(M.J.E. / Composer listings / Agnew / Errors)

Piano music by Roy Agnew
Complete edition published by The Keys Press
Edited by Prof. Larry Sitsky

List of probable errors in the edition


Compiled by Michael Edwards


Introduction

      After many years during which most of Australian composer Roy Agnew's piano music has been out of print, it is currently being reprinted in successive volumes in a complete edition by The Keys Press, edited by Prof. Larry Sitsky.
      This is something I welcome, since Agnew is one of my favourite composers, and I think it is regrettable that his music has been neglected in more recent years; but at the same time, I believe the new edition, for reasons that I cannot determine, contains many dozens of mistakes. In the case of pieces where I also have the original edition, I was able to verify surmises I made by studying the new edition without initially comparing it with the original, and I found that in almost every instance I was completely correct. This gives me the confidence to identify many mistakes I believe I have found in pieces of which I do not have the original edition, although such surmises of course have to be more tentative, because I cannot verify them.
      On this web page I am going to list these probable mistakes and discuss my reasons for supposing them to be mistakes.
      Larry Sitsky has made deliberate alterations to the music, and these are usually marked in the score with various typographic devices as editorial alterations or additions, although in some cases not as clearly or precisely as I would have liked. While I do not always agree with what he has done, and in most cases see no reason whatever to alter Agnew's original version, I have to accept that this is a Sitsky edition of Agnew, not an "urtext" of Agnew. I will therefore not be focusing on these deliberate changes, and will refrain from discussing my opinion of them, and will concentrate on those changes or mistakes which I feel sure are unintentional. (Where the markings to identify what is editorial seem unclear, though, I will discuss that. I accept, with a few misgivings, the inclusion of editorial alterations; but in such cases I think it is very important that it be easy to distinguish what the editor has done from what the composer did. This edition fails in that regard in a few instances, in my opinion.)
      Whether Prof. Sitsky was, or is, aware of these mistakes I'm sure I've found, I cannot be sure, and it greatly puzzles me, because I would have thought that routine proof-reading before the music went to the printer would pick them up. I feel sure that many, even most, of these mistakes are typesetting mistakes rather than editing mistakes on Prof. Sitsky's part. Support for this idea can be found in the fact that there are a number of mistakes which are purely typographical ones, such as improper vertical alignment of notes, and the overall appearance of the scores suggests that the typesetting may not have been done as carefully as it might have been. Even a brief glance at the original and new editions of a score, without studying any detail at all, makes the difference in typesetting style obvious. Even so, proof-reading surely should have picked up most of the actual errors, if not the shortcomings in typesetting style.
      In a very few cases the original edition has mistakes too, and I will include and discuss these too, regardless of whether the new edition perpetuates the mistake or corrects it. I will in each such case mention that it was in the original edition. In short, I will mention anything I am aware of in any editions that will contribute to a more accurate edition, should that be produced one day.
      What mystifies me about this situation is that Larry Sitsky is beyond doubt one of Australia's foremost musicians and musical scholars, and he has made a speciality of Roy Agnew's music, and his authority on the subject cannot be contradicted lightly. Yet I must contradict much of what appears in this new edition; the mistakes just make themselves too obvious.
      What may make this seem even more audacious of me is the fact that I have no formal musical qualifications whatsoever, and I am not in the least known in the musical world at large. Yet I have a considerable familiarity with a huge range of music from over the last two centuries or so, and a particular familiarity with much of Agnew's music, so I believe I am in a position to assert the presence of these mistakes. And I will give full reasons for each instance, and I believe my reasoning will stand on its own merits - not on my own qualifications or reputation, neither of which I have.
      In some cases, I am not sure that something is a mistake, but feel only that it possibly is. I have included all such doubtful cases too, and have indicated that they are doubtful. But if a corrected version of this music is ever printed one day, I believe there is enough doubt about these marginal cases that they should at least be investigated and preferably compared with the original editions, which seem, at least in the ones I possess myself, to be almost completely free of errors; and of course the composer's manuscripts, if available, would be a very valuable source to compare them with.
      In yet other cases I feel certain there is a mistake, but two or more possible corrections seem plausible. In such cases I may suggest a reading which seems most likely, and may also mention one or two alternatives. In all cases I will fully back up my statements with reasons.
      But of course before any corrections are done, the composer's manuscript and/or original editions should be consulted to resolve the matter. Needless to say, this would be an extremely time-consuming task, taking perhaps dozens of hours, comparing the original sources and the new edition literally bar by bar, for page after page; but, in my opinion, for a reliable edition to be produced of this music, it is a task that simply must be done. I see no way around this, no short-cuts.

      The items in the following list fall broadly into two categories: errors I have discovered by comparing the new edition with the old one (which I sometimes surmised from the new edition alone, before making the comparison with the old edition, which bolsters my feeling of competence to compile this list), and errors in pieces where I do not have access to any other edition, where for strong musical reasons I feel sure a misprint has occurred.
      Following on from this, the pieces themselves fall into two categories: those where I own an earlier edition, and those which I am seeing for the first time in the new edition. Given that many of my comments are based on a comparison with the earlier editions, I would like to list the early editions I have, so that anyone who wants to pursue the matter can tell exactly what I am referring to when, in the following notes, I refer to "the original edition".


Original editions consulted in compiling this list

     Sonata Poème                       Allan & Co. Prop. Ltd.       1936
     Capricornia (Sonata Legend)        Augener Ltd.                 1949
     Australian Forest Pieces           Nicholson & Co., Limited     1913
     Youthful Fancies                   W. H. Paling Pty. Ltd.       1936
     Poem no. 1                         Allan & Co., Pty. Ltd.       1922
     Poem no. 2                         Allan & Co., Pty. Ltd.       1922
     Toccata Tragica                    Allan & Co., Pty. Ltd.       1922
       (There is another version of this piece published by Curwen called
       Poème Tragique, which was not available to me.)
     Three Poems                        Augener Ltd.                 1927 
     no. 4 of Four Preludes             Oxford University Press      1927
     Three Preludes                     Augener Ltd.                 1927
     Etude                              J. Curwen & Sons Ltd.        1924
     A May Day                          Oxford University Press      1927
     Rural Sketches                     Augener Ltd.                 1927
     Two Pieces ("An English Dance",
       "A Country Lane")                Oxford University Press      1927
     Contrasts                          The Arthur P. Schmidt Co.    1927
     Deirdre's Lament                   Allan & Co., Pty. Ltd.       1922
     Rabbit Hill                        Oxford University Press      1928
     Dance of the Wild Men              Allan & Co., Pty. Ltd.       1921
     Toccata                            J. & W. Chester, Ltd.        1933
     Drifting Mists                     Augener Ltd.                 1931
     Whither, and Exaltation            Augener Ltd.                 1931
     Noontide                           W. Paxton & Co., Ltd.        1935
     Holiday Suite                      W. H. Paling Pty. Limited    1937
     Album Leaf                         Augener Ltd.                 1949
     Sea Surge                          Augener Ltd.                 1949
     Two Pianoforte Solos
       ("In Meditation",
       "Looking Back")                  W. H. Paling & Co., Ltd.     undated

      So I will now present my itemized list of errors, and give full reasons for each item. I do this not in a spirit of being carping or fault-finding, but in a cooperative spirit, in the hope that it will one day contribute to a corrected version of this music being printed one day, unlikely though that may seem.
      Items were occasionally difficult to describe clearly in words - but if you check them in the actual edition itself as you read them, I think they should be reasonably easy to follow. The list is designed to be read in conjunction with the actual scores, and will be meaningless to anyone who doesn't have a copy of Agnew's music. It would be good if I could include actual musical notation for the more complicated instances; but, since I don't have the resources for doing this, nor the knowledge of how to do it, I will have to make do with a careful description in words. I apologize if some of my descriptions seem very wordy and detailed; but there was no other way. In such cases (fortunately very few) I have given priority to clarity over brevity.
      Please note: in the limited-edition Sonatas volume (which I hope won't remain a limited edition of only 250 copies permanently), where bar numbers are provided, I refer to passages by identifying the bar number. However, bar numbers are not provided in the other volumes, so I will identify passages by giving in turn the page number, system number, and bar number within that system. In numbering bars I have counted only complete bars, and ignored incomplete bars containing an upbeat. This situation usually occurs only at the start of a piece.
      In the following list, I've marked each item with either an asterisk or a question mark: the asterisk marks items I feel almost certain are mistakes; the question mark denotes items where I do not feel sure, but it seems uncertain enough that it should at least be looked into more closely, and compared with the original sources. Where I feel sure there is a mistake, but not sure of how it should be corrected, and have suggested two or more possible alternatives, I have marked the item with an asterisk: it does not indicate that I am certain about the preferred version I suggest, but merely that I am certain there is a mistake.
      In referring to musical notes, I do not have musical symbols available for use in this list: therefore I use the symbol "#" for sharps (as in F#), the letter "b" for flats (as in Bb), "x" for double-sharps (as in Fx), and "bb" for double-flats (as in Bbb). A note-name without these symbols is always to be taken as natural (as in D - that is, D natural). If I need to emphasize that a note is natural, as against sharp or flat, I will say something like "D-nat.", since I cannot even approximate a natural sign with any characters available to me.
      I will deal with the Sonatas volume first, then work through the other volumes of shorter piano works in order. I will provide links on this page at various points to go straight to particular volumes or pieces, including a complete list of Agnew's piano works and full links right now.

Michael Edwards,
Victoria, Australia.

E-mail me about this music.


NOTE:
      Click here if you need an explanation for the strange appearance of the e-mail address which will appear when you click on the e-mail link, or if you don't know what you need to do to make the e-mail address work properly.



List of volumes

The 6 Sonatas (limited edition volume)
    Symphonic Poem (La Belle Dame Sans Merci)     Fantasie Sonata     Sonata (1929)     Sonata Poème
    Sonata Ballade     Sonata Legend (Capricornia)     "Second Sonata" (Ossianic) (incomplete work)
Volume 1
    Australian Forest Pieces
       1. Gnome Dance     2. When Evening Shadows Fall     3. Forest Nymphs at Play
       4. Night in the Forest     5. By a Quiet Stream     6. The Forest Grandeur
    Youthful Fancies
       1. The Merry-Go-Round     2. The Gurgling Brook     3. The Sunlit Glade     4. Sleeping Child
Volume 2: The Complete Poems
    Two Poems (1922)
       1.     2. (To the Sunshine)
    Poème Tragique (Toccata Tragica)
    Three Poems (1927)
       1.     2.     3. [G#-minor-ish]
Volume 3: The Complete Preludes
    Four Preludes (1925)
       1.     2. [G# minor]     3. (The Wind)     4.
    Three Preludes (1927)
       1. [A major]     2. [Eb minor]     3. [B major]
Volume 4
    Etude     A Dance Impression     A May Day     An Autumn Morning     Capriccio
Volume 5
    Pangbourne Fields
    Three Lyrics
       1. The Falling Snow     2. A Quest     3. The Happy Lad
    Rural Sketches
       1. The Shepherd on the Hill     2. The Fairy Dell     3. A Starry Night     4. At the Fair
Volume 6
    Two Pieces
       An English Dance     A Country Lane
    Contrasts
       1. A Child's Dream     2. Country Dance     3. Winter Solitude     4. Elegy     5. April on the Hills
Volume 7
    Deirdre's Lament     Elf Dance     Rabbit Hill     Rhapsody
Volume 8
    Dance of the Wild Men     The Windy Hill     Toccata
Volume 9
    Drifting Mists
    Whither, and Exaltation
       Whither     Exaltation
    Before Dawn
    Noontide
    Trains
Volume 10
    Holiday Suite
       1. Spiders     2. Holidays     3. Lullaby     4. The Party     5. March of the Soldier Ants
    Album Leaf
    Sea Surge
    Two Pieces
       In Meditation     Looking Back
Volume 11: Previously Unpublished Pieces
    Nocturne     Will o' the Wisp I     Will o' the Wisp II
Volume 12: Piano Duets
    Green Valley     The Village Fair

[ last / first volume of shorter works ]

The 6 Sonatas (limited edition volume):
 Symphonic Poem (La Belle Dame Sans Merci); Fantasie Sonata; Sonata (1929); Sonata Poème;
Sonata Ballade; Sonata Legend (Capricornia); "Second Sonata" (Ossianic) (incomplete work)

Symphonic Poem (La Belle Dame Sans Merci)      -----===** not yet complete **===-----

Note: In this piece, there are passages which are repeated almost note for note; when errors occur in corresponding bars in both passages, they are discussed in adjacent entries, or even in the same entry when that is convenient, breaking the normal pattern of going through the bars in order.

* Bar 14, crotchet 4, r.h.:
      Error: Lower B should be Bb.
      Reason: Compare with l.h., where the B is flat; also compare with the same chord in the next bar. The melody voice as a whole moves in octaves, so the B-nat. doesn't fit for that reason, also. If the B were intended, it would probably have been written as Cb.

* Bar 17, crotchet 3, l.h.:
      Error: Probably Gb-Ab (major 9th) should be an Ab octave - or possibly a Gb octave.
      Reason: The surrounding l.h. notes are octaves, and the maj. 9th doesn't seem to fit the texture of the passage as a whole.
      Ab seems the more likely alternative, because it continues the pattern of root-position dom.-9th chords. But a case could be made for Gb, since the chord Gb E Gb Ab Bb C Eb makes sense too. After all, this chord must be slightly different from the preceding chords anyway, whichever way you read the bass, since the presence of the E in the r.h. means that, even with the Ab in the bass, the chord would not be a pure dom.-9th chord like the preceding chords.

* Bar 27, last chord in r.h.:
      Error: The lower C needs a natural sign (to agree with upper note of the octave).
      Reason: The outer notes of this series of chords in the r.h. move in octaves both before and after, so a false relation here doesn't make sense. If it is intended though, a confirming sharp sign would seem advisable.
      Another possibility might be that both C's should be sharp, in which case the natural sign for the top C should be changed to a sharp sign. (Compare with bar 175, next paragraph.) But the E included in the chord seems to suggest C, because it would continue the pattern of parallel maj. 3rds. At this point both hands are independently following streams of parallel chords based on the whole-tone scale, and C would definitely fit this pattern better than C#.

? Bar 27, last chord in r.h. :
      Error: Possibly a G# or Ab has been omitted from the chord.
      Reason: All the preceding and following chords in the r.h. are augmented triads, which suggests that maybe this one should be too.

* Bar 175, last chord in r.h.:
      Error: Both C's need natural signs.
      Reason: This would agree with the parallel passage in bar 27, where one of the C's is natural, and probably both should be. See the previous but one entry for a fuller discussion of this point. If, in spite of this, both C's should be C# here, then no change needs to be made here.

* Bar 29, second-last quaver in r.h.:
      Error: The G needs a natural sign.
      Reason: All the preceding r.h. chords are augmented triads, following the l.h. parallel major triads; it seems this last r.h. harmony should be an augmented triad, too. This is supported by bar 177, which is exactly parallel, where the G is given a natural sign.

*? Bar 32, crotchet 1 in r.h.:
      Error: The C probably needs a sharp sign.
      Reason: C# seems to fit the harmony better than C. Compare with bar 180, which is parallel, and presumably meant to be the same, and which has the sharp sign. The following C would then need a natural sign, of course.

*? Bars 46 and 193:
      Error:
      Reason:

*? Bars 46 and 193: These bars are obviously closely parallel, apart from being in different keys. This suggests to me that maybe the inner r.h. voice may be wrong in one of them: perhaps, in 193, the 2nd and 3rd notes should be Gb-F, not F-Fb, to agree with 46; or, if I've got it the wrong way round, then the 2nd and 3rd notes in 46 should be D-Db, not Eb-D. (This version seems less likely to me.)

*? Bar , :
      Error:
      Reason:

*? Bar 50, last quaver, r.h.: Quaver rest needed to complete the inner voice. (It appears in the parallel bar 199.)

*? Bars 51 and 199: These bars are obviously parallel: indeed, situated in quite lengthy passages which are parallel. Because some of the parallel bars appear to be exact, and others only approximate, and because I can't be sure whether these particular bars are intended to be exactly or only approximately the same, any claim about possible errors in one or both of these bars is tentative only; but my feeling is that there may be a couple of errors here, which perhaps could be checked with the original source.

      If my hunch is right here, it would seem that either bar 51 should





become: or else bar 199 should become:





(The points of difference are in the l.h.: two notes in the inner voice, stems pointing down, are one note-degree different, after transposition has been taken into account, and the composition of the chord in the second-last quaver is different. Also (I feel sure about this one, at least), the first D in the l.h. in bar 199 should be Db, to agree with bar 51 and to preserve the Db-major tonality.)

*? Bar 71, crotchet 1, r.h.: Minim should be G#-D-F#. (This would parallel the tenor line exactly the way the surrounding r.h. chords do. It seems that there is intended to be a series of parallel chords here of 4 notes each: the 3 notes in each r.h. chord plus the top notes in the l.h.)

*? Bars 82-84, l.h.: Several crotchet stems omitted. (This is a passage Larry Sitsky rearranged. However, assuming that it should be effectively the same as Agnew's version, merely redistributed between the two hands, the lower l.h. notes all need crotchet stems added to give them the right duration (except where they tie into a following quaver); but many of these crotchet stems are missing, thinning out the passage too much. A few of the upper l.h. notes need crotchet stems, too, for similar reasons.)

*? Bar 83, quavers 1 and 2, l.h.: The first G should be natural (tied to the G in the previous bar), and the second G, flat. (Compare with the rest of the bass-line, which moves in even crotchets; also, compare with Agnew's version given in the Appendix.)

*? Bar 83, quaver 8, l.h.: The final E should be flat, not natural. (Similar reasons as in preceding paragraph.)

*? Bar 93: The meaning of the footnote doesn't seem clear to me: Did the editor add the triplet notes out of nothing, or convert crotchet chords to triplet figuration? Or is the footnote possibly saying that he merely added the triplet signs?

*? Bar 101, crotchet 1, r.h.: Db should be natural (to agree with l.h.). The following quaver D would then need a flat sign.

*? Bar 102, crotchet 4, l.h.: B needs a natural sign. (This would agree with B in r.h., and make the l.h. chord fit into the series of parallel dom. 7ths.)

*? Bar 106, crotchets 1-2, r.h.: First D needs a natural sign, second a flat sign (to make them agree with l.h.)

*? Bar 123, crotchet 3, l.h.: B needs either a natural or a confirming flat. (The former would agree with r.h.; the latter would confirm the false relation.) Perhaps the low E needs a natural, too, because the 8va sign might be regarded as cancelling the previous E natural.

*? Bar 138, crotchet 3, r.h.: A needs a sharp. (Clearly necessary to agree with overall harmonic sense; all the other A's are sharp.)

*? Bar 151, crotchet 1, r.h.: E needs flat sign (to agree with l.h. quaver 2 notes earlier), or else a natural (to confirm disagreement with l.h., which seems unlikely to me).

*? Bar 156, 6th quaver, r.h.: If false relation in bar 157 is correct (unlikely), and intended here also, the upper G needs a natural sign.

*? Bar 157, 6th quaver, r.h.: Upper C should be sharp, not natural (assuming a per. 8ve is intended, as in the preceding parallel bar).

*? Bar 163, crotchet 3, l.h.: Lower Eb in octave semiquaver at end of bar needs a flat sign. (This would agree with the top note, and the E natural in the semiquaver run earlier in the bar needs to be cancelled.)

*? Bars 164-165, r.h.: A 5-note chord is tied between these two bars, but only 4 tie lines are present; the top Ab's need to be tied also. (I presume all 5 notes are intended to be tied.

*? Bar 173, crotchets 1-2, l.h.: In the first crotchet, the lower note should have a tie leading into it (the Db at the end of the previous bar has a tie leading out of it, and the parallel passage in bars 24-25 contains a similar tie); in the second crotchet, the upper D needs a reminding natural sign.

*? Bar 175: Perhaps the word "steady" (in editorial square brackets) needs to be added above the top stave at the beginning of the bar. (It is present in bar 27, which is identical.)

*? Bar 178, crotchet 4, r.h.: The last G should be sharp. (This would agree with the parallel bar 30, and also agree with the same figure an octave higher in the first half of the same bar.)

*? Bar 178, crotchet 4, l.h.: The bass clef should appear before the two semiquavers, not after. (Gb-B and C in the treble make no sense here, and don't agree with bar 30, which places the clef correctly.)

*? Bar 201, crotchet 2, l.h.: Possibly C-E should be Cb-Eb. (This would agree with the first chord in the bar, and also correspond more closely to the parallel bar 53; although this comparison may not be valid, because the bars are only similar, not identical, it is persuasive to me, all the same.)

Fantasie Sonata      -----===** not yet complete **===-----

Although I have never seen the original edition of this work, at least advertising for it on the back cover of other Agnew works (published by Augener) gives the title as "Fantaisie Sonata" (extra "i" in "Fantasie").

Note: In this piece, there are passages which are repeated almost note for note; when errors occur in corresponding bars in both passages, they are discussed in adjacent entries, or even in the same entry when that is convenient, breaking the normal pattern of going through the bars in order.

*? Bar 2, dotted minim, l.h.: Perhaps E should be Eb. (This would correspond with the following chord a semitone higher, and many of the chords in this sonata are dom.-7-based. E natural just sounds wrong to my ear.)

*? Bars 12-15, r.h.: Should the upper notes of the trills be 1 or 3 semitones higher? (In the first trill, for example, should be F-F# or F-G#? - and similarly for the other trills.) (Without a doubt, the auxiliary notes should be 3 semitones higher, according to the standard rules of musical notation. But trills of an augmented 2nd are very unusual, and I have seen this notation in other music where it seems that the note a semitone higher is intended. In any case, the notation given here is slightly ambiguous to me.)

xx *? Bars 30 and 224, second-last r.h. chord in each: Lower D needs a natural sign. (D# not only disagrees with the upper D natural, but is contraindicated by the immediately preceding and following Eb's.)

xx *? Bars 45 and 239, quaver 6 in each, l.h.: Upper B needs a flat to cancel the earlier natural. (It is obviously intended that all the l.h. octaves in this passage be Bb.)

xx *? Bars 45 and 239, quaver 9 in each, l.h.: In the earlier bar, the notes are Bb-Gb; in the latter a Bb octave. Should either be changed to agree with the other?

xx *? Bars 46 and 240, quaver 7 in each, r.h.: The earlier bar has F, the later F# (arising from an earlier sharp still in force); naturalize the F# in the later bar. (F would agree with the F in l.h.)

xx *? Bars 46 and 240, quaver 8 in each, r.h.: E needs a natural sign to cancel an earlier Eb (obviously not intended here).

51, quaver 1, r.h.: The ties at the end of the previous bar should be shown continuing into the Eb octave at the start of this bar.

xx *? Bars 59 and 253, second semiquaver group in each, r.h.: Should the Bs be natural? (This would fit in with the overall pattern of the semiquaver tremolos being in major 3rds.)

xx *? Bars 60 and 254, last semiquaver in each, r.h.: Should the B be A (natural)? (This would continue the pattern of maj.-3rd tremolos.)

*? Bar 63, start of bar, l.h.: Arpeggio sign should start at bottom of chord (in agreement with the same chord in bar 257).

*? Bar 67, start of bar, l.h.: Dots missing on F# octave.

xx *? Bars 79 and 273, second group of semiquavers in each, r.h.

xx *? Bar 80, last semiquaver, r.h.: Should G# be F#? (This would continue the pattern of maj.-3rd tremolos.)

xx *? Bar 82, semiquaver 3, r.h.: C needs a flat sign (to continue maj.-3rd tremolos, and to agree with Cb in l.h. at the beginning of the bar).

*? Bar 106, r.h.: In tied chord Eb-A-F, the tie for the top note is missing. (Clearly intended to match similar chords 2 and 4 bars earlier.)

*? Bar 107, second chord, r.h.: Dots on crotchet missing.

*? Bar 109, last quaver, l.h. in top stave: Db should be marked "loco". (Without the "loco", the melodic shape (compared to the preceding and following bars) is spoilt.)

*? Bar 162, first chord, l.h.: Should E be Eb, to make an Ab-min. triad? (No concrete reason, except that it seems right that way to my ear.)

*? Bar 170, just after bass clef, r.h.: E should be Eb. (Agrees with Eb an octave lower, and fits in with the overall harmony in that bar better.)

*? Bar 172: I have to be a bit vague here, because I can't pinpoint it; but there is something about this bar that seems wrong. Perhaps it's the false relations created by E and B naturals against E and B flats. Is it possible that a couple of flat signs have been left out of this bar, or perhaps one or two wrongly inserted?

*? Bar 189: Should this bar consist simply of the single quaver (even though there is no 1/8 time signature)? (The empty space before the quaver makes it look as if something might have been accidentally omitted from the type-setting.)

*? Bar 223, end of bar, r.h.: A treble clef should be inserted at the very end of the bar. (Only so would the following bar make sense, as well as agree with bar 29.) The treble clef at the end of bar 224 should also be removed.

*? Bar 224, dotted crotchet, l.h.: C should be C# (to agree with bar 30; also, C# seems to fit the harmony better than C would).

*? Bar 226 xx

*? Bar 227, quaver 5, l.h.: The low C should be C# (agrees with bar 33, as well with the C# an octave higher immediately following).

*? Bar 245, last quaver, r.h.: The E octaves should be Eb (to agree with bar 51; also fits into the current harmony better; also, E tied into the next bar doesn't make sense with Fb simultaneously sounding).

*? Bar 247, second dotted crotchet, l.h.: F-E should be E octave. (This is better harmonically and texturally, and agrees with the earlier parallel passage at bar 53.)

*? Bar 252, last semiquaver, l.h.: Should the last semiquaver be in square brackets, to indicate an editorial insertion? (The corresponding note in bar 58 is so marked.)

*? Bar 256, start of bar, l.h.: Bb dotted minim missing in bass. (It is present in the corresponding bar 62.)

*? Bar 257: Perhaps this bar should be marked "(the middle notes very softly)" (to match bar 63, so marked).

*? Bar 259, start of bar, l.h.: Arpeggio sign should start at bottom of chord (in agreement with the same chord in bar 65).

*? Bar 262, dotted minim, l.h.: F# in bass needs to be marked with a tie continuing from the previous page (which ends with such a tie). Or, if it is not to be tied, the note needs a sharp sign; but the parallel passage in bars 67-68 includes the tie.

*? Bar 265, l.h.: First crotchet in tenor needs a dot. Second and third notes in tenor voice are positioned on stave one quaver earlier than they should be.

XXup toXX bar 268

*? Bar 286, first l.h. chord: Clearly this chord is tied from the previous bar; but only the lower three notes have ties running into them, not the top Ab.

*? Bars 287-288, beginning of bar, l.h.: This is the first of two bars Larry Sitsky modified. (He expanded 1 bar by Agnew into two bars.) The Appendix gives Agnew's version, which includes a l.h. chord not found in Larry Sitsky's version. Was the dropping of this intentional, or a misprint? If the latter, the following notes, tied together, should be inserted in bars 287-288 to parallel Agnew's dotted semibreve chord:





(In bar 288 this would necessitate a third, bass-clef stave, or else a number of leger lines to accommodate the low Ab. Or perhaps Debussy's device of combining two clefs on the same stave at the same time would do the trick.)

*? Bars 289-292, quaver 7 in each bar, r.h.: The top note of the l.h. dotted quavers is taken by the r.h. in each bar, but the r.h. notes lack the necessary dot.

*? Bars 295 and 314, first note in l.h. in each bar: The bass notes as given by Agnew are B octaves at the bottom of the piano. Larry Sitsky has added an F to both of these octaves; but is it possible that a sharp sign has been accidentally left out, making the F's F#'s instead? (F# in both places seems to fit the harmonic context better than F; surrounding F's are all sharp, and 2 Gb's are present in bar 295.)

Sonata (1929)      -----===** not yet complete **===-----

*? Bar 5, quaver 5, middle stave: The minim A should be Ab. (Compare with the same voice in the next bar, and the manuscript for this page reproduced in the Appendix.)

*? Bar 5, quaver 6, middle stave: The quaver Ab should be A. (According to next bar and manuscript.)

*? Bars 5 and 6, l.h. semibreve in each bar: The arpeggio signs should start at the bottom notes of the two chords, not the second-lowest (as shown in manuscript page in the Appendix).

*? Bar 7, quaver 5, middle stave: Gb quavers should be natural. (Gb-F#-Gb makes no sense; manuscript page in Appendix agrees.)

*? Bar 8, quaver 6, middle stave: Lower Eb (quaver) should be natural. (The falling and rising 2nd in this motif is always a min. 2nd; manuscript page in Appendix agrees.)

*? Bar 11, quaver 6, middle stave: Bb (quaver) should be natural. (Same reason as in previous paragraph.)

*? Bar 15, last quaver, r.h.: E should be Eb. (Cb-E is enharmonically wrong; parallelism with preceding and following chords suggests a maj. 3rd here.)

*? Bars 42-43, middle stave: Bb as well as Eb should be tied over the bar-line. (Bar 43, but not bar 42, is included in the manuscript page in the Appendix, but both these notes in bar 43 have continuation ties leading to them.)

*? Bars 43 and 44, quaver 6 in each, top stave: "R.H." is given for both upward arpeggios, but the manuscript page in the Appendix gives "L.H.". Only the second "R.H." is marked as editorial.

*? Bars 54 and 55, quaver 1 in each, r.h.: Top E should be Eb (the bottom E added by the editor is flat); or else, if the false relation is intended (which seems unlikely), a confirming natural would prevent the appearance of a misprint.

*? Bars 60-61, l.h.: Bass clef should be inserted at the end of bar 60, and treble clef at start of bar 61 changed to a bass clef; bass clef at end of bar 61 should be removed. (The l.h. passage in bar 61 only makes sense melodically and harmonically if construed as being in the bass clef.)

*? Bars 97 and 98, crotchet 2 in each, r.h.: The lower C's read as if they are C# (because of the earlier sharp in the C space), in spite of the false relation this causes with the higher octave; but comparison with the B octaves on the following beat suggests that an octave may have been intended, thus requiring that the lower C's be natural. A confirming natural or sharp seems necessary for the lower C's in question.

*? Bar 111, crotchet 5, r.h.: Is there a slight possibility that the E crotchet should be Eb to agree with the l.h.?

xx *? Bar 173, quaver 7, r.h.

xx *? Bar 173:

xx Bar 173

*? Bar 177, B octave in top stave: Should the "8va" apply to this, too, or should be marked "loco"? ("Loco" would seem better from the point of view of the texture of this chord as a whole.)

*? Bar 177, 2nd-last quaver in middle stave: The notes are F#-E#, but the context seems to suggest the entire triplet should be in octaves. Should either of the notes be changed to make an octave with the other?

*? Bars 178 and 179, crotchet 2 in each bar: Because of an earlier E#, the E octave in the bottom stave actually reads as E-E#, reading upwards, creating an aug. 8ve; but it seems from the context that the note should be a per. 8ve, either on E or E# (I would consider E# more likely, judging by all the other E#'s in these bars, and the total absence of E naturals). If so, then either the top E should have a natural in front of it, or the bottom one should have a sharp in front of it.

*? Bars 185 and 186, quaver 6 in each bar, l.h.: There are 2 C's in each chord: one with a sharp, one with no accidental. The one with no accidental needs a natural in front of it to cancel the C# in the preceding chord.

*? Bar 187, 2nd chord, l.h.: An A should possibly be added to the l.h., to match the pattern followed in the immediately preceding and following l.h. chords. If it is placed to the right of the Ab already present, the A in the next chord will need to be flattened.

Sonata Poème      -----===** not yet complete **===-----

The original edition of this piece was titled "Sonata Poeme" (without the grave accent in "Poeme")

Note: In this piece, there are passages which are repeated almost note for note; when errors occur in corresponding bars in both passages, they are discussed in adjacent entries, or even in the same entry when that is convenient, breaking the normal pattern of going through the bars in order.

37: There should be a diminuendo in this bar (a hairpin sign, not "dim."), lasting the entire duration of the bar (found in Allan's edition).

xx *? Bars 55 and 176: These bars correspond, yet in bar 55 the word "increase" occurs at the start, whereas with bar 176 it occurs near the end of the previous bar. The older Allan's edition does this too. I'm wondering if this is a mistake, and whether the manuscript should be checked to see whether either position should be changed to match the other.

xx*? Bar 57, crotchet 2, r.h.: The A should be G#. (G# fits the harmony, whereas A doesn't; it agrees with the Allan's edition.) (Although G# fits into the harmony, an intriguing thought occurs to me. I wonder if this could be a mistake, because F# would also fit the harmony, and it would fit the shape of this motif better, too. Comparison with bars 32, 38, 46, 106, 108, 110, 125 (and many others) would support this. I wonder if the manuscript actually gives F#, and the G# is a mistake which crept into the Allan's edition years ago.)

xx *? Bar 178, crotchet 2, r.h.: This bar agrees with the Allan's edition, but the speculation I made about a possible alternative reading in the previous paragraph would apply here also, in a passage which is identical.

*? Bar 71: The hairpin should begin below the r.h. G, not at the beginning of the bar (according to Allan's edition).

*? Bar 72, crotchet 2: The Bb in the treble should be B. (Agrees with parallel passage 2 bars earlier; necessary if the slur is to be regarded as a tie; agrees with the Allan's edition.)

*? Bar 79: The "Ped." and "*" signs are marked as editorial, yet they appear in the Allan's edition, and are thus presumably by the composer.

*? Bar 133, crotchet 3, l.h.: The G immediately after the treble clef should be Gb. (All the other G's in this bar are flat. This also agrees with the Allan's edition.)

*? Bar 141: The hairpin occurs too early, according to the Allan's edition: it should begin above the second F, and end above the B.

*? Bars 148 and 151, crotchet 1 in each, upper stave: All the notes given agree with the Allan's edition, and the version given certainly doesn't sound obviously wrong, but the following thought occurs to me. The downward run in these two bars is parallel, except for being transposed down a tritone the second time. Assuming the passages should be the same, comparison of the 2nd note in each of the two bars suggests that perhaps either the C# in bar 148 should be D, or else the Ab in bar 151 should be G. Would it be worth checking it with the manuscript?

*? Bar 155, crotchet 2, l.h.: Although it is doubtful, the thought occurs to me that just maybe the A in the bass should be Ab. (The surrounding chords are root-position dom.-7ths, and having Ab in the bass would continue the idea of the bass-line moving down by a min. 3rd, a progression that occurs both before and after this bar. This is purely my own speculation; the Allan's edition agrees with the new edition here.)

*? Bar 177, crotchet 4, r.h.: The F# should be F. (F# makes no sense harmonically or melodically, and the Allan's edition supports F, as does comparison with bar 56.)

*? Bar 178, minim 1, l.h.: The F in the minim chord should be F#. (F makes no sense harmonically, and the Allan's edition supports F#, as does comparison with bar 57.)

*? Bar 185, crotchet 3, l.h.: Possibly the G# should be G. (Although I cannot be so definite here as in other cases, the G# sounds wrong to my ear, and doesn't seem to fit the style of harmony that prevails in this passage. G also agrees with the Allan's edition.)

*? Bar 190, crotchet 5, l.h.: G should be Gb. (All G's are flat in r.h.; this also agrees with the Allan's edition.)

*? Bar 190, crotchet 6, l.h.: F-C-F should probably be A-Eb-F. (The dom. 7th effect this creates fits the harmonic idiom better than would the stark per. 4th and per. 5th of F-C-F. This also agrees with the Allan's edition.)

*? Bar 191: There should be a crescendo hairpin (according to the Allan's edition), starting with the first semiquaver, and ending with the last semiquaver.

*? Bar 192: The hairpin is not positioned quite right (according to the Allan's edition): it should start with the 3rd, not the 4th, semiquaver (is that being too fussy? - it's difficult to know where to draw the line); and it should end with the last semiquaver, not the end of the bar.

*? Bar 196: The hairpin is not positioned quite right (according to the Allan's edition): it should start with the 2nd, not the 1st, semiquaver; and it should end with the 2nd last semiquaver, not the 4th last. (There are a few more minute discrepancies in the positioning of hairpins between the two editions, but it would be tedious and pedantic to list them all; I might already have gone too far with that. I cannot believe that the hairpins in bars 191-201 are all supposed to be slightly different, as I seem to have implied, and that a couple of these bars are supposed to lack one altogether. I wonder if all the bars should have one, uniform in position.)

Sonata Ballade      -----===** not yet complete **===-----

Note: In this piece, there are passages which are repeated almost note for note; when errors occur in corresponding bars in both passages, they are discussed in adjacent entries, or even in the same entry when that is convenient, breaking the normal pattern of going through the bars in order.

*? Bar 7, crotchet 3, r.h.: Either the top note of the quaver should be C, or the bottom one B. (The context suggests a passage all in octaves.) Probably B (if chosen) should be Bb (suggested by harmonic context).

*? Bar 10, crotchet 3, l.h.: Should the last G be Gb? (Suggested by comparison with bar 9, and having G repeat doesn't seem to sound right, nor does the Eb-maj. triad that G would complete sound right in the context.)

xx *? Bar 51, crotchets 2-3, l.h.:

*? Bars 61-62, r.h.: There should be a lower tie as well as an upper tie connecting the two B's. (The two sets of note-stems make two voices in unison. The parallel passage in bars 185-186 are notated thus.)

xx *? Bar 71, crotchets 3-4, l.h.: The D's should be Db. (Compare with bar 70, where the l.h. takes exactly the same notes, albeit it earlier in the bar.)

*? Bars 73 and 197, crotchet 4 in each, both hands: The bars are exactly parallel, the later one being a semitone lower. If they are meant to be exactly the same (apart from the semitone transposition), it would seem, in crotchet 4, either that bar 73 should read Ab-Bb-G, or that bar 197 should read Gb-A-F.

*? Bar 93, crotchet 3, r.h.: The G octave semiquaver should be a quaver. (Comparison with nearby bars with the same motif (bars 90 and 98) rule out the other possible version for this beat: a dotted quaver followed by a semiquaver.)

*? Bar 123, crotchet 4, r.h.: The very last note in the bar should be C#, not C, and so should the first note in the next bar. (This fits in with the preceding C's in the bar, which are all sharp, and it makes the motif correspond to the same motif appearing in the l.h. in the previous two bars.)

*? Bar 129, crotchet 6, r.h.: The last chord should be a crotchet, not a quaver.

*? Bar 143, crotchets 3-4, r.h.: The sharps for the C's in the last two chords should probably be repeated because of the 8va sign starting at this point. (I know a bar-line ends the scope of an accidental; but does an 8va sign? I'm not sure; but most music I've seen regards a new accidental as necessary in such a situation.)

*? Bar 149, crotchet 3, l.h.: The E natural clearly marked seems a complete anomaly to me, and I would think it should be Eb. (It makes no harmonic sense at all, sounds ugly - to my ear, at least - and is not supported by the corresponding passage in the preceding bar. The repeated flat sign for the following Eb, cancelling the natural sign, makes it appear deliberate, not a mere misprint; but it does not persuade me in the slightest that it is correct. Perhaps the original edition and/or the manuscript should be checked.)

xx *? Bar 175

*? Bar 175, crotchet 5, l.h.: The Bb should probably be given a flat sign. (This raises another question about the scope of accidentals I'm not sure about: if an accidental occurs, the clef changes, then (after further notes) it changes back again, and the original note that had the accidental occurs again (as happens here), is the accidental still deemed to be in force, or does it have to be repeated? I would assume the latter, just to be on the safe side, but I don't actually know what the theoretically correct situation is here. Most music I've seen would repeat the accidental.)

*? Bar 178, crotchet 2, r.h.: The Bb immediately after the treble clef needs a flat sign.

xx *? Bar 189, crotchet 4, r.h.: The lowest D needs a sharp sign. (Because of all the other D#'s, it surely cannot be meant to be D natural.)

xx *? Bar 190, crotchet 4, l.h.: The lowest D needs a sharp sign. (Same reason as previous paragraph.)xxxx

*? Bar 196, crotchet 1, l.h.: The augmented triad is obviously tied from the end of the previous line. It needs tie lines running into it to make this clear.

*? Bar 212, last chord, r.h.: The final chord should be a quaver, not a semiquaver.

*? Bar 213, crotchet 3, r.h.: The E needs a flat sign in front of it. (The context of parallel major triads suggests that it is not intended to be natural.)

XX compare 39,51,175

Sonata Legend (Capricornia)      -----===** not yet complete **===-----

The original edition of this work is called "Capricornia (Sonata Legend)", whereas the new edition reverses main title and sub-title.

*? Bar 17, crotchet 2, r.h.: The lower Db should be D. (G major triad requires it; D agrees with the upper octave in the same beat; Augener edition has D. The manuscript page in the Appendix agrees also. It also compares properly with the parallel passage two bars later, despite a slightly different bass line there.)

*? Bar 21, crotchet 4, r.h.: The chord Fb-Cb-Fb should include an Ab too. (The full Fb-major triad occurs in the sample manuscript page included in the Appendix. The Augener edition omits the Ab also, so this mistake is obviously of early origin, and perpetuated in the new edition.)

*? Bar 25, crotchet 4: the word "broaden" should occur one quaver later. (In Augener edition; the manuscript page in the Appendix agrees.)

*? Bar 27, crotchet 3, r.h.: The F octave needs natural signs. (In spite of the 8va sign ending, and possibly terminating the scope of the earlier Fb accidental, the flat might still be read by some as being in force.)

*? Bars 33 and 34, crotchet 2 in each, r.h.: The two grace notes and the F minim following need to be slurred together (in both bars). (Given in Augener edition.)

*? Bar 38, 2nd quaver, r.h.: Lower Bb in this chord was added by Larry Sitsky. Could it be that it was meant to be B (or Cb)? (It would fit the harmony better. Also, it would enable the two low notes to be played by the thumb, which appears to be intended; A-Bb-Eb-Bb would probably be unplayable by pianists with right hands any smaller than Rachmaninov's. Compare with the corresponding chord in bar 40, which does follow what I am saying.)

*? Bar 40, 1st quaver, r.h.: D should be Db. (Agrees with parallel passage in bar 38, and agrees with Augener edition.)

*? Bar 52, crotchet 1, l.h.: The first group of semiquavers goes like this in the Augener edition:





This is not clear from the notation, nor from the end-note on p. 127, which, although referring to Larry Sitsky's alteration, makes it sound as if the original went thus:





*? Bar 57, crotchet 4, l.h.: F should be Fb. (Agrees with all the other Fb's in this bar, and with the Augener edition.)

*? Bar 60, start of bar: "very broad" is not in the Augener edition, but is not marked as an editorial addition either.

*? Bars 60-61, r.h.: Composer's slurring (different from Larry Sitsky's) was omitted (first two notes in each bar slurred together), even though this edition is generally giving both the composer's and the editor's slurring. (Composer's version is given in the Augener edition.)

*? Bars 60 and 61, crotchet 4 in each, r.h.: F in each bar should be Fb. (The prevailing harmony requires it, and this is given in the Augener edition. The natural signs look deliberate, but are not flagged as editorial alterations.)

*? Bars 63-64, r.h.: The slurring (composer's, not editor's) might be wrong in these two bars: the two slurs should be joined to form one long one. (Given thus in the Augener edition.)

*? Bar 65, crotchets 1-2, r.h.: The two crotchet chords have 4 notes each: only the top 3 are tied, but the bottom C and Bb need to be slurred together. (Given in Augener edition.)

*? Bar 73, quaver 1: The "Ped." indication is marked as editorial, but it would appear from the Augener edition that it is in fact the composer's indication.

*? Bar 73, last quaver, l.h.: The tenuto marking under the chord B-C#-D is omitted. (Given in Augener edition.)

*? Bar 82, quaver 1, l.h.: The staccato dot is marked as editorial, but it is in fact found in the Augener edition, and thus presumably by the composer.

*? Bars 84-85, r.h.: Bar 84 is at the bottom of a page in the Augener edition, bar 85 at the top of the next. The slur beginning in bar 84 extends only to the low tied F, which reflects the Augener edition - sort of. But the slur in that edition doesn't end neatly on the note F like you might expect, but ends high up in the air as if it were to be continued on the next page; but it isn't continued there, which is probably an oversight. It seems to me that the intention was to continue on the next page and end on the high Eb at the end of bar 85. This agrees with the roughly parallel passage in bars 76-77, where the continuous slur is given in both the old and new editions.

*? Bar 108, crotchets 3-5, l.h.: This isn't a correction, but it might be of interest that I think I know the meaning of the single brackets in front of the F#'s, which the footnote says are of obscure meaning, or a typesetting error. It seems to me that they are simply brackets around the sharp signs (with the closing ones omitted) confirming the F#'s are sharp in spite of the simultaneous F naturals in the r.h. The sharp signs are bracketed because they are not strictly needed, because the first sharp is in force for the entire bar; the bracketed sharps are effectively reminders to guard against an unconscious turning of the F#'s to naturals to match the r.h. I often see this use of brackets around accidentals in chromatic music where false relations occur, although usually the closing one is found, also.

"Second Sonata" (Ossianic) (incomplete work)

* Bars 1-3:
      Error: The first system is written with no key signature, and a change to 3 flats is shown at the end. But the 3-flat signature should be present from the start. All the notes are otherwise shown correctly, and even the one accidental present is shown correctly - but of course the absence of the key signature changes the entire harmonic meaning of the passage.
      Reason: Most compellingly, the manuscript page shown in the volume shows the 3 flat signature to be present from the start of the piece. Also, some of the harmonies in these bars make less sense if read without a key signature.

* Bar 2, 3rd crotchet in r.h.:
      Error: There should be two voices here: the upper one consisting of the "Scotch snap" figure in octaves, Eb-C, and the lower one consisting of a crotchet on the Ab-C shown as part of the semiquaver in the upper voice.
      Reason: It is more logical, although to be sure Agnew does not always do this, so this is not in itself a compelling reason. But far more compelling is the fact that the manuscript page reproduced in the volume shows it thus.

* Bar 8, 1st note in l.h.:
      Error: The very low Bb should probably be a B-nat.
      Reason: While Bb would in itself make reasonable sense, B-nat. is better for a couple of reasons, both arising from a comparison with the next bar: B-nat. would make the melodic shape of the l.h. start with an octave, as in the next bar; and it would also make the harmony a dim.-7th chord instead of a dom. min. 9th, and that would agree with the next bar also. Indeed, the whole passage seems to be based largely on chords of the dim. 7th.

[ Sonatas / next volume ]

Volume 1:
 Australian Forest Pieces; Youthful Fancies

Australian Forest Pieces:
 1. Gnome Dance; 2. When Evening Shadows Fall; 3. Forest Nymphs at Play;
4. Night in the Forest; 5. By a Quiet Stream; 6. The Forest Grandeur

Note: In much of the advertising in various volumes, the fourth piece in this set of pieces, "Night in the Forest", is omitted from the listing for the set of pieces. However, the piece itself appears in its proper place in the volume.

           1. Gnome Dance

* p. 1, bottom of page:
      Error: The composer's note, reading "Pedals with discretion", was omitted from new edition.
      Reason: The note simply appeared at the bottom of the page in the original edition, not marked with an asterisk or other sign referring to a particular location in the text.

* p. 1, system 3, bar 3 (first-time bar):
      Error in both editions: Probably the crotchet rest (in both staves) at the end of the bar should be a quaver rest.
      Reason: This mistake occurs also in the old Nicholson edition; I wonder if it is also to be found in the composer's manuscript. At the very start of the piece, before bar 1, there is a D quaver upbeat 2 octaves above middle C; it seems to me that it is intended that it be used in the repeat also. If so, the crotchet rest in the first-time bar should be changed to a quaver rest.
      If the repeat of the upbeat is not intended, then the first-time bar should be left as it stands, but an opening repeat sign inserted at the beginning of the first complete bar of the piece, after the upbeat. I consider this reading to be vanishingly unlikely, because the upbeat note is a proper part of the figure (many later bars in the piece support this statement). But, whichever it is, the note-values don't add up properly as the text now stands in both old and new editions.

* p. 2, system 2, bar 2 - 8va sign:
      Error in original edition: This 8va sign was left out of the original edition. The new edition correctly restores it, with a footnote explaining.
      Reason: The footnote in the new edition explains it. Comparison with the previous bar indicates that the 8va sign should indeed be there.

* p. 2, system 2, bar 3 - beginning of semiquaver run in r.h.:
      Error in original edition: The slur begins on the second note in the original edition, but (correctly) on the first note in the new edition.
      Reason: This editorial correction is probably right: in the original edition, the phrase began one semiquaver later; but the new version seems better. In both editions, other similar passages have the slur starting on the first note. Is it worth checking the manuscript?

* p. 3, system 5, bar 3 - four semiquavers in lower stave:
      Error: Four staccato dots were omitted from the semiquavers in the new edition.
      Reason: Found in the original edition.

           2. When Evening Shadows Fall

* p. 4, system 5, bar 1 - 2nd chord in r.h.:
      Error: Gb should be natural.
      Reason: According to the original edition, and G-nat. seems to fit the harmonic progression better, paralleling the bass in min. 3rds. Having two notes change from one chord to the other sounds good, but having only the bass line change sounds rather weak and lame. A comparison with the parallel passage at p. 5, system 3, bar 4 bears this out, in spite of other discrepancies between these parallel passages which I accept as correct nonetheless.

* p. 5, system 5, bar 1 - first chord in l.h.:
      Error: The lower G should be changed to Eb.
      Reason: Given thus in the original edition, and G doesn't agree with the Gb's in the r.h.

           3. Forest Nymphs at Play

Note: In this piece, there are passages which are repeated almost note for note; when errors occur in corresponding bars in both passages, they are discussed in adjacent entries, or even in the same entry when that is convenient, breaking the normal pattern of going through the bars in order.

* p. 6, system 2, bar 3;
* p. 7, system 1, bar 5 - rests in l.h.:
      Error: The quaver and crotchet rests have been mistakenly exchanged in position, although it makes no practical difference.
      Reason: In 6/8 time it would be usual to have the crotchet rest before the quaver rest, and the original edition does this.

? p. 6, system 1, bar 2;
? p. 6, system 2, bar 3;
? p. 6, system 5, bar 4;
? p. 7, system 1, bar 5 - l.h.:
      Error in both editions: In the new edition, in cases 1 and 3 of the bars just listed, the l.h. holds its chord for a dotted minim; but in cases 2 and 4 the chord is a dotted crotchet, followed by a crotchet rest and quaver rest. But in the original edition, the l.h. note is a dotted minim in all cases except the first.
      Reason: The treatment in the original edition has an irregularity that doesn't seem deliberate. It seems likely that it should have used a dotted minim in all four cases. The configuration used in the new edition seems the next mostly likely to me.
      In the new edition, while the treatment is not uniform, it has a certain regularity - both occurrences of the theme contain the bar twice, and the l.h. note is long the first time in each occurrence of the theme, and short the second time - so this could be quite correct after all, and may be an alternative way of correcting the probable error in the original edition, besides using dotted minims in all cases.
      As to the varied treatment of the bar itself - if both methods are meant to be included in the piece (in some configuration): although certainly bars can be varied when they are repeated in a theme, this particular variation seems pointless, and it's just enough that it would make me check the composer's manuscript to see if it does this. I'm half-persuaded that dotted minims for all four cases might be most probably correct; having such a long l.h. rest in the middle of this phrase seems to me to break it up a bit too much.

* p. 6, system 3, bars 2-3
* p. 6, system 4, bars 1-2
* p. 7, system 2, bars 4-5
* p. 7, system 3, bars 2-3 - top voice in r.h.:
      Error in both editions: This is very similar in structure to the previous entry: the new edition treating cases 1 and 3 one way and cases 2 and 4 another way, and the original edition treating it irregularly.
      Reason:

6/4/1-2, 7/3/2-3, r.h.: These two passages are parallel to 6/3/2-3 and 7/2/4-5, 4 bars earlier in each case. But in these earlier passages, the top B is tied over into the second bar for a second dotted minim, whereas the passages 4 bars later drop the second B and the tie. The Nicholson edition does the same, except that in 7/2/4-5 the tie is omitted, giving two separate notes; this is probably a mistake, which the new edition corrects. But, given the parallel nature of the passages, I wonder if (in both old and new editions) a second B and a tie should be given in the bars identified at the start of this paragraph.

----------------------=========== (unfinished)

* p. 7, system 3, bar 2 - "pp" sign:
      Error in original edition: This is a "p" in the original edition, but the new edition changes it to "pp" - probably correctly.
      Reason: This bar occurs three other times in the piece, and they are all marked "pp" in both editions. There is no reason to suppose this instance is meant to be different in its dynamic marking.

           4. Night in the Forest     -----===** not yet complete **===-----

8/3/4, crotchets 3-4, r.h.: In the Nicholson edition, the Ab is tied to the G#; however, in the new edition, the tie is much higher up, looking merely like a slur applied to the chords as a whole.

           5. By a Quiet Stream     -----===** not yet complete **===-----

9/1/1, 9/3/2, 10/2/2, l.h.: These three bars are exactly parallel. The new edition gives the l.h. as this each ====== time:





Yet, in the second and third instances, the Nicholson edition gives this:





(Both times, the top minim, in the upper voice, is wrongly given a dot also.) On the first instance it gives this:





which is the same except that the dots on the minim chord have been omitted (probably by mistake). This suggests to me that a more correct reading would be to omit the crotchet rest on each occasion, and to dot the minim chord in the lower part; this would also avoid a break in the l.h. chords which rather obtrusively breaks the continuity of the part (assuming the break is not covered up by the pedal, which I would probably do anyway).

9/1/2, 9/3/3, 10/2/3, l.h.: This is slightly similar to the preceding instance, but my claim here is more tenuous. Once again, the minim rest seems to break the l.h. part up too much. On each occasion, the Nicholson edition has no rest, but gives the preceding chord as a minim, not a semibreve, as it should do if there is to be no rest. Quite possibly the intention is "minim chord followed by minim rest", however much it may not seem right to me. But is it worth checking this with the manuscript, or other sources?

9/5/4, 10/1/4, 10/4/2, r.h.: In the Nicholson edition, these bars (all identical) are given thus:





The use of minims near the end of the bar is strange, and part of the value carries over into the next bar by a quaver; but the meaning of this seems reasonably clear. The new edition reduces these minims to dotted crotchets, to keep their value entirely within the bar. If the unorthodox use of minims is considered undesirable, and was eliminated from the new edition for this reason, it would seem to me that the intent would be better reflected by the use of tied quavers, thus:





This would match the earlier minims in each bar better, which also carry over their usual territory by a quaver, and yet were kept in the new edition.

           6. The Forest Grandeur     -----===** not yet complete **===-----

*? p. , system , bar - :
      Error:
      Reason:

11/1/3, 11/3/1, 12/3/3: beginning of each bar, l.h.: In the tenor voice in each of these bars, the original minim shown in the Nicholson edition has been changed to a quaver rest followed by a dotted crotchet; however, whereas most such editorial alterations have been marked as such, there is no such marking here.

11/4/1: The Nicholson edition is obviously in error in giving this passage as:











The note-values don't add up to 4 crotchets (even accepting the final crotchet as a separate voice not to be added to the value of the preceding notes). The correction as given in the new edition is one way of resolving this; but the following seems to me to be more likely:











Perhaps the manuscript itself might resolve it one way or the other; but the rhythm as given in the new edition sounds wrong to me.

12/1/1-2, above treble stave: I think the new edition is probably right here, but perhaps the marking "in time" should appear above the first chord in the next bar. (This is how it is given in the Nicholson edition; but perhaps this is mistaken: above the D# seems more logical.)

12/2/1, minim 2, l.h.: Eb should be E. (In Nicholson edition, and necessary to agree with r.h. and also with other l.h. notes before and after.)

12/4/1, second-last quaver, l.h.: The E at the bottom of the chord immediately following the treble clef needs a natural sign. (In Nicholson edition, and needed to agree with the harmony in that entire bar.)

Youthful Fancies:
 1. The Merry-Go-Round; 2. The Gurgling Brook; 3. The Sunlit Glade; 4. Sleeping Child

           1. The Merry-Go-Round

Note: This volume gives the title simply as "Merry-go-round".

(No errors found.)

           2. The Gurgling Brook

* p. 17, system 5, bar 2 - last half of bar in r.h.:
      Error: The 8 semiquavers should go A-Bb-A-Bb-A-Bb-A-Bb, with no D's included.
      Reason: As given in the original edition; it also fits better with the generally running-down feeling of the end of the piece.

* p. 17, system 5, bars 2-3 - l.h.:
      Error: The tied C's should be D's.
      Reason: Found in the original edition, and necessary because of the D Aeolian tonality of the piece.

           3. The Sunlit Glade

(No errors found.)

           4. Sleeping Child

* p. 20, system 3, bars 4-5; * p. 20, system 4, bars 1-2 - phrasing:
      Error: Two phrases are marked in broken lines, indicating an editorial insertion; but the composer's phrasing was omitted.
      Reason: According to the original edition, the first of these two phrases ends one note later than the editor's phrasing, and the second starts one note later also - but these are not marked in the new edition.

* p. 20, system 4, bar 1 - 2nd note in r.h. upper voice:
      Error: The D quaver should be a C.
      Reason: Found in the original edition, and - although this is a subjective opinion - it seems to sound more correct also.

[ previous / next volume ]

Volume 2: The Complete Poems:
 Two Poems (1922); Poème Tragique (Toccata Tragica); Three Poems (1927)

Two Poems (1922):
 1.; 2. (To the Sunshine)

           1.

* p. 1, system 2, bar 1 - 1st note in l.h.:
      Error: The low A should be an F.
      Reason: In original edition. Compare with other bars in this piece which have a similar l.h. arpeggio: they all start with an octave.

* p. 1, system 3, bar 2 - 2nd chord in r.h.:
      Error: The natural sign in front of the top B should be removed.
      Reason: It makes no sense whatever here; Bb would preserve the tendency towards harmony based on the whole-tone scale, as seen in some of the preceding passages. The presence of the tie suggests that the second note should be the same. Compare with the last bar on page 2, which is similar; it keeps the octave on the corresponding chord. Also, the original edition agrees with the reading I propose here.

* p. 1, system 4, bar 1 - crotchets 2 and 3 in r.h.:
      Error: A slur should extend from Bb to Ab.
      Reason: It is given thus in the original edition.

* p. 1, system 4, bar 3 - lowest note in first r.h. chord:
      Error: The C in this chord should be Cb.
      Reason: Compare with the previous bar, and also with the corresponding bar on p. 3, system 2, bar 1, where the chord is the same as what I propose here, although transposed.

* p. 1, system 4, bar 3 - lower voice in l.h.:
      Error in both editions: The Db tied from the previous bar should have a dot.
      Reason: This is necessary to fill out the value of the bar fully, and no crotchet rest is given to suggest that an undotted minim is the proper value. The dot is omitted in the original edition too, but this clearly seems to be a mistake the new edition simply carried over.

* p. 2, system 3, bars 2-3 - bass line:
      Error: A slur should extend across Bb, C, Db.
      Reason: Found in original edition.

* p. 2, system 4, bar 2:
      Error: A crescendo hairpin should extend the length of the bar.
      Reason: Found in original edition. All other bars with a similar shape have the sign, also, except for one near the end, which is very soft, and therefore not strictly comparable.

* p. 2, system 5, bar 3 - 2nd last quaver in l.h.:
      Error: The E near the end should be Eb.
      Reason: Required harmonically (maintains the augmented triad harmony), and also in the original edition.

* p. 3, system 2, bar 1 - lower voice in l.h.:
      Error: What looks like a B in the lowest voice in the l.h. is really a Bb tied from the previous bar - the tie has not been extended to this system as it should be.
      Reason: Similar to the instance at p. 1, system 4, bar 3, to which it is parallel.

* p. 3, system 2, bar 1 - lower voice in l.h.:
      Error in both editions: The Bb discussed in the previous entry should also have a dot.
      Reason: This is similar to the instance at p. 1, system 4, bar 3, where the dot is also left out, as discussed above. The absence of the dot in both places is not sufficient to persuade me that it is not a mistake. Leaving them out just makes no sense. Perhaps the composer omitted these dots, and printed editions of the piece perpetuated the omission.

* p. 3, system 2, bar 2 - 4th and 5th quavers in r.h.:
      Error: These two notes should be tied together, not slurred.
      Reason: Perhaps the slur line is intended to be a tie. But ties usually touch or nearly touch the actual noteheads, and a similar line on the opposite side of beams suggests slurring, not tying.

* p. 3, system 2, bar 3 - last note in lower voice in r.h.:
      Error: C should be D.
      Reason: The original edition gives D. Compare with corresponding bar, 2nd bar on p. 2, where both editions gives Eb, which is the corresponding note - pitchwise, although it doesn't correspond exactly in enharmonic spelling.

* p. 3, system 5:
      Error: The pedal signs are not found in the original.
      Reason: If they are editorial additions, they are not marked as such, although the edition claims to distinguish editorial additions with typographic devices. While the first marking seems reasonable, the second one makes little sense as far as I can see, especially letting the penultimate B in the bass merge with the final chord.

           2. (To the Sunshine)

Note: In this piece, there are passages which are repeated almost note for note; when errors occur in corresponding bars in both passages, they are discussed in adjacent entries, or even in the same entry when that is convenient, breaking the normal pattern of going through the bars in order.

* p. 4, system 2, bar 1 - l.h. semibreve:
      Error: The chord should have a strong accent above it (like a reversed "V").
      Reason: Found in original edition.

* p. 5, system 1, bars 1 and 2, and p. 7, system 1, bars 2 and 3 - arpeggio signs:
      Error: The arpeggio signs are given differently from the original edition in all four of these bars. While this is flagged as an editorial alteration, it is not clear from the score how the original version appears. A footnote might be the only clear way of doing this.
      Reason: In the new edition, the arpeggio marks are in the l.h. stave, and extend from Ab up to Cb a 10th higher. But in the original edition, the Cb is in the r.h. and the arpeggio marks are in the r.h. stave and extend from the Cb to the Bb at the top.
      It is not the change itself that I am explicitly listing here, because my intention is not to comment on the editorial changes; it is the fact that the composer's version is not clear in the score that I am listing here.

* p. 5, system 1, bar 3 - 4th quaver in r.h.:
      Error: The B-nat. should be Bb. Found in the original edition, and also in the new edition later on in the parallel passage on p. 7, system 1, bar 4.
      Reason:

* p. 6, system 1, bar 3 - 3rd crotchet in r.h.:
      Error: There should be a crotchet rest in the lower voice, to fill out the value of the bar.
      Reason: Found in original edition, and obviously necessary.

* p. 6, system 2, bar 2 - r.h.:
      Error: There should be a C dotted minim in the lower voice in the r.h. (or added to the l.h. chord, following Sitsky's rearrangement of the notes in this passage).
      Reason: Found in the original edition. Compare with the previous two l.h. chords, which include corresponding 9ths.

* p. 6, system 4, bar 2 - 2nd half of bar in l.h.:
      Error: The chord in the lower voice G-C# should be a minim, not a crotchet.
      Reason: Clearly necessary to fill out the value of the bar in that voice. Found in the original edition, and in the new edition in other similar bars, including the very next bar, which is a repetition of this one.

* p. 6, system 5, bar 3 - lowest notes in r.h. throughout the bar:
      Error in both editions: The B's should be Bb.
      Reason: The original edition omits the flat sign at the start of the bar, thus rendering all the Bb's as B's; but it is clearly a mistake. B-nat. doesn't make harmonic sense here, and doesn't fit in with the overall style of harmony based partly on the whole-tone scale. Also compare with bar 10 (the bar at the bottom left of p. 4), which uses Bb. There are also similar passages pitched a per. 5th higher where the note F appears, which corresponds exactly to Bb.

* p. 7, system 1, bar 2 - first chord in l.h.:
      Error in both editions: The A in the chord lacks an accidental, thus reading as A-nat.; but clearly it should be Ab. The new edition perpetuates the error in the original edition.
      Reason: Comparison with the following bar, and with the two bars earlier in the piece that correspond to these two bars (p. 5, system 1, bars 1 and 2), makes Ab clearly correct. A-nat. would not make sense harmonically.

* p. 7, system 5, bar 1 - r.h. at beginning:
      Error: There should be 4 little tie marks in mid-air, continuing from the end of the previous system.
      Reason: It is usual, when a tie extends past the end of a system, to show it at the start of the new one, even if the ties end in mid-air. The original edition does so at this point.

* p. 7, system 5, bar 1 - dotted minim in l.h.:
      Error: The sharp sign should be removed.
      Reason: The sharp sign is wrong because it is tied from a G-nat. in the previous bar. Even if the tie were read as a slur, G# makes no harmonic sense here, doesn't agree with the parallel passage at p. 5, system 4, bar 2, and is not found in the original edition.

* p. 7, system 5, bar 2 - below l.h. stave:
      Error: Instruction "Hold pedal to the end." is missing. It should appear underneath the G dotted minim.
      Reason: Found in original edition.

* p. 7, system 5, bars 2-4 - on the two actual bar-lines:
      Error: On the two actual bar-lines separating the last three bars, on the lower stave, there should be four little ties hanging in mid-air - one set over each of the two bar-lines - representing tied-over harmony whose notes are not actually printed in these bars.
      Reason: Found in the original edition.

* p. 7, system 5, bar 3 - l.h.:
      Error: The semibreve rest shouldn't be there.
      Reason: It's not in the original edition, presumably to represent the fact that the pedal is still holding previously sounded harmony.

* p. 7, system 5, bar 4 - beginning of bar :
      Error: The grace notes should strictly speaking be full-sized notes. A slur should extend from the first of these notes to the final top semibreve taken by the l.h. crossing over.
      Reason: Found thus in the original edition, although this doesn't make for a bar which adds up properly - which is probably why the new edition changed them to grace notes, which theoretically take up no time.

Poème Tragique (Toccata Tragica)      -----===** not yet complete **===-----

Note: In this piece, there are passages which are repeated almost note for note; when errors occur in corresponding bars in both passages, they are discussed in adjacent entries, or even in the same entry when that is convenient, breaking the normal pattern of going through the bars in order.

Note: This piece was originally published in two separate editions which have differences between them, some of which are noted in The Keys Press edition. I have only the Allan's edition, printed in Melbourne and entitled Toccata Tragica, but not the Curwen edition published in London under the title Poème Tragique. The Allan's edition seems mostly free of mistakes, as far as I can tell, and I have used it as a standard of comparison with the Keys edition. But because I am aware that there are two early editions with differences between them, and because I have seen only one of these, I will also rely on my own musical judgement and familiarity with Agnew's style to evaluate possible errors in the new edition.

*? p. 8, system 1, bar 2 - 4th semiquaver note in r.h.:
      Error: The semiquaver A should be a B.
      Reason: Found in the original Allan's edition. Also, compare with the next occurence of the figure: B would make both of them end with a min. 3rd.

*? p. 8, system 3, bar 2 - 2nd and 3rd quavers in l.h.:
      Error: The two D's should be tied together.
      Reason: They are tied in the Allan's edition, and comparison with other similar bars suggests that the notes are intended to be tied.

*? p. 8, system 4, bar 2 - 2nd and 3rd quavers in l.h.:
      Error: The two E's should be tied together.
      Reason: This tie is omitted in the Allan's edition, but the context suggests that the tie should be included.

*? p. 9, system 4 - several instances in both staves of the last semiquaver of a group and the crotchet following:
      Error: These two notes should be tied together in each instance.
      Reason: Found in the Allan's edition, and the context seems to suggest that the notes should be tied. In a couple of case, the ties are quite visible. In each of the other cases on this system, a tiny little mark like a dot appears between the two notes in question, in a position where a dot would have no meaning. This suggests that the dot may in fact be a poorly typeset tie, done so short that it looks like a dot.

*? p. , system , bar - :
      Error:
      Reason:

*? p. , system , bar - :
      Error:
      Reason:

Three Poems (1927):
 1.; 2.; 3. [G#-minor-ish]

           1.

* p. 21, system 3, bar 2 - crescendo sign at end of bar:
      Error: It should appear above the r.h. stave, thus applying only to the r.h. notes.
      Reason: In the original edition.

* p. 22, system 3, bar 1 - quavers in l.h.:
      Error: They should be beamed into two groups of 3 and 4 notes respectively.
      Reason: In the original edition.

* p. 22, system 3, bar 3 - between the staves:
      Error: Two hairpins are missing: a crescendo beginning with the first r.h. C and finishing at the first Ab; and a diminuendo beginning just after the second Ab and ending with the final C.
      Reason: In the original edition.

* p. 22, system 4, bar 2 - between the staves:
      Error: There should be a "pp" just before the G in the r.h.
      Reason: In the original edition.

* p. 22, system 5, bar 3 - "ppp" sign:
      Error: It should be above the r.h. stave, thus applying to the r.h. only.
      Reason: Found thus in the original edition - in fact, a little way into the bar, not directly above the first notes of the bar. Admittedly, it makes little practical difference, because the l.h. is the only active part beyond the first note of the bar.

           2.

Note: In this piece, there are passages which are repeated almost note for note; when errors occur in corresponding bars in both passages, they are discussed in adjacent entries, or even in the same entry when that is convenient, breaking the normal pattern of going through the bars in order.

Some of the possible errors in this piece are not inherently obviously wrong, but result from comparison of different passages in the piece which are identical other than the purported errors. Accordingly, I will group together corresponding bars of the piece, and discuss them together, without specifying which bars might actually be wrong, if any are.

* p. 23, system 1, bar 1;
? p. 24, system 2, bar 2;
? p. 25, system 2, bar 3 - last note in upper voice of r.h. in each bar:
      Error and discussion: Each occurrence of this bar is transposed a maj. 3rd lower than the previous one, so in discussing this passage, allowance must be made for that. But once that allowance is made, the three bars are exact replicas of each other, except for the note named above. This note is E-nat., B, and G in those three bars. But uniformity would require either Eb, B, and G, or else E-nat., C, and Ab. In other words, either the first instance is wrong, or both the later ones are. Since either note would be about equally suitable harmonically and melodically, we can't use that to decide the matter with any certainty. If we assume the notes are meant to be uniform (which seems a reasonable possibility), then presumably a majority vote would be the most probable resolution of this - that is, the first instance is the incorrect one.
      It might be that we shouldn't assume that the notes in question should necessarily be uniform, and presumably the composer's manuscript (which I don't have access to) would shed light on this. But the original edition (which I do have) supports uniformity, and indicates that the first instance is the one that's out of step: that is, E-nat. should be Eb. In the other two instances, both editions agree, which would strengthen the versions given in those places.

? p. 23, system 1, bar 2;
* p. 24, system 2, bar 3;
? p. 25, system 3, bar 1 - second note in upper voice of r.h. in each bar:
      Error in both editions and discussion: This is a very similar case to the previous entry, so we can compare the three instances of this passage in the same way. This indicates that the 2nd case is the odd one out - and this time, the original edition contains the error too. Two editions giving the same reading indicates that maybe the composer's manuscript contains it too - which need not preclude the possibility that the composer made a slip of the pen. That would be very easy for a composer to do in music as chromatic as this.
      Going through the comparison shows that in the second case the G should be F, and that would be F# because it would then be tied to the preceding note (the slur becoming a tie). (If you go the other way, then the first instance should be B-nat. (strictly speaking Cb, enharmonically - but enharmonics are treated with some freedom in this piece, for ease of reading), and the third instance would be D# or Eb - but I consider this reading very unlikely: not only is it outvoted, but the harmony that results doesn't seem to fit into the general harmonic style, which uses min. 7ths far more often than maj. 7ths which result from this other interpretation. Once again, a possibility is that the cases are meant to be different in that note - but that also seems rather unlikely.)
      There's one more factor that strengthens the reading I've opted for. Everything falls into place if, in the second case, you assume that the slur leading into the note is really a tie, and that the note has simply been positioned a degree too high by the typesetter of the original edition, and that the typesetter of the new one simply copied that, assuming it to be correct.

? p. 23, system 2, bar 2;
* p. 24, system 3, bar 3 - first note in lower voice of r.h. in each bar:
      Error and discussion: Once again, this is similar to both the preceding entries, although this time the bar occurs only twice, shifted down a maj. 3rd the second time.
      With only two occurrences of the bar, neither reading can outvote the other - but this time the original edition maintains a consistent reading for both bars, and can give a casting vote.
      The note is given in the new edition as C the first time and G the second time. Consistent readings would be either Cb and G, or else C and G#, and it is the latter which the original edition supports. In the absence of further information (such as the composer's manuscript), I would accept this as correct, although either reading would make equal sense in context - which is part of what makes deciding the matter more uncertain. But reading it this way makes the new edition correct in the first instance of this passage, and wrong in the second.

* p. 24, system 5, bar 1 - "f" sign:
      Error: It should appear above the r.h. stave, in front of the A quaver, not between the staves.
      Reason: Written thus in the original edition.

* p. 25, system 1, bar 2 - 2nd last quaver in l.h.:
      Error in original edition: The new edition gives this B a natural sign, and the original one omits it, thus making it a Bb because of the influence of an earlier flat sign. I regard the new edition as correct here.
      Reason: B gives a better melodic shape (admittedly a rather subjective perception), and makes better harmonic sense (suggests dom.-7th harmony on Bb combined with min. 9th and min. 13th). Also, the notes surrounding the B form a pattern that appears earlier an octave higher, and both editions agree in rendering it as B-nat.

           3. [G#-minor-ish]

* p. 26, system 2, bar 1 - grace note in r.h. in the middle of the bar:
      Error: The grace note needs a sharp sign.
      Reason: Fits better harmonically; found in original edition too.

* p. 26, system 5, bar 3 - Ab in l.h.:
      Error: It needs a staccato dot.
      Reason: In original edition; compare with the preceding and following bars.

* p. 27, system 1, bar 3 - quaver rest below l.h. stave near end of bar:
      Error in both editions: The rest should be under the last chord in the l.h., not before it. The new edition obviously carried this over from the old one.
      Reason: A matter of proper alignment.

* p. 27, system 2, bar 3 - 3rd semiquaver in r.h.:
      Error: The C# needs a G added to it, making a double note.
      Reason: Found in the original edition. Comparing this figure with other similar figures in surrounding bars shows that a double note is needed here. On the face of it, I would have considered G# a possibility for the extra note - but the original edition's G-nat. seems to settle it reasonably.

* p. 27, system 3, bar 3 - "pp" sign:
      Error: Should be above the r.h. stave, not between the two staves.
      Reason: In original edition.

* p. 28, system 4, bar 3 - lower r.h. voice:
      Error in original edition: The original edition gives each group of 3 notes here as demisemiquaver, double-dotted quaver, quaver. This is probably incorrect, and the new edition renders it correctly as demisemiquaver, dotted semiquaver, quaver.
      Reason: The way the original edition renders it adds up to 3 crotchets, not 2, and thus it won't fit in the bar.
      Another interpretation is possible, although very unlikely - and the notation for that interpretation is still not as clear as it might be. If the quavers were considered to belong to a different voice, the remaining notes would add up, and would be similar to the figure in this voice in the immediately preceding bars. The quavers would then need to alternate with quaver rests to complete that voice. But harmonically and texturally this reading would seem to be improbable; and comparison with the passage two bars later in the original edition, which is parallel (in the r.h. at least), rules this out, because there the passage is given the simpler way, and agrees with the new edition.

* p. 28, system 4, bar 3 - last group of semiquavers in l.h.:
      Error: The slur should cover all three notes, not just the first two.
      Reason: Shown in the original edition, and would fit in with other similar figuration.

* p. 29, system 3, bar 3 - between staves:
      Error: "retard" (or "rit") should appear, beginning beneath the second semiquaver in the r.h.
      Reason: Found in the original edition.

* p. 29, system 4, bar 3 - "pp" marking:
      Error: Should appear above the r.h. stave.
      Reason: In original edition.

* p. 29, system 5, bar 3 - last semiquaver in l.h.:
      Error: The note should be tied to the following crotchet. Also, the natural sign should be removed, making it G#.
      Reason: Both these corrections are in the original edition. G-nat. makes absolutely no harmonic sense whatever - even allowing for the fact that Agnew sometimes ends pieces with unusual chords. G-nat. would also destroy the shape of the motif it appears in, which is a very strong motif throughotu the whole piece, and it almost always moves in 2nds, very rarely including 3rds. Because of the ties, this means the piece would end with G in the bass, which completely goes against the strong G#-minor tonality of the end of the piece. And G makes no sense in the chord itself, which includes a G# in the r.h. The whole chord looks clearly intended to be a G#-minor triad with an added maj. 2nd - and Agnew several times ends a piece with a triad plus an added 2nd.

[ previous / next volume ]

Volume 3: The Complete Preludes:
 Four Preludes (1925); Three Preludes (1927)

Four Preludes (1925):
 1.; 2. [G# minor]; 3. (The Wind); 4.

           1.

Note: In this piece, there are passages which are repeated almost note for note; when errors occur in corresponding bars in both passages, they are discussed in adjacent entries, or even in the same entry when that is convenient, breaking the normal pattern of going through the bars in order.

? p. 1, system 1, bar 2, and p. 2, system 4, bar 1 - arpeggio sign for last chord:
      Error: In the first location, the arpeggio sign added by the editor covers the l.h. only; but in the second location, it starts with the l.h.'s top note and goes up to the note in the r.h.
      Reason: There is no reason to suggest that this difference is intentional, and that it is not a typesetting error.

? p. 1, system 2, bar 3, and p. 3, system 1, bar 2 - final note in r.h.:
      Error:Possibly the final F# is really intended to be F-nat., and the natural sign that would cancel the earlier sharp sign omitted by oversight.
      Reason: F# doesn't seem to fit in with the underlying Db-major harmony. Admittedly this is uncertain, because Agnew often does use all manner of conflicting harmonies, false relations, and the like. And my claim is weakened further by the fact that the same thing occurs in two separate places. But it seems strange enough to me to be worth checking against the original sources.

           2. [G# minor]

* p. 4, system 3, bar 1 - r.h., first two chords;
* p. 6, system 1, bar 1 - l.h. chord;
* p. 6, system 5, bar 2 - r.h. chord;
* p. 7, system 1, bar 2 - r.h., first two chords:
       Error: In each chord just listed, there are notes which are enclosed in editorial brackets. But it is not clear which note is intended to be flagged as an editorial addition, or how many notes are affected.
      Reason: The notes in each case are a 2nd apart, and it is difficult in such a situation to select one of them with brackets when they are a 2nd apart. Perhaps footnotes in such cases would clarify what notes were added, and what ones written by the composer.

* p. 6, system 2, bar 2 - C# octave in bass:
      Error: The octave dotted minim C#-C# is aligned wrongly with the other notes.
      Reason: The octave cannot be aligned precisely without its stem merging with the downward-pointing stem of the higher chord in the l.h. stave; but it should be shift slightly to the right - not to the left.

* p. 6, system 3, bar 1 - C# octave in bass:
      Error: This octave is tied from the previous bar, but the ties leading into it are missing.
      Reason: Not continuing the tie begun on the previous system is ambiguous.

* p. 7, system 4, bar 1 - semibreve in l.h.:
      Error: All five notes are tied from the previous system, but two of the ties are not continued into this bar. Ties should be added for G-nat. and C#.
      Reason: Ambiguous without this being done.

           3. (The Wind)

? p. 8, system 3, bars 1 and 2 - A-minor triad in each bar:
      Error: Possibly the A-minor triads should be A-major ones.
      Reason: This would continue the parallel harmony fromm the surrounding passages. Obviously much of the harmony in this piece is based on parallel harmony. Compare with the parallel passage on p. 9, system 5, which, although transposed, continues the parallel harmony in a similar way.

* p. 9, system 3, bar 2 - 3rd-last quaver in l.h.:
      Error: The lower G in the octave needs a natural sign.
      Reason: Clearly it is supposed to be a per. 8ve, not a dim. 8ve. But the G# from the beginning of the bar is still in force. Compare with the 2nd bar in the piece, which is parallel, although a semitone lower.

* p. 9, system 5, bar 1 - 3rd last note in r.h.:
      Error: A natural sign is wrongly inserted.
      Reason: A-nat. doesn't make any sense here. The entire passage is parallel dom.-9th harmony, which would require Ab. Compare with the next bar, which is identical. Also compare with p. 8, system 3, which is the same in this matter (although transposed) (although it does have a discrepancy of its own, as discussed earlier).

* p. 11, system 2, bar 1 - 3rd quaver:
      Error: The A should be A#.
      Reason: The entire r.h. passage is based on augmented triads, and A-nat. doesn't fit. Compare with the passage two bars later, which is the same, although transposed up a min. 3rd - it maintains the augmented triads throughout.

* p. 11, system 4, bar 3 - very low double-note in l.h.:
      Error: The notes are given as A-B, but probably should be B-B.
      Reason: The overall context indicates that this bass part is generally in octaves - a maj. 9th doesn't make sense.
      There may be a question about whether A or B is the correct reading. Considering that most of the harmonies are constructed as dom. 7ths and min. 9ths with extra notes added rather in the manner of Scriabin, that suggests B here far more strongly than A. A would parallel the A in the higher chord, and the overall effect would be weaker, as well as making little harmonic sense.

* p. 12, system 1, bar 1 - 2nd bass note:
      Error: A dot should be added to the minim.
      Reason: Necessary to complete the value of the bar. There is nothing to suggest this note should be treated any differently from the preceding bass notes, all of which fill out their respective bars entirely.

* p. 12, system 1, bar 1 - r.h. 8va sign:
      Error: The 8va sign is too long, and covers three more notes than it should.
      Reason: As written, the descending shape of the arpeggio is destroyed. Finishing the sign at the end of the first group of demisemiquavers restores the pattern.

* p. 14, system 3, at start:
      Error: The brace at the front of the three staves is missing.
      Reason: That there are 3 staves in this system does not seem to suggest any reason for omitting the customary brace, although I have very occasionally seen scores that omit it for 3-stave systems only. (I think, from memory, that Ives' "Concord" Sonata does this. The Agnew edition doesn't though, in any other locations.)

           4.

* p. 15, system 4, bar 2 - last chord in r.h.:
      Error: The D should have a natural sign in front of it, cancelling the preceding D#.
      Reason: This would continue the pattern of parallel dom.-7th chords. The original edition shows it thus.

* p. 16, system 3, bar 4 - 3rd quaver in l.h.:
      Error: This note should be Ab, not A-nat.
      Reason: This would fit in with the general style based on dom.-7th chords, and would agree with the Ab in the r.h. Shown as Ab in the original edition.

* p. 16, system 4, bar 2 - 3rd quaver in r.h.:
      Error: The "+" flagging an editorial alteration should not be here.
      Reason: The arpeggio symbol it is above is in the original edition too.

* p. 16, system 5, bar 5 - pedal sign:
      Error: The pedal sign is too long.
      Reason: In the original edition, it ends underneath the final rest, not at the end of the bar.

Three Preludes (1927):
 1. [A major]; 2. [Eb minor]; 3. [B major]

           1. [A major]

* p. 17, system 2, bar 5 - first l.h. note, F#:
      Error: There should be an extra voice added simultaneous with this F#: an F# dotted minim, as well as the F# quaver belonging to the voice already shown.
      Reason: Found in original edition.

* p. 17, system 5, bar 2 - l.h.:
      Error: The slur from the A to the B in the upper l.h. voice should not be there, or else should be dotted, indicated an editorial addition.
      Reason: It is not found in the original edition.

           2. [Eb minor]