the set completed scarcely a decade later.

NBA = The other edition which began to emerge after the war was that of the Neue Bach Ausgabe, the new complete works.  Series IV (organ works) had eight planned volumes, but a ninth was necessary because of the 1985 discovery of the so-called “Neumeister Chorales,” now attributed to Bach's early years; much later came the appearance of two additional volumes featuring works from the Bach circle that could plausibly (if doubtfully) be attributed to him.  All told, it took fifty years for Series IV to be completed.  This expansion of the series indicates a tendency to cast the net ever wider—an understandable temptation when the NBA project as a whole is an obligatory expense at many libraries around the world.  The sales numbers may be comparatively small, but they are pretty much guaranteed.  (Bärenreiter issues offprints of the musical text of all eleven volumes, and it is in this form that the NBA shows up on the music racks of organs.)  Now a new revision (NBArev) promises at least two volumes of organ chorales, which I assume will essentially replace the flawed Ser. IV. Bde. 2-3, the earliest of the original volumes to appear.

Truly, of the making of many Bach editions there is no end.  Two very interesting editions are ongoing as I write:

Leupold = This is a very serious scholarly edition that does a very good job of catering to the very serious student.  All the volumes that have appeared so far have been edited by George B. Stauffer, certainly a prominent name in the last generation of Bach scholarship, and Stauffer does his best to make the editorial issues clear to the user.  It's not clear to me how many volumes this edition will eventually comprise, as some are to be issued in two very distinct versions (“Standard Urtext” and “Practical Urtext”—a concept which seems a little dubious to me).

21st B&H = And now Breitkopf & Härtel is at it again, with an entirely new edition planned to comprise ten volumes.  With so many accumulating, it seems odd to call this one a “welcome addition,” but in my estimation it is just that—and the edition I would recommend to organists wanting a chance to look anew at works they have played for years (although in my experience using any unfamiliar edition will force that new glimpse).  This is certainly an edition for the new century—taking advantage of digital advances (with online resources and enclosed CD-ROMs which allow users to print out the variants they want while avoiding the bulk and waste of paper for those who don’t require them).   To quote the Preface, “In addition to presenting the musical text with comments, this disk allows synoptic depictions and a cogent search process for specific measures, thus providing a better and faster overview than would be possible with a printed version.”

And surely that’s enough to be getting on with.  But here I want to focus just on the oldest of these, and just its last volume, which appeared in three substantially different manifestations—first in 1881 (three decades after the rest of the set), then again in 1904, and finally again in 1940.  Each issue was the work of a different editor—in 1881 by Griepenkerl’s successor Ferdinand Roitszch; Max Seiffert’s 1904 revision coinciding with his important discovery of new sources; and Hermann Keller's in 1940 at the moment that his Bärenreiter set was abandoned.  Even from the start, Vol. IX was something of a catch-all volume, with a mixture of chorale-based and free works.

Between them, the three different versions of Peters Volume IX contain some 38 individual works, but only twelve works appear in all three.  Several of the works included by Roitszch in 1881 were later ruled to be misattributions.  Seiffert excluded three of these (BWV 692; BWV Anh. 57 and 171), and three that escaped the 1904 purge were tossed out by Keller (BWV 561, 580, and 587).  Further, seven of Seiffert’s twelve new additions were deleted by Keller (BWV 742, 743, 747, 752, 754, 757, and 763), although five of those have subsequently found a place in the NBA.  (Only one of Keller's seven additions was not retained in the NBA (BWV 1027/4a); the music is not printed, but it is given its own section in the critical report to Ser. IV, Bd. 11.)

Excluding the thorny question of which Clavier pieces were not intended for organ anyway, if one takes the Bach organ repertoire at its widest breadth (as does the late, lamented Peter Williams, for example, in his excellent survey, The Organ Music of J. S. Bach—and really his second edition doesn't completely supersede his first) I find that there is actually no single “complete” edition that comprises the repertoire in toto.  Even if one has ready access to the BG and NBA, there are still missing works (not likely to appear in either Leupold or 21st B&H).  I note, for example, two works that have appeared only in Seiffert’s 1904 version of Peters Vol. IX (BWV 752, and 763) and you will search in vain for them elsewhere (unless you are content with homemade editions posted on the IMSLP).  As more and more performers perform the whole corpus as Bach organ “marathons” [Google it ], it would be nice to know exactly how the placement of the finish line is determined.

As Williams has astutely remarked, “[i]t is a curious irony that the uniform appearance presented by any edition of Bach’s organ works distorts them in that it does not give a true impression of the disparate nature and origins of the pieces themselves....  In giving pieces of edited music to the public, editors misrepresent them, despite earnest endeavors to do the opposite.” [p. 274].  The impressive bindings of such series conceal the bewildering array of textual situations for the repertoire contained therein.  Even that repertoire won’t stand still for a generation.

20. The chord that should get lost

So far I have generally avoided posting about my own work, but as it is only with my own work that I have really been able to be “behind the curtain,” I thought I would give an example where the reading I (as editor) wanted wasn’t what eventually made it to print, and where even the textual note about it didn’t ultimately satisfy me.  I don’t think this is a case of “telling tales out of school,” but a reminder (to myself, at least) that all sorts of hidden factors may stand between the editor’s intended text and that which is published.  I know that there must be many cases like these, but how can we know unless people share their experience?

I have the highest regard for my general editor on the William Walton Edition—the English conductor David Lloyd-Jones.  His path-breaking edition of Boris Godunov in the 1970s has had long-lasting effects on the way the opera was performed, restoring the quirkiness of Mussorgsky’s text, and he has produced critical editions of a wide range of works—principally nineteenth- and twentieth-century Russian and British works, but with a distinguished foray into Berlioz, too—all moonlighting alongside a distinguished conducting career.  By offering me a Walton volume to edit, he gave me my first big break professionally; by offering me a second volume, he shored up my confidence to continue.  I owe him an incalculable debt; and yet here is an instance where we disagreed—in this case just a single chord, but a chord I would so much like to lose.

The second of the volumes I did for the edition was an unusual one, consisting of concert suites derived from Walton’s film scores, including those for the Laurence Olivier Shakespeare films Henry V (1944), Hamlet (1948), and Richard III (1955).  These suites were made in the 1960s by Muir Mathieson (1911-1975), who had conducted the recording sessions for each of these films.

SOURCE:  1946 78s; scan of  US release front cover

There seems to be a critical consensus that the most successful of these films (and of these Walton scores) is Henry V.  Right after the release of the film, a four-movement concert suite was prepared (credited at the time to conductor Malcolm Sargent, although I have not found the slightest evidence that Sargent had any hand in it), but Walton’s publisher never offered it for sale, so it had only a marginal impact.  (The two movements for strings only were put on sale and were more widely circulated.)  A more successful commercial re-use was a recording of musical excerpts paired with Olivier reciting newly-recorded speeches on four 78 rpm discs [shown at right].  This recording had a good bit more music than the suite.  Significantly, it included the dramatic “Charge and Battle” music, a musical and cinematic climax of the film.  As the title indicates, these are actually two different musical cues—separated in the film by the launch of a volley of arrows.  (Walton marked the end of his autograph of the charge with the caption “Bombs gone!”)  The charge sequence is tremendously exciting to watch—particularly the virtuoso tracking shot lasting some fifty seconds as the French horses trot, then canter, then gallop across the field.  (A contemporary reviewer noted that the audience at the press-screening gave a spontaneous ovation at this moment, even though “they were applauding t’other side.”)  The whole charge sequence is analysed shot-by-shot and bar-by-bar in the 1957 Roger Manvell and John Huntley text The Technique of Film Music.

SOURCE:  Manvell & Hartley, The Technique of Film Music, p. 91 (the very end of the “Charge”); I have reformatted the page here to better fit a wide rather than tall aspect-ratio.  The penultimate bar is incorrectly transcribed:  Walton writes this as two bars of 3/4, with each 8th-note here really a quarter.

When Mathieson set about producing a new suite, he used the “Sargent” suite as his starting point, but deleted the choral parts (and consequently the music requiring a chorus) and reduced the scoring to double-wind, hoping to make it more attractive to smaller orchestras.  He also inserted the “Charge and Battle” music (plus another section, appended after the battle—but thereby hangs a tale for another time) as the centerpiece of a five-movement suite—and very effective it is, too.

Mathieson has skillfully spliced the cues together, but in doing so he added a chord.  The “Charge” cue ends abruptly on a downbeat—just an eighth-note chord.  (It was followed on both the film and the 1946 RCA recording by the launch of the arrows; to hear those, click either of those hyperlinks.)  The “Battle” cue continues in 3/4 time although with an eighth-rest on the downbeat.  Mathieson elides the two, so that the downbeat chord that ends “Charge” takes the place of the eighth-rest on the downbeat of “Battle.”  The problem comes in the second bar, which Walton indicates only by a ditto mark:

SOURCE:  detail of Walton's autograph of the beginning of the “Battle” cue (147c) of Henry V, taken from a screenshot of the page at the website of the Beinecke Rare Book and Manuscript Library, Yale University [http://brbl-zoom.library.yale.edu/viewer/1237439].  The measure numbers in red ink across the top were added by Mathieson as he prepared his suite.

Mathieson applies this ditto sign to everything in his newly-elided bar, so the downbeat chord appears a second time.

I sought to remove this extraneous chord from the new edition, as it forms no part of the film or any of the early sources, even if it was unambiguously a part of Mathieson's arrangement.  David Lloyd-Jones’s response—reasonable as ever—was that as Walton had conducted a recording of the Mathieson arrangement at the time of its publication, the appearance of the chord on that recording could be taken to be Walton's acceptance of the variant reading.  (Hear him conduct it here; listen closely and you might hear my teeth grinding in the background.)  Given that this recording was made twenty years after the music was composed, and that Walton was hardly the most detail-conscious of composers, I was not persuaded that the presence of the chord indicated that he had even noticed it, let alone endorsed it.

And so the offending chord appears in the William Walton Edition, over the objections of the volume editor:

 

SOURCE:  marked-up digital scan of William Walton Edition vol. 22, p. 48 (detail).

The corresponding textual note documents that this was Mathieson’s addition, but does nothing to suggest—as I wanted to—that one should at least consider eliminating it.  I believe our volume is an improvement over the first edition of Mathieson’s arrangement (especially because ours restores Walton’s original triple-wind scoring), but here is one place where I think we didn’t go far enough in restoring the composer's text.

19. (im)posing

which is the sixth installment of the Settling Scores

In this post I consider an edition of a central item in the Bach keyboard repertoire, BWV 971, the Italian Concerto from Clavierübung II.  The edition at hand was the work of Rosalyn Tureck (1914-2003), and it was published by G. Schirmer in 1983 under an imprint revealing that it was to be the first of many:

SOURCE:  cropped scan of the title page of Tureck’s 1983 edition

The series nameplate combines three terms that do not necessarily mix well:  critical edition, performance edition, and urtext.  (Later issues in the short-lived series add facsimile to the namplate, and this is accurate as each one includes a facsimile reproduction of her principal source(s), although sometimes these are reduced in size or too faint to be very useful.)

A critical edition is particular type of scholarly edition:  it is the product of the studious application of some text-critical method involving multiple sources.  In musical editions these methods are usually deployed in pursuit either of the earliest version penned by a composer, the last version as revised by a composer, or (increasingly, so it seems to me) a version that the composer never got down on paper but for which there is some authoritative evidence; but a critical edition need not seek a version known to the composer, but might seek a subsequent version (like Mendelssohn’s or Vaughan Williams’s re-scorings of Bach’s St. Matthew Passion—in either case the usual Bach sources would be generally irrelevant to the text-critical process).  Whatever goal a critical edition is aiming for, I am all in favor of it when it is done well.

Many critical editions also claim the label urtext, which was coined to refer to the original reading of a text, but which in musical editions has come to describe little more than editions purged of interpretive editorial accretions—slurs and other articulation marks, dynamics, fingerings, pedalings, bowings, etc.  In that sense an urtext may not prove to be “critical” in the sense described above.  I would argue that an edition based on a single source—that is, in which there has been no collation of variant readings—cannot be a critical edition in its strict sense, although it may well be an urtext in at least the marketing sense.  (Although a number of webpages credit the publisher Günter Henle as the first to apply the term to music, the OED has a 1932 citation from the TLS referring to “the nearest thing possible in Chopin’s case to an Urtext,” and Breitkopf & Härtel used it for a series of publications in the 1890s.)

A performance edition aims to make visible to the user the various unwritten or obscure conventions that are otherwise invisible on the page—or, indeed, performance practices that are quite far removed from the composer.  Editorial additions to the text in such editions vary widely, and sometimes recognizing them for what they are can be difficult, as you have to know what a composer “looks like” to know when something has been added; at other times the interpretive suggestions are bleedingly obvious.  And—as I indicated in my very first post—I don’t disparage such editions in the least, as they are vital resources in understanding performance conventions of their time.  The goals of urtext and performing editions are  essentially at odds.  While it is a standard practice to adapt a text to modern notation conventions (for example, using treble and bass clefs only for keyboard music), many editions I see marketed as urtexts make more extensive concessions to aid in performers’ interpretations.  Tureck’s does this, but at least she explains in her preface exactly the sorts of interpretive marks she has added.   Nonetheless, her approach to these editions seems to transfer them out of the urtext realm into something else altogether.

The two other issues of the Tureck’s series were the lute suites BWV 996 and 997 adapted for the guitar.  These bear a curious note in which the editor insists that “[t]he Suites in this series, edited for classical guitar, are not arrangements.  This edition preserves the original form of Bach’s compositions” [p. ii].  Indeed?  An urtext for the wrong instrument?  If the word arrangement was too slippery for her to assert authority, why not transcription?  Clearly adjustments have been made in order to make this playable at all:  “Where notes are considered unplayable on the guitar the editor does not omit them.  For the sake of musical completeness they are included within parentheses in the musical text.  8ve signifies an octave above the original register, an editorial solution” [BWV 996, p. v].  An “editorial solution,” that is, to a problem created by the editor.  These are transcriptions—very good ones, perhaps, made by an authority on Bach performance—but they unnecessarily claim a specious scholarly objectivity.  She is posing to be more than she is.

So too her edition of the Italian Concerto (which Tureck significantly describes as edited “for Harpsichord or Piano”), which presents not merely a critically-established text, but overlays it with her interpretation.  The fidelity is not thus to the text of the work but rather to a learned artist’s understanding of it.  She glosses the Italian Concerto so that we can see what she perceives when she reads the music.  Not that there’s anything wrong with that:  I am reminded of Malcolm Bilson’s complaint that everyone uses urtext editions but few know how to read them. Tureck’s profuse prefatory remarks give the impression that hers is—for the first time ever—the text Bach meant.

“No autograph copy of the Italian Concerto exists.  Although several manuscripts in other hands are extant, the most reliable source is Bach’s own corrected copy of the first printing, published in 1735, in which revisions are set down in his hand. [Bach-Digital description here.]  The second printing appeared in 1736.  The inestimable value of Bach’s text is self-evident.  It is a rare instance in the keyboard works of direct contact with the original textual and performance intentions of Johann Sebastian.  The editor employs this musical text [i.e., the 1736 edition or the hand-corrected 1735 edition?] for this edition which, besides being an urtext edition, is also edited for performance on the piano or the harpsichord according to all the original indications in Bach’s corrected copy.  The stem directions, which in the editor’s opinion are of prime importance, have also been preserved as closely as possible.... [p. iii]  It has long been the custom to present a ‘clean’ score with urtext references, leaving the performer to find the way to performance solutions.  This procedure has served two functions:  (1) it has rescued editions from erroneous music texts and anachronistic performance directions[,] and (2) it has reflected scholarly research and orientation.  The bare urtext editions give the performing musician and teacher contact with the scholar’s approach and with increasingly reliable scores which provide a textual foundation upon which an authentic performance art may be developed....”  [p. iv]

The eyes glaze over at some point, and she relies increasingly on an authoritarian passive voice:

“In addition to current editing procedures, performance practices must now be introduced if musicians are to employ an urtext which will contribute to an authentic performance style.  Heretofore, the performer has been left uninstructed, an impossible practice for music composed some 250 years ago. Innumerable specific, historical performance practices are identifiable, and substantial data concerning them are available....  This edition integrates the textual sources with Bach’s own performance indications and historical style, based on Baroque performance practices for harpsichord. These practices, when combined with an uncompromising purity of Baroque style, considerations of the musical structure, and a fitting piano technique, have valid applications on the piano.” [p. iv]

Paul Badura-Skoda blasts this edition in his Interpreting Bach at the Keyboard, particularly in a section he calls “The Urtext Problem: An Imaginary Interview” [p. 188-200], where he imagines all the questions he wants to be asked and so that he can respond pithily.  In particular he takes Tureck to task for her radical proposal that in the second movement the accompanying alto line may be ornamented.  See her note before the beginning of the movement:

SOURCE:  digital scan of Tureck ed., p. 8

Whatever else one might say of this edition, it doesn't have the look of an Urtext.  As promised, it is laden with interpretive information.  These alto-line embellishments are a step too far for Badura-Skoda:

B.-S.  Look, if a professional musicologist comes up with a sensational claim such as, for example, that in the second movement of the Italian Concerto the alto voice should be freely embellished throughout, then he or she should adduce some evidence to support it.  I know of no Italian or German treatise written during Bach's lifetime that suggests the embellishment of accompanying (in contrast to imitative) inner voices.  Thus one might expect from Dr Tureck some evidence or explanation for such a claim... 

DR C.  Doesn’t she point to the fact that the thirds in the lower system of the Andante have two tails [stems] throughout, thus proving that they are two distinct voices? 

B.-S.  This is simply a naïve remark and doesn’t prove anything.  Everyone who is acquainted with the keyboard music of the eighteenth century knows that it was common notational practice to add tails to all thirds, sixths, and so on.  This even occurs in Haydn and Mozart.  If what she says were true, one could also, in the Italian Concerto, set about enriching the accompaniment of bars 30f., 46f., and 129f. of the first movement with ornaments. 

DR C.  But Tureck points to bar 17 of the second movement, where, in the first edition, there really is a Pralltriller sign in the middle voice... 

B.-S.  ... which is almost certainly an engraver’s error...  [p. 195f.; the ellipses are his]

He belabors this point at some length, adding that her suggested execution of the trill is wrong “because the Pralltriller formula she adduces is not found in a single treatise before 1757, seven years after Bach died.”

This notwithstanding, the only thing that irks me about Tureck’s edition is that she overreaches, imposing her way as the one true path to Bach.  Badura-Skoda does the same.  Maybe this is an occupational hazard for performers, but particularly for one who communed with the composer as “the high priestess of Bach.”  And, like any editor, Tureck is delighted when she can restore some textual variant in order to give us the truth:

“As a result of this comparative analysis, the editor brings to light an error in the first movement which appears in well-known 19th and 20th century editions including the Neue Bach Ausgabe.  At measures 13-14 the figure in the soprano had been altered to match measures 175-176 in the closing da capo section.  This figure is restored in this edition to its original version.”  [p. iii]

The figure to which she refers is this:

SOURCE: composite; my own transcriptions (clefs updated) from the first edition (1735) from IMSLP #417409

Tureck makes a big deal of this, but other editors before her have noted the oddity and opted to regularize.  Badura-Skoda’s response: “In the other Bach concertos initial and final tuttis of a movement are identical, which means that this discrepancy goes back to an engraver’s error that Bach overlooked” [p. 198].  Just off the top of my head I can think of one closing ritornello markedly different from the opening—the first movement of Bach's second “Brandenburg” concerto. (When another example occurred to me, I find that it is slippery—so slippery that, rather than digress here, I will plan to get back to it a future post; if you want to whet your appetite, you can see it here.)

In any case, “Original version” may be too much for Tureck to claim, as it appears that this figured had changed a bit before the 1735 publication.  A manuscript copied by Johann Christoph Oley (1735-1789) held by the Boston Public Library (details at Bach-Digital here, but a scan is available at the IMSLP) presented an earlier reading of the text which Oley subsequently updated to match the published version (altering even the title page to conform—see the account in the NBA Kritischer Bericht).  Wherever they might be placed metrically, these twiddles seem to have been second thoughts.  In these and other instances Oley has originally written 16th-notes, and then crammed the extra note in to make a pair of 32nds.  (Note that here, in addition to the cramped space, often the stem of the added note does not cross the closest beam, although in the places where Oley originally wrote 32nds (as throughout the second movement), the stems cross all the way to the main beam.)

SOURCE:  cropped screenshots from Oley MS, at IMSLP #302163

Even here, Oley doesn’t quite match the printed text: in bb. 73-74, the 32nds are early (by which I mean, on the first half of each half-beat); in bb. 175-76, the first pair is early, while the other two pairs accord with the printed reading.

Tureck’s labors devoted to Bach interpretation are admirable—especially her “progressive anthology” series, An Introduction to the Performance of Bach (3 vols, Oxford,  1960).  Most interesting there is a sort of etude in which she has rewritten the C-major 2-part invention (BWV 772) so that the hands are essentially reversed “for the development of flexible thinking in two parts”:

SOURCE:  cropped page scan  from Tureck, An Introduction to the Performance of Bach, vol. 2, p. 14

This series is naturally a product of its time.  (See especially her discussion of the sources of the Aria Variata, BWV 989, in vol. 3, p. 7)  The same is true of the Schirmer series from the 1980s.  This edition of the Italian Concerto remains in print, and it is part of a long tradition of instructional editions.  Its presumption to be more than it is probably does no harm—but a phrase from an earlier Badura-Skoda publication come to mind (co-written with his wife, Eva):  “[i]t would be a good thing if the use of the word ‘Urtext’ were protected by law...” [p. 129], rather like Champagne and organic in some jurisdictions.  It sells well, but is the label accurate? 

18. Moving targets (Episode #2)

Although I meant to get back to this a long time ago, this is only the second in a series considering different editions of the same work issued by the same publisher but without any notice of textual discrepancies between them.  Sometimes these changes are hardly more than cosmetic, but sometimes they are real nuisances, and sometimes inexplicable meddling.

SOURCE:  scan of 2013 printing

The example in this post will be the two Alphonse Leduc editions of Glazunov’s Saxophone Concerto (1934), and I will be considering only the versions issued “avec accompagnement de piano” rather than the full score.  The first edition appeared in 1936, bearing the plate number “A.L. 19,256.”  A scan of a 2007 reprint of this edition (judging from the imprint date on the last page, anyway) is available on the IMSLP.  I cannot date the second edition, but the terminus ante quem is 2013, the imprint date of the copy I have to hand, and would roughly coincide with Leduc’s acqusition by the multinational Music Sales Group.  (Worldcat doesn’t help much to date this second edition:  all of the Leduc printings listed there are said to be 19 pages, while this is 22pp.)  The copyright date on the second edition remains 1936, and indeed both the cover (at right) and the plate number (AL 19 256) also remain the same.  I strongly oppose their re-use of the same plate number when the editions are manifestly different productions.  In any case, my use of plate number is not really accurate:  even if the first edition was engraved on copper plates, the second edition is a re-setting via computer notation software.  It is the tell-tale short cuts typical of computer-setting that I want to highlight here, as I think the earlier version makes for a clearer read.  Here is an example from the very first page:

SOURCES:  bb. 6-7; cropped digital scans of first ed. Leduc piano score, p. 1 (left) and second ed. Leduc piano score, p. 1 (right) 

Crowded as the first edition may be in b. 7, the five voices of the counterpoint are quite clear, and the pianist understands which notes belong to which line.  In the second edition, there is no delineation of the five voices, and the beaming in the new top “voice” even obscures that on beat 2 there is an entry—the highest, and thus arguably a climax—of the main motive of the introduction.  The viola and cello lines are subsumed into a sort of new “tenor” voice.  Most troubling is the movement of the f marking ahead by a beat, which suggests even a bringing out of the inner voice at beat 2.  The orchestral score presents the opposite situation—the f appears first in the first violins (beat 2) and in the rest of the ensemble a beat or more later.

What is frustrating about this is that it is merely the product of laziness:  it would take a little more time at the computer to arrive at the configuration of the first edition, and the setter apparently didn’t think it was worth it.  The version on the right is marginally easier to play, but I think that is the only thing I can say in its favor.

There are a number examples where to me the changes in the notational configurations in the two Leduc editions do not amount to improvements, but I will consider just a few here.

SOURCES:  bb. 47-48; cropped digital scans of first ed. Leduc piano score, p. 3 (left) and second ed. Leduc piano score, p. 4 (right) 

That one is hard to explain.  To my eye, the first edition is clearer, although I grant that the reading on the right makes the crossing of the upper parts more obvious.  Still, the lower stave looks curious, with the stem-down downbeat and no rest above it.

SOURCES:  bb. 131-32; cropped digital scans of first ed. Leduc piano score, p. 7 (left) and second ed. Leduc piano score, p. 9 (right) 

This one is my favorite, a true comedy of error.  I think I can reconstruct what happened in this instance.  I think the computer-setter set the lower staff first, then—while setting the upper staff and converting it to the bass clef—decided to move the mano destra line also into the upper staff.  The comedy is that the setter never deleted this line in the lower staff, so it is duplicated erroneously, uselessly—indeed meaninglessly.  Maybe it’s just me, but I chuckle to myself about such absurdities.

SOURCES:  bb. 227-28; cropped digital scans of first ed. Leduc piano score, p. 12 (top) and second ed. Leduc piano score, p. 14 (bottom) 

Here I would argue that both versions are unplayable as notated (at least at the 100-beats-a-minute tempo).  The second edition moves the bass-line up an octave—something a player might do anyway—even when Glazunov scores this for cellos and basses an octave and two octaves below this register (but not at all in the register notated here).  As an accompanist often playing these sorts of orchestral reductions, I generally prefer to know what the general texture is (even when unplayable by me) and adapt as I have to than to have someone else attempt to much simplification for me.  The first edition just gives more.  It is instructive to compare this passage as it appears in the recent Bärenreiter edition:

SOURCE: bb. 227-28 of piano reduction by Martin Schelhaas (Bärenreiter 8732a), p. 13; the bar numbers are different because this edition counts the cadenza as one bar.  (It includes three different versions of the cadenza, and an interesting tale it is.)

Throughout the Bärenreiter edition the piano reduction has been entirely rethought with an eye toward simplifying it down to something reasonable—even though the editors acknowledge that the Leduc first edition piano reduction seems to have been the work of Glazunov himself rather than “A[ndre] Petiot,” to whom the Leduc editions have given the credit.  Thus the Bärenreiter “Urtext”—and a very good edition it is in many respects—has jettisoned the authentic piano part.  Although the “Urtext” label appears on the cover, the title page is more accurate:  “With an Urtext Solo part... Piano Reduction based on the Urtext.”  (There is a critical report included in the corresponding urtext full score, but there is of course not one for the new piano reduction, which also lacks a description of the sources.  Martin Schelhaas seems to have used the second Leduc edition as his starting point, but the result is a wholly new and worthy reduction.)  This piano reduction also includes an additional soloist part (i.e., non-urtext), edited for performance by Carina Raschèr, daughter of Sigurd Raschèr, for whom the work was written.

This publication is not the strongest in the Bärenreiter catalogue (as even without hunting for them I found small errors in both the musical text of the full score and the critical report), and it is even more curious for its mixture of urtext and practical approaches.  (I will return to that particular dilemma in my next post.)  Nonetheless, it is not the focus of this post, and it is in any case better than the second Leduc edition.  I find it extremely irksome that Leduc would reissue the work, bearing the same plate number as the first edition but with an inferior presentation of the text.  The Leduc standard has been lowered, although perhaps not to half-mast.

17. Just kidding?

The fifth installment of the Settling Scores

Johann Nikolaus Forkel’s 1802 biography of Bach has many over-the-top claims which bash the reader over the head with Bach’s supreme genius.  Today I consider just one of these, drawn from information related by C. P. E. Bach.  I’m prompted to write about it now because of some evidence that I have stumbled across in recent months supporting my doubts about it.

In the chapter concerning “Bach the clavier player,” Forkel records that

“He even went so far, when he was in a cheerful humor and in the full consciousness of his powers, as to add extempore to three single parts a fourth part, and thus to make a quartet out of a trio.”  [trans. in The New Bach Reader, p. 435; cf. C.P.E.'s letter in ibid, p. 397]

And he repeats it in the chapter concerning “Bach’s character”:

“If he was in a cheerful mood and knew that the composer of the piece, if he happened to be present, would not take it amiss, he used, as we have said above, to make extempore, either out of the figured bass a new trio, or of three single parts a quartet.  These, however, are really the only cases in which he proved to others how strong he was.”  [Ibid., p. 460]

Maybe there is some basis in fact to this story—even if it was only a single occasion—but it seems to me literally incredible otherwise.  Forkel even acknowledged elsewhere that the sort of polyphony that allows the adding or reducing of parts has to be very specially constructed:

“In his compositions in four parts, you may sometimes even leave out the upper and lower part and still hear in the two middle parts an intelligible and pleasing music.  [Only the two middle parts, Herr Forkel?  An example would be nice....]  But to produce such harmony, in which the single parts must be in the highest degree flexible and yielding towards each other if they are all to have a free and fluent melody, Bach made use of peculiar means, which had not been taught in the treatises of musical instruction in those times, but with which his great genius inspired him.  These means consisted in the great liberty which he gave to the progress of the parts.  He thereby transgressed in appearance, but not in reality, all the long-standing rules which, in his time, were held sacred.”  [Ibid., p. 443]

As an organist I am occasionally in a situation where a descant is added for the final verse of a hymn.  Sometimes the descant line is printed in the hymnal on an extra stave above the four-part harmony.  What I notice invariably—because it is indeed inevitable (that word that dogged my previous post)—is that in order to give the descant line a musically-satisfying melody, it will at times borrow note-progressions from the alto or tenor (or even the soprano), creating intrusive parallel unisons or octaves.  Really, the right way—if I may be so bold—to add a descant is to compose it as a counterpoint to the melody, and then write a harmonic background for those two lines together.  There are many examples of this done well, but too often I’m playing the other type.  And it beggars belief that a hypothetical trio by Forkel’s “composer of the piece, if he happened to be present” would accommodate an added fourth line with its own integrity while not making substantial alterations to the original parts.  I have naively accepted this as literal truth for too long.  No longer:  two examples I recently noticed suggest to me that four voices was too many for Bach to shuffle around in his head.  Heresy?  Maybe so.  But look at these:

1)  strict canon:  “Christe eleison” from the Missa in A, BWV 234

I love this movement.  For years I have used it on the very first day of Theory I, when students generally have no theoretical background, may only read one clef (if that), and may never have seen a full score.  I throw it at them and ask them to observe:  What do you see?  And there’s much to be seen.  It is a strict canon beginning with the bass soloist, and with ensuing entries in the tenor, alto, soprano, and finally two flutes in unison—each of these entries a perfect fourth higher than the last, and with the sustained harmonies in the strings never really relaxing into anything that feels like a resolution.  Even the apparently simple question “What key is this in?” defies a simple answer.  Also interesting is that the canonic line echoes its opening arpeggio (and more) twice even as the voices accumulate, so that the arpeggio occurs not just five but nine times, as if in stretto.  (This is partially illustrated below, but if you’re curious you’ll save time just looking up the movement yourself.  Good stuff.)

One day, sitting in the Subaru service department with the NBA Kritischer Bericht at hand—as you do—I was killing time looking at the variant readings for this movement.  The report documented a series of systematic corrections in the autograph, commenting merely (and I paraphrase) “the corrections in bb. 80, 82, and 85 are related:  Bach altered his conception after the fact, as the fourth canonic phrase initially began with a leap of a fourth.”  [p.27]  Each note circled below (the autograph score on the left, the NBA text on the right) was originally a fourth lower in the autograph.

Composite of BWV 234/i; SOURCES: (left) cropped scan of autograph score f. 3v (bb. 75-85) from ULB Darmstadt scan;
(right) scan of Barenreiter TP 266 (off-print of NBA Ser. II Bd. 2), p. 11 (bb. 80-85); for both, I have added the red circles.

Here is the bottom system of the same page of the NBA (bb. 83-85), which presents the complete “Stimmtausch block” of the canon, marked-up to show the canonic phrases.  [I appropriate the term from Robert L. Marshall’s landmark study of Bach’s compositional process (v. 1, p. 134f.]  “1” is how the line begins (originally in the basses, now in the flutes), continuing to “2” and so on to accumulate five canonic elements (although each voice is a fourth higher than the previous one).

SOURCE: cropped scan of Barenreiter TP 266 (off-print of NBA Ser. II Bd. 2), p. 11 (bb. 83-85)

The NBA does not go on to explain why Bach would have made the changes.  Bar 85 makes the reason clear:  the A in the alto (the pick-up to canonic phrase “4”) was originally an E, but the leap up from E to A would cause parallel fifths with the tenor leap from A to D.  This became an issue only when he wrote the fifth canonic phrase, but apparently he didn’t notice it in b. 82, so had to go back to fix it there; the retrospective change in b. 80 was only necessary to preserve the strict canon, and was thus presumably the last to be made.  The lesson here?  Bach could juggle a number of voices in his head as he constructed the canon, but clearly had to get it down on paper to get it right once too many voices had accumulated.  In other words, he had not conceived phrase 5 when he first notated phrase 3.

2)  permutation fugue:  chorus Himmelskönig, sei willkommen, BWV 182/ii

This is a similar example—this time more complicated because the lines had to work as invertible counterpoint rather than just a canon.  Arthur Mendel discussed this example in a 1960 Musical Quarterly article; his point was really just that an autograph that Spitta had taken to be a fair copy was really a composing score, as the types of corrections were not copying errors but rather directly related to the substance of the counterpoint:

Composite of BWV 182/ii bb. 1-5; SOURCES: (top) cropped scan of f. 1v of autograph score of BWV 182 from Bach Digital; (bottom) cropped scan of Mendel’s reconstruction of the first reading of these bars (Musical Quarterly (1960), his Ex. 1, p. 292); for both I have added the accolades on the left.

 Mendel notes a pattern of corrections:

“...we see that while there are no corrections in the first two measures, in measure 3 the last note in the soprano has been changed from an original a1 to d2, and there is a corresponding change in the alto in measure 4 and the tenor in measure 5.  In the bass in measure 6 [not shown above], however, at the beginning of the second brace, there is no corresponding correction; here the corrected reading found in the other three voices was written in to begin with.  If we look a little further, we can see that at the end of measure 4 not only the alto but also the soprano, and at the end of measure 5 not only the tenor but also the alto, have been corrected.  But again in measure 6, the tenor bears no correction corresponding to those in the alto in measure 5 and in the soprano in measure 4.”  [p. 292]

and then reconstructs Bach’s compositional process:

“Bach starts out as follows, writing in the first brace the whole four measures of the soprano, then the rests plus three measures of the alto, then the rests plus two measures of the tenor, and finally the rests plus the subject in the bass.  [This produces Mendel’s Ex. 1, given in the above composite.]  But already on the fourth beat of the bass's subject-entrance (the first beat of measure 5) there occur consecutive octaves between bass and soprano.  Apparently he next changed the soprano, inverting its motion to read d2-f#2 instead of f#2-d2.”  [pp. 292f.]

He then charts further changes to get from his putative original to the eventual (I hesitate to say final with Bach) reading.  He concludes

“It is surprising to find that in writing such a permutation fugue (he had already written several that we know, and the scheme remained a favorite one with him) he had not worked out the invertibility of his four melodic elements until he set pen to paper to write a score that is neat enough to have been taken by Spitta for a fair copy.” [p. 293]

Granted, Bach was clearly very good at this.  The anecdote transmitted by Forkel suggests a contrapuntal understanding of such profundity (being able to spontaneously convert a trio into a quartet by the addition of an extra line) that writing out such examples into fair copy would not be surprising at all.  For too long I’ve taken the Forkel story at face value, but when I stop to think of it it really can’t be true.  “Just kidding,” as my students sometimes say when corrected; maybe Forkel would say the same.

ADDENDUM  20 August 2017

In the course of my Bach cantata pilgrimage, I have run across an interesting example going the other way—something originally in five real parts (SATB with a descant) in which Bach later deleted the fifth part, and had to make only very minor adjustments in two spots for the four-part version:  compare the closing chorale of BWV 12 with its later use in BWV 69.

ADDENDUM  28 November 2017

Then again, the seven-part scoring of the final chorale of BWV 70 more than amply demonstrates Bach’s ability to write integrated-yet-disposable lines.  Even so, there were some small alterations to the inner parts when this harmonization was included as a four-part chorale as no. 347 of the Breitkopf edition of Bach's chorales (1784-1787).  See NBA Ser. III, Bd. 2, Teil 2, where this chorale appears on p. 200.

16. Forty years ago in a galaxy far, far away....

“That a work of creation struggles and insistently demands to be brought into being is a fact that no genuine artist would think of denying.  Often, the demand may impose itself in defiance of the author’s considered interests and at the most inconvenient moments.  Publisher, bank-balance, and even the conscious intellect may argue that the writer should pursue some fruitful and established undertaking; but they will argue in vain against the passionate vitality of a work that insists on manifestation.  The strength of the insistence will vary from something that looks like direct inspiration to something that resembles a mere whim of a wandering mind; but whenever the creature’s desire for existence is dominant, everything else will have to give way to it; the writer will push all other calls aside and get down to his task in a spirit of mingled delight and exasperation.” [pp. 140f.]

Thus the always fascinating Dorothy L. Sayers in The Mind of the Maker, her examination of the human creative mind via a Trinitarian analogy.  She argues that even before a creator may have a clear conception of it, a “creature” (artwork?) has an existence of its own and insists on being realized in form knowable outside of the creator’s mind.  Elsewhere in the same treatise she writes

“[Let us imagine that] Our perfect writer is in the act of composing a work—let us call it the perfect poem.  At a particular point in this creative act he selects the ‘right’ word for a particular place in the poem.  There is only the one word that is ‘dead right’ in that place for the perfect expression of the Idea.  The very act of choosing that one ‘right’ word automatically and necessarily makes every other word in the dictionary a ‘wrong’ word....  Now, the mere fact that the choice of the ‘right’ word is a choice implies that the writer is potentially aware of all the wrong words as well as the right one....  Potentially and contingently, his intelligence ‘knows’ all the wrong words.  He is free, if he chooses, to call all or any of those wrong words into active being within his poem....  But the perfect poet does not do so, because his will is subdued to his Idea, and to associate it with the wrong word would be to run counter to the law of his being.  He proceeds with his creation in a perfect unity of will and Idea, and behold! it is very good.”   [pp. 104f.]

And all of this is in an analogical discussion of the origin of evil....  But it is this ex post facto sense of inevitability that I want to highlight.  That was the word used in a particularly purple Leonard Bernstein passage, expounding on why Beethoven was a “great artist”:

“Always probing and rejecting in his dedication to perfection, to the principle of inevitability.  This somehow is the key to the mystery of a great artist:  that for reasons unknown to him or to anyone else, he will give away his energies and his life just to make sure that one note follows another inevitably.  It seems rather an odd way to spend one’s life; but it isn’t so odd when we think that the composer, by doing this, leaves us at the finish with the feeling that something is right in the world, that something checks throughout, something that will follow its own laws consistently, something we can trust, that will never let us down.”  [p. 93]

My interest in variant texts is sometimes an idle curiosity about what publishers have seen fit to present a buying public, but most of the time it is the drive to know more about uncertainties and second thoughts that composers faced as they tried to bring a work to fruition.  The final product may seem as if it were inevitable, but even in works generally accepted as masterpieces of their kind, the progress toward the familiar version—whether or not this is the Fassung letzter Hand—is not inevitable nor even inexorable, and may unfold with various fits and starts.  A few summers ago I was fascinated to read Dominic McHugh’s account of the challenges that Lerner and Loewe faced in trying to morph G. B. Shaw’s Pygmalion into a musical.  My Fair Lady was a hit musical by the time they were done with it, but at many points along the line it was not at all obvious what to do—even how to end it.  With a good portion of the work done, they aborted the project, only to return a year later to bring it to fruition.  And similar tales could be told of a great many pieces—many more, I’m sure, than I will ever know about.

The artist’s perennial “divine dissatisfaction” notwithstanding, I tend to think writers and composers—with rare exceptions like Felix Mendelssohn—generally know when they’ve produced something that is pretty much what they were trying to say, or at least the best version of it they can get on paper at the moment.  Or at least something they can live with.  (Sure, they may change it later.)  So I was startled a few months ago to be reminded of a musical moment which in my estimation is pretty much “perfect,” but of which the composer was unsure even in the recording studio, recording five substantially different takes—substantially in that the musical substance was different each time.  It wasn’t just a matter of the performers playing to his satisfaction, but rather of not being certain which performers should play what and when.

And so we go back exactly forty years ago this month:  March 1977, when 86 musicians of the London Symphony Orchestra sat for eight days of sessions under the baton of composer John Williams to record the music for the soundtrack of STAR WARS.

I was a little too young for STAR WARS (and somehow I feel like I have to capitalize it), being just over two years old when it came out.  But I had two older brothers, and my childhood was inevitably saturated with STAR WARS stuff, including the LP record The Story of STAR WARS, which juxtaposed soundbytes from the film with narration, and a generous share of Williams’s score.  I must have listened to that album many times, as when I saw the film again in the theater in 1997 I was surprised how the lines of dialogue excerpted in that LP jumped out at me as intimately familiar, despite a general unfamiliarity with the complete film.  Coinciding with the 1997 re-release, I bought the re-released soundtrack for the first of the films, figuring it was the sort of thing someone of my generation and background needed to know better:  STAR WARS as cultural obligation.

Now as my children get older (and as the franchise of films keeps expanding), I sensed the same obligation.  So I borrowed the DVD from a friend and one night sat down with the kids to watch—knowing I was going to be pausing every 20 seconds to explain (at the 7-year-old and 5-year old level) what was going on.  There was the familiar 20th-Century Fox fanfare, and then the pregnant silence under the words “A long time ago in a galaxy far, far away....”  And then that memorable cymbal clash + B-flat explosion, the fanfare that opens the main title music blasting at us, over the iconic opening crawl.  (Watch it here, almost as in 1977.)  Suddenly I felt tears welling up, and I couldn’t say why exactly.  Sentimental memories from childhood?  But those opening three bars are as close to perfect music—doing the right thing at the right time—that I know; and I’m usually a snob about such things.

They liked the movie, and so a few days later I took the soundtrack CD off the shelf and put it on in the car while taking them to school.  A while later, driving and without even noticing that the music had gone off, suddenly a voice crackled over the speakers:  “Take sixteen.”  And then for the next two minutes or so, there was the first musical cue of the film, starting with the main title music.  Then “Take seventeen,” and there it was all over again.  The 1997 release (and apparently some subsequent reissues) includes a hidden extra, an “archive” of the five session takes of the main title music.  I had heard all these years ago but had totally forgotten them.

As I said above, there are substantial differences in the takes.  To use a philological term, the takes are actually variant readings.  The most stunning of these to me was that three of the five takes started with a pick-up chord before the B-flat fanfare—a “flat-VI” G-flat major chord, swooping up with a crescendo into the familiar downbeat.  Really?!?  That famous first chord almost wasn’t the first chord?!?  Searching around on the web, these tracks are not generally available from legal streaming services.  (For the moment, at least, these can be accessed on archive.org (starting at 05:00 on this track [Track 119]), although I can't imagine that they’ll always be accessible there.)

An authorized score for this music has been published as a concert suite for full orchestra, but I avoided consulting it until after listening and transcribing (sometimes from half-speed files) what I could hear, and consulting others with acute ears.  When I finally compared my five transcriptions with the published score, there were even more surprises.  Here is the middle of the score—brass and percussion—for the three-bar opening fanfare as published:

SOURCE:  cropped scan of p. 1 of STAR WARS:  Suite for Orchestra, I. Main Title

There are a number of differences between the published version and the five takes from the 1977 sessions; indeed, none of those takes is accurately represented by the published score, at least as far as my ear can tell—even Take 19, which was the one used for the film soundtrack for this passage (complete with a very obvious cracked trumpet note on beat 3 of b. 3).  For example, while it is a most effective and even obvious idea to have the first trumpet on the high B-flat (written C here) tonic for the initial chord, I don’t hear this on any of the takes from the 1977 soundtrack.  In all but Take 20 that pitch is there in woodwinds and tremolo strings, but not trumpet.  That trumpet top B-flat is very clear on the soundtrack of The Empire Strikes Back (1980), and so perhaps it was a very early revision.  (I also have my suspicions that the Empire main title is using a larger brass section, maybe 4 trumpets and 4 trombones; I have no way of confirming that, but it sounds thicker to my ear, even when the parts are in unison.)

Another way in which the score differs from the 1977 takes is the thirty-second notes in (at least) the trombone part in b. 1.  I have puzzled over this:  is it just sloppy playing?  Were the 32nds on the part, but the players just weren’t able to articulate them fast enough?  When I listen to it at half-speed, I hear something that sounds like triplets (i.e., only three iterations in each of those half-beats).  Are the 32nds being muddled up by the players, or were they reading triplet 16ths?  With the trumpets in b. 2, it is harder to tell, as in each of the five takes I hear the trumpets play four iterations, but except for Take 19 (the one on the film), these notes slightly too early and too slow, as if triplet-16ths.  The effect is this:

SOURCE:  my transcription of the rhythms suggested by the playing on Take 16, 17, 18, and 20; only on Take 19 are the trumpets late enough for me to believe they are reading (but not quite playing) 32nds.  With the trombones, sometimes the first of the triplet groups isn’t clearly articulated (Take 17 especially).  Also, in Take 20 there is no initial chord from the trumpets (nor indeed anyone else).

As it happens, on the soundtrack of Empire, the trumpets are very clearly 32nds, but the trombones—on which it is admittedly more difficult to make such quick notes speak cleanly—are still pretty muddy.

I mentioned above the G-flat swoosh into the downbeat on Takes 16, 17 and 18.  (This upbeat is even slightly elongated in Takes 17 and 18—just a bit longer than a beat of the ensuing tempo.  And I haven’t tried to calculate where this extra beat would need to begin on the film to preserve the rest of the synchronization, but I think it would have to be when the screen is still completely black just before the text STAR WARS appears.)   For these takes the harp glissando that appears in b. 3 of the published score happens instead as a component of this initial swoosh (and presumably with the harp set on a G-flat major scale, although it is devilishly hard to discern).  When the upbeat was deleted for Take 19 (and 20), the glissando was moved to the end of b. 3 and adapted to the dominant harmony.

The upbeat of the first three takes also has a woodwind flourish—a scale leading up to the high B-flat.  The strings seem to have this too, and certainly—as in the familiar version—from the downbeat the violins prolong high B-flats (in octaves) with a tremolo for most of these three bars, doubled by the triangle roll.  (Trill?  What do you call it on the triangle?)  That shimmering background is a memorable hallmark of this musical moment.  Thus it is astonishing to discover that Take 20—the last of the takes, the one just past the “keeper” (no. 19)—begins very sparsely:  no big chord, no string tremolo, no woodwinds.  Just the cymbal clash, the rolling triangle and the unison trombones.  The horns, trumpets and tuba accumulate gradually, but there is no hint of the rest of the ensemble until the pick-up to b. 4.  (This starts at 14:38 on the track on archive.org, and its spareness really must be heard to be believed.)  Was this seriously considered for the iconic introduction?  Shocking as it may seem with the benefit of hindsight, this was a plausible alternative in the studio.

I haven’t attempted to scrutinize the remainder of the main title music on these five tracks so carefully, but as far as I can tell, all of the variants in the first portion (except for the relocation of the harp gliss. mentioned above) were accomplished by having various players remain silent at designated places.  Thus all of these could be accomplished with the same parts on the stands.  In that case, maybe in Take 20 they recognized that they had just gone too far with the “tacet” instructions.  In any case, these are all orchestration details.  Did Williams orchestrate this passage, or his collaborator Herbert W. Spencer?  Was Williams just editing from the podium?   Probably these questions could be answered—the composer is still with us, after all—but answering them is beyond the scope of my blog.  (There's a reason I work with dead composers:  they can’t answer back.)  Still, examining these more closely has been a good reminder for me that the “inevitable” may rarely seem that way at the time.

ADDENDUM   22 March 2017
Brendan Finan responds to this post on his blog at http://www.brendanfinan.net/wordpress/a-star-wars-speculation/


ADDENDUM   20 July 2019
A follow-up with information from someone who has seen the sources: 
http://www.settlingscoresblog.net/p/star-wars.html