(M.J.E. / Composer Listings / Scriabin / Orchestral work errors)
List of possible errors in
orchestral works by Alexander Scriabin
Compiled by Michael Edwards
Scriabin is one of my very favourite composers, and his orchestral works
must count amongst my favourite orchestral works of all time, and I have the
scores for all of them, and have at various times studied them quite closely.
As a result, I became very familiar with Scriabin's orchestral and harmonic
style. At various times, therefore, I was able to see that certain things which
I saw in the scores were probably misprints, because they did not fit in with
Scriabin's meticulously detailed and consistent style.
Although I realize publication of a new edition of these works is unlikely,
there could possibly be reprints of the already-existing editions; and I suppose
there is at least the possibility that corrections could be incorporated,
especially if they are very small ones, such as the misprints I've found.
Accordingly, I thought I would compile a list of them, just in case the
publishers of editions want to make these corrections in future reprints, and
just in case by some fluke this list comes into their hands. (Once the list is
complete, I can always try contacting people who might know what to do with it.)
That is what this page is here for: it is a list of the errors I believe
I've found in some of these scores, together with my reasons for feeling sure
that they are indeed mistakes. There are not many mistakes, and I feel the
scores are pretty accurate, on the whole - but not perfect. The few mistakes I
found are very small ones, and a study of the context usually makes it clear
what is intended.
I think most of the items are quite clear; however, the list is designed to
be read in conjunction with the score itself, and will be meaningless to anyone
who doesn't have a copy of the score.
In the following list, I've marked each item with either an asterisk or a
question mark: the asterisk marks items I feel fairly sure 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 at more closely by an expert to decide.
In referring to musical notes, I do not have musical symbols available for
use in this list. (Possibly they do exist in a form that is usable in H.T.M.L.
code (the "language" web pages are written in) - but until the ridiculous lack
of standards in H.T.M.L. is sorted out, I am determined to keep my web-site
files as simple and standard as possible, as I don't want to render my web pages
unreadable to some readers because I've used techniques their browsers can't
interpret correctly.) 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
Please note that the scores do not come with bar-numbers indicated. For
ease of referring to particular bars, I identify them by counting bars
separately for each system on a page, and where necessary systems are counted
for each page. So I will refer, for example, to page 3, system 2, bar 4 - and
E-mail me about this music.
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.
Symphony no. 3 in C minor, op. 43 - Le Divin Poème
(The Divine Poem)
* p. 3, beginning of system - instrument listing:
Error: It is highly desirable, if not completely essential, for an
orchestral score to clearly indicate the exact instruments needed to perform the
score. There are two main ways in which composers do this:
(a) The first page of the score includes a staff for every instrument
included at least in the movement, if not the entire work, regardless of whether
or not it has anything to play on that first page.
(b) A page appears before the first page of score which actually lists the
make-up of the orchestra required.
Some scores might even do both of these; but, in general, a score that does
(a) will be likely not to include a list of instruments; and a score that does
(b) will quite likely list on the first page only those instruments that play on
the first page.
The Divine Poem clearly uses method (a) - but does so incompletely.
The Percussion instruments used in this work are not included on the first page
of the score, or on the first page of any of the movements.
Reason: The problem is that it is not possible to tell from the
front of this score exactly what instruments are required. Unless a conductor
happens to know the precise instrumentation for this work, he or she will have
to page right through the score, making note of any instruments that occur which
are not included on the first page. (It is the less-used Percussion instruments
or "extras" that are most likely to be omitted from the first page of a
Either the missing instruments (the Percussion) should be included on the
first page, or else a complete listing of the orchestra should precede the first
page of score.
The Percussion instruments omitted are: Tamtam; Campanelli. The Campanelli
(Glockenspiel) are not used in the first movement - only the second - but the
second movement does not list its Percussion on its first page, either. The
Glockenspiel should be listed either on the first page of the entire score, or
the first page of any movement in which it appears.
(At least one recording of this work also includes cymbals, but this is not
indicated in the score, and would seem to be the addition of the conductor of
the recording, Yevgeny Svetlanov. At the risk of provoking purists, I must say
the cymbals do add a bit of punch at certain climaxes, and I feel this improves
the overall effect. The climaxes in other recordings without cymbals do sound a
bit flat by comparison. Perhaps this merely reflects the fact that one tends to
continue to prefer the performance of a work one gets to know first, and it
continues to sound "right", even if you find out later that it is wrong in some
* p. 3, Violoncelli, Contrabbassi, bar 2 - last note:
Error: The last note, a minim, is positioned wrongly within the bar:
It is placed vertically with the last crotchet of the bar, instead of the
second-last one, as it should be.
Reason: It's a simple matter of lining up notes vertically that
* p. 43, bar 3 - Arpa II:
Error: In the Harp glissando, the E should have a natural sign in
front of it.
Reason: Because of the key signature, it reads as Eb, but this
doesn't fit in with the harmony at that point.
The glissando is partly written out in notes. It is usual in this style of
notation to give a complete octave of notes in order to show the exact settings
for the Harp pedals; it is especially important in scores such as this one which
do not directly show pedal settings for the Harp. However, no C is included in
the fully-notated portion of the glissando. It is reasonable, from the harmonic
context, to suppose that the C's in the glissando should be natural. Neither Cb
nor C# would agree with the chord - C-nat. appears nowhere in the chord
elsewhere in the orchestra, but C-nat. in the Harp would fit best, giving a
whole-tone flavour which would fit the harmony very well. The Harp passage four
bars later confirms this, because the chord is the same, and this time the
glissando does include a C-nat.
* p. 43, bar 5 - p. 44, bar 1; Oboi, 1., 2.:
Error: The quavers for Oboes 1 and 2 at the end of the first-named
bar and the beginning of the second-named one should be joined together,
crossing the bar-line. (It continues over the page, but continuous beaming for
quavers can be indicated over a page-turn by extending the beams on each page a
little beyond the notes themselves in the desired direction, it being presumed
from this notation that the beams are joined.)
Reason: All the surrounding Oboe figures are so joined in groups of
4, sometimes across bar-lines, and there is no reason to suppose the pattern
should not continue here as well.
? p. 47, bar 6; p. 48, bars 2, 3 - Contrafagotto:
Error: In each of these three locations, the last beat of the bar
consists of the following: quaver rest, dotted semiquaver rest, low D
demisemiquaver. Possibly the D should be preceded by a C demisemiquaver, and
the dot removed from the semiquaver rest.
Reason: The part the Contrabassoon is playing is doubled either at
the unison or the higher octave by the 3rd Bassoon and the Double Basses, which
parts include this C, which is clearly an integral part of the recognizable
motif which appears at this point. It might seem that Scriabin deliberately
left out the Contrabassoon on this note, preferring not to have the instrument
enter on such a low note. However, the note it does enter on is only a
whole-tone higher (D).
I don't claim this is a definite error, and accept that it is probably
intentional - but, if I were revising the edition, I'd at least want to look
into this further, and see what the manuscript says, and perhaps get the opinion
of a contrabassoonist on this.
* p. 53, bars 3-4 - Viole:
Error: A triplet sign is missing on the triplets in these two
Reason: When triplets begin after preceding notes of normal value,
it is usual to mark at least the first two or three with a triplet sign.
* p. 53, bar 4 - Trombe in B, 1., 2.:
Error: Two Trumpets share this staff. The second and third notes
are in octaves, obviouslly taken by the two instruments; but the first note is a
single quaver, and it is not indicated whether it is played by 1., 2., or
Reason: It is reasonable from the context to suppose that both
Trumpets will play the unison quaver; but, for absolute clarity, it should
nevertheless be indicated.
* p. 54, bar 2 - Viole, Violoncelli, Contrabbassi:
Error: The bar in these parts is fully occupied by a dotted minim;
this note should be an undotted minim, followed by a crotchet rest.
Reason: Letting the notes in these parts continue into the third
beat clashes harmonically with what the rest of the orchestra begins on this
beat. It is the beginning of a new idea, and the harmony is completely
different. To suppose that a momentary bitonal effect is intended here is
completely contrary to the harmonic style this work uses.
? p. 55, bar 5 - Violini II:
Error: The last note is indicated as a crotchet. Possibly it should
be notated as a quaver followed by a quaver rest.
Reason: The note in question is pizzicato, and all the preceding
pizzicato notes are notated as quavers followed by quaver rests. True, this
doesn't prove that this is a misprint - Scriabin might have notated it as a
crotchet. But if you look two pages back, at a very similar passage, the last
note here is notated as a quaver followed by a quaver rest. It should
perhaps at least be compared with the composer's manuscript, or with other
editions of the work.
* p. 62, system 1, bar 6; system 2, bar 2 - Fagotti, 1, Violoncelli,
Error: This bar consists of four crotchets in each of these parts.
Each such group of crotchets should be accompanied by a quadruplet sign.
Reason: Irregular rhythmic groups should be marked with the
appropriate notation when they first appear in any passage. Or this bar, for
these instruments, could change to 4/4 for one bar, and then back to 3/4,
although this would seem slightly less suitable for a quadruplet that appears
only for a single bar at a time.
* p. 64, bar 6 - Violoncelli:
Error: The two crotchet rests which seem to indicate a rest in one
of several parts should be removed.
Reason: The Cellos at this point are divided into three parts: the
first part is resting for the entire bar, as indicate by the whole-bar rest
above the staff; at the beginning of the bar, for the first two beats, the lower
two parts play minims on two different notes, which are stemmed together; on the
third beat, the upper of these two parts rests, as indicated by the crotchet
rest on the third beat, while the lowest part plays a crotchet note. This
accounts for all three parts in the entire bar - so the two crotchet rests
appearing on the first two beats of the bar are superfluous, and indeed
confusing. (Note that the crotchet rest on the third beat is
necessary: the upper note on the minim (the middle of the three Cello parts)
continues to it, and without it, it would not be absolutely clear what the
middle Cello part should do on the third beat.)
* p. 67, bar 3 - Viole:
Error: This bar, in this staff, should have a quadruplet sign over
the upper of the two divisi parts.
Reason: Necessary because the Viola part in this bar consists of 4
crotchets. In surrounding bars, other staves change to 4/4 for one or more
bars, then back again; but this is not suitable in this case because of the
dotted minim in the other part on the same staff.
* p. 67, bar 3 - Viole:
Error: In the upper of the two divisi parts, consisting of a
quadruplet of crotchets, the last note, Bb (with a superfluous flat sign in
front of it) should in fact be Bbb.
Reason: Compare with Cor Anglais and the 3rd Horn. Allowing for
transposition, they play concert Bbb here. It is obvious that the notes in
question are intended to go down the chromatic scale.
* p. 70: bar 5 - Fagotti, 1.; bar 10 - Fagotti, 2., 3.; bar 10 - Violoncelli
Error: In all these locations, in these staves only, the time
signature 4/4 should appear.
Reason: The passage is one in which some instruments, for a few bars
at a time, play four notes against the three crotchets the rest of the orchestra
is playing. Scriabin has chosen not to indicate these with quadruplet signs,
but with a 4/4 time signature instead; therefore this time-signature change
should appear at the start of all such passages.
* p. 71, bar 3 - Fagotti (both staves), Violoncelli (both staves), Contrabbassi:
Error: On each of these staves, the time signature 3/4 should be
inserted at the beginning of the bar.
Reason: See the previous item for a description of the rhythmic
situation immediately preceding this location. After some instruments have
played a few bars in 4/4, they now revert to the standard 3/4 for this movement,
and the 3/4 signature should be inserted to indicate this.
* p. 74, bar 5 - Violoncelli, last note in bar:
Error: The note should be G, not A.
Reason: The passage is an easily-recognizable thematic motif, and
only G is correct here.
* p. 78, bar 6 - Violini I, upper staff:
Error: The two triplet groups should have a triplet marking.
Reason: As before: when triplets first appear in a passage, they
should be indicated two or three times.
? p. 109, bar 1 - Flauti, 3.; p. 110, bar 5 - Flauti, 2.:
Error: Possibly the first crotchet in the bar should have a trill
sign in the respective Flutes.
Reason: A recording I have has a trill in the 3rd Flute in the first
place. That doesn't prove it's correct, but it's enough circumstantial evidence
to me that it should at least be looked into.
The situation is a little less clear in the second location. I remember a
trill at this point, too, but the 1st Flute is also trilling at this point (and
is marked doing so in the score), and on the same note, and therefore it would
be difficult to tell whether the trill I recall is only in the 1st Flute part,
not the 2nd Flute. However, a trill in the 2nd Flute is suggested by analogy
with the earlier passage just cited, involving the 3rd Flute.
? p. 112, bar 4, p. 113, bars 1-2 - Arpa I, II:
Error: The Harp parts include several Bbb's in these bars, which is,
strictly speaking, incorrect notation, because Harps can tune notes only sharp,
flat, or natural - not double-sharp or double-flat.
Reason: Scriabin does not appear in this score to use fully
enharmonic notation in the conventional manner used for Harp writing. It might
be a bit too strong to state that non-enharmonic notation is wrong; but
at least it is less convenient from a Harpist's point of view, and leads to the
danger of a composer writing something that is unplayable on the instrument.
The Bbb's must be played as A-nats., and the most confusing aspect of this
passage is the fact that an Ab appears in the bar following the bars cited here,
thus necessitating a quick pedal change that is not directly indicated (this
score never giving pedal settings), and is not immediately obvious from the
* p. 122, right margin of the page:
Error: All staves on the page are barred together at the very ends
of the staves.
Reason: The page contains two systems, so the bar-line at the right
margin should be broken where they wrongly join the two systems.
* p. 143, bar 1 - Viole (top staff):
Error: The trill sign is extended too far: it stops at the end of
the bar, although there is a crotchet rest underneath it.
Reason: Clearly, the trill sign should extend only up to the point
immediately before the crotchet rest.
? p. 144, bar 2 - Arpa II:
Error: Should the Harp chord be an octave lower? (Very doubtful,
Reason: This is based on a recording I have where it sounds as if
this chord is an octave lower. However, it is as least as likely that the
player on that recording was in error as that my edition of the score is
* p. 159, bar 2 - Violoncelli:
Error: The second note is positioned too early within the bar,
although its rhythm is correctly notated other than that.
Reason: The note must be moved a quaver forwards to be positioned
vertically with notes in other instruments it sounds simultaneously
Jeu Divin (Divine Play)
? p. 175, bar 5 - Horns 5 - 8 (lower 2 of four Horn staves):
Error: Horns 5 and 6 are playing mf, but Horns 7 and 8 are
Reason: This is not necessarily wrong, and it could be done for
reasons of balance, or to get a particular effect. However, it is sufficiently
unusual in this score as to at least make me wonder if it might be a misprint.
Perhaps I'm showing my ignorance of the subtleties of orchestration, but I see
nothing about this passage that seems to indicate that this would be
* p. 193, bar 3 - Violini I, first staff:
Error: "pochiss. cresc." should be written beneath the staff in this
Reason: The marking appears in all other instruments of the
orchestra which are playing in that bar, and there is no reason to think that
only part of the 1st Violins would behave differently.
* p. 201, bar 3 - Arpa II:
Error: The time signature 3/2 should appear in the 2nd Harp part at
the beginning of the bar.
Reason: The rest of the orchestra changed to 3/2 some bars earlier,
but the 2nd Harp continued to be notated in 3/4; but at this point it is
unmistakeably playing in 3/2, with 6 crotchets per bar.
Le Poème de l'Extase (The Poem of Ecstasy), op. 54
See the Note at the bottom of this page concerning this
[No errors found yet.]
Prometheus: The Poem of Fire, op. 60
See the Note at the bottom of this page concerning this
work. Bar numbering is provided in this score, at the top left corner of each
system, and I will use the numbering given, rather than counting afresh for each
* p. 3, beginning of system - instrument listing:
Error: This score is not prefaced with a list of instruments
required, and relies on its first page of full score to indicate the instruments
required. (See the note for the beginning of
The Divine Poem for more explanation
about the methods scores use to indicate the required instrumentation.)
However, in Prometheus, not quite everything is actually included on the
first page of score.
Reason: The basic problem is that this score fails to indicate
clearly exactly what instruments are needed to perform it. Some of the many
"extras" required in this work are omitted from the first page: namely, Camp.
(Campanelli, or Glockenspiel), Campane. (a deeper bell of some sort - whether
this is a set of regular Tubular Bells, or a special set of bells I once read
Scriabin had had made especially for this work, I'm not quite sure), Arpa I,
Arpa II, Organo, Sopr., Alti, Ten., Bas. (the four parts of a Chorus, that
* p. 7, bar 44 - Piano, r.h.:
Error: The fourth and fifth r.h. chords are tied together. There
are five notes in the chord, but only four ties, that for the A being
Reason: It is clear the chord should be completely tied, and there
is no reason to suppose that the A alone is to be restruck.
* p. 11, system 1, bar 92 - Piano, l.h.:
Error: The third note is misaligned, and is too far to the right; it
should appear beneath the second chord in the r.h.
Reason: It's a matter of vertically lining up notes that sound
* p. 21, bar 167 - Piano, r.h.:
Error: The r.h. has an "8va" sign over the entire bar; but it should
stop before the third-last note.
Reason: This would allow the figuration to have the overall downward
direction it seems obviously intended to have.
? p. 21, bar 167 - Piano, r.h.:
Error: In the same figuration referred to in the previous item, it
seems possible that the 4th-last note, a C#, should be D#.
Reason: This is a very spurious supposition; but slight support for
it can be found in the fact that, as this figuration descends from one octave to
another, both the higher and lower octaves contain a D# instead of a C#.
However, equally possibly, the discrepancy might be intended so that the 9th and
7th intervals that are created instead of octaves give a certain harmonic and
textural colour. So it is only an item I would like to check from other sources
- not one I am confident is mistaken.
* p. 23, bar 182 - Piano, r.h.:
Error: The 1st and 7th quavers are misaligned, and should be moved
slightly to the left.
Reason: The quavers need to be exactly above the l.h. notes they
sound together with - namely, the first semiquavers in the 1st and 3rd
* p. 28, bar 222 - Luce:
Error: The Luce part has two lines: one which changes very slowly,
rising up the scale by a note or two every several dozen bars; and a second
voice which changes with every change of harmony, this one moving to the note
which is the root of the current harmony. Judging by this, the A in the upper
voice is wrong, and should probably be F#.
Reason: There is actually slight room for doubt about what the note
should be; but, in any case, the only other possibility would be C, and A is
absolutely ruled out.
In the harmonic language used in this composition, many of the chords are
based on the so-called "Prometheus" scale, which, in C, for example, would run
C D E F# A Bb. Any chord which uses most or all of these notes would be
regarded as having a root of C, regardless of which note was in the actual bass.
(It is directly analogous to the fact that, in a C-major triad, C is the root,
regardless of which note is in the bass.) By this criterion, the harmony I am
claiming to wrongly have A in the Luce part is definitely based on F#, the
Prometheus scale in this key running thus: F# G# A# B# D# E. The harmony at
this point uses all these notes, and the only other note is a G-nat., which is
clearly nothing more than a passing note (and probably should, strictly
speaking, be written as Fx).
The note in the actual bass, however, is C. Regarded as an alternate
spelling of B#, this does belong to the Prometheus scale in F# - but there is an
anomaly in that Scriabin almost always spells his notes correctly, and is not
inclined to respell notes to a more common name - he will even let a chord have
a fistful of obscure and complex accidentals if such notation is correct, rather
than adopt a simpler spelling which is incorrect, or which inteprets the harmony
differently. If the chord at issue here is in the F# Prometheus scale, the note
which is in the bass here should be spelled as B#, and Scriabin would normally
do so, even if it was the bass note. In general, if he spells a note wrongly
(according to his own system), it certainly causes me to look at the harmony
again and consider whether it should be interpreted differently. (In his often
complex harmonic language, there can sometimes be more than one way of
interpreting a chord, more than one note that can be considered the root, and
these would call for different enharmonic spellings of some notes.)
Thus for him to spell this bass note as C instead of B# is
uncharacteristic, and leads one to wonder if he regarded C as the root of the
chord, in which case the F# is spelled correctly. However, the rest of the
harmony does not fit the Prometheus scale based on C. It is because of this
that I consider that the Luce part should probably be F# rather than C, although
there's slight room for doubt. (Not all chords in this composition are based on
the Prometheus scale. And a chord that is in the Prometheus scale when F# is
regarded as the root is seen as something completely different when C is
considered the root. At any rate, from the C perspective, it certainly is not
the Prometheus scale.) However, be that as it may, I can see no way that A is
correct in the Luce part. It is correct for the next bar - so I assume
this misled the composer or typesetter into putting an A incorrectly in this bar
* p. 36, bar 256 - C.B.:
Error: In this bar, the part is divided into two. In the lower
part, after the 3rd note, a G#, there should be a semiquaver rest.
Reason: It's needed to make the value of the lower part add up
properly, and align properly with the upper part. Without the rest, the G#
semiquaver appears to occupy the space of a quaver.
* p. 37, bar 264 - Fl. I.:
Error: The last four notes are a little too far to the right,
aligned wrongly with other parts.
Reason: To correct this, the 3rd-last note should be positioned
where the 4th-last note is.
* p. 37, bar 264 - Viole:
Error: In the second beat, an A crotchet (trilled, and actually part
of a triplet) is followed by a semiquaver (which is part of the same triplet.
There are two things wrong here:
(a) The semiquaver is a double stop, consisting of G and A. The A is at
the same pitch as the A crotchet, and the two are tied (correctly). But the G
should be removed, leaving just the A.
(b) A semiquaver rest should follow the semiquaver just mentioned.
(a) The G component of the double-stop doesn't fit the harmony or style,
and the double-stop itself, if played, would jar with the context. The overall
texture and feeling of the passage clearly suggests that a line of trills
separated by very short rests is all that is required in this part.
(b) The crotchet which first appears is two-thirds of the triplet, and the
remaining third should add up to a quaver. Since the note itself is only a
semiquaver, clearly a semiquaver rest has to follow it for all the values to add
up correctly. The same rhythmic figure occurs on all three beats of this bar,
and the other two instances do include the semiquaver rest.
* p. 38, bar 267 - Celli:
Error: The lowest note of the last chord in the bar should have a
flat sign in front of it.
Reason: Clearly the harmony requires Ab rather than A-nat.
* p. 38, bar 269, Cl. II., III.:
Error: The sharp sign in front of the second semiquaver, in the
upper part, should not be in parentheses.
Reason: The parentheses suggest that the sharp sign is cautionary
only - that, even without the reminder, the note would be sharp anyway; but the
C definitely must have a sharp sign, and would be natural without it, and
therefore the sign is not cautionary. (The natural signs appearing in the same
bar are cautionary, though.)
? p. 43, bar 295 - Fl. Picc., Camp.:
Error: Possibly the last note in the bar in each of these parts
should be an octave higher.
Reason: This would preserve the melodic shape of the figure, which
is a very distinctive one of some importance in the work. But possibly Scriabin
intentionally compromised this, maybe feeling that it took these instruments
into an uncomfortably high range.
* p. 45, bar 308 - Arpa I:
Error: The r.h. chord probably should have two harmonic signs above
it, not just one.
Reason: There is no reason to think that only one of the notes is
intended to be a harmonic. The previous two bars are obviously similar, and
those have two harmonic signs each for the r.h.
* p. 47, bar 319 - C.B.:
Error: Possibly the Eb in the Double Bass part should, in this bar
only, be notated as D#, in spite of it being Eb in both the preceding and
Reason: See the item at bar 222 for a
discussion of how Scriabin selects the enharmonic spelling for notes. His
system is somewhat obscure by normal standards; but he is usually very careful
and consistent about this, and rarely uses a wrong spelling (judging by the
complex system he possibly devised himself).
According to the criteria he consistently uses in this work, and, indeed,
in all his compositions, there is no doubt that this note should be written as
D#. The harmony and melody in this bar are almost completely based on the
Prometheus scale on A, which goes: A B C# D# F# G. Seen in this light, Eb is a
The harmonic structure of two bars back is very similar, and Scriabin
does write D# there. The four bars from 317 - 320 are, in fact, based
alternately on Prometheus scales on A and Eb. When you have two Prometheus
scales based on notes a tritone apart (one of the basic harmonic techniques used
in this work), one of them must spell the other's key-note differently
from the other scale itself. Here the Eb/D# is the note whose spelling changes
from one scale to the other.
* p. 50, bar 336 - Piano, l.h.:
Error: In the last beat, the l.h. contains a group of three
semiquavers beamed together, followed by a fourth semiquaver and a
semiquaver rest. The group of three semiquavers should have a triplet sign.
Reason: This would normally be the way I would interpret this bar.
But there is a slight notational anomaly, which is probably nothing more than a
typesetting flaw, which could cause one to interpret the rhythm differently.
The triplet beam and stems go downwards, and the single, fourth semiquaver,
which is a sixth lower than the last triplet note, is stemmed upwards. It is
also printed so close to the last triplet note that the notes are very nearly on
top of each other. (The opposite directions of the note-stems make it possible
to put the notes this close together.)
This could give rise to the impression that the notes are intended to be
played together, and that the triplet is not really a triplet, but three normal
semiquavers taking up three quarters of the beat. If this passage is counted as
two separate voices, it really does need a couple of extra rests to complete the
value of each voice within that beat - but, except for that omission, it is
possible to interpret the rhythm as straight semiquavers. And if you look at
the passage quickly, before you have time to figure it out, it is rather
ambiguous as to what is intended.
However, I firmly believe that the first interpretation I gave is the
intended one, and that the near-vertical alignment of the last two notes is in
error. Confirmation of this can be seen earlier, in bar 328 on p. 48, where a
very similar passage occurs in the piano l.h. staff; but this time the notes are
separated properly, and a triplet sign provided - and the meaning is completely
clear. The analogy with bar 336 is obvious, and I would read it in a similar
If a triplet sign were provided here, and the near-vertically aligned notes
separated a bit more, it would make it much clearer.
* p. 50, bar 339 - Cel.:
Error: The third note, B, should be D. It should also be on the
lower staff (using one leger line).
Reason: B is wrong harmonically, but D fits. The Celesta figure is
a well-defined motif, and B does not give the right melodic shape to the figure,
whereas D does. Also, the figure, when it appears in the Celesta part, is
usually distributed between the two hands in such a way that the hands play
notes alternately, especially when the intervals between the notes are wide (a
7th or more), and giving this D to the l.h. fits that pattern better.
It seems to me that this may be a typesetter's error: the wrong note and
the correct note I am proposing both have in common that they are beyond a
single leger line, and the typesetter may have inadvertently transferred the
note from the top of the l.h. staff to the bottom of the r.h. staff, keeping
(but inverting) its position relative to the staff, and thereby altering its
pitch by a minor 3rd.
* p. 52, bars 346, 348 - Trb. III., Tuba:
Error: There is no indication as to which instrument(s) play the Eb.
Reason: The staff is shared by two instruments, and therefore, when
only one note appears at any time, it should be shown which instrument is
playing the note, or whether both play it together - either by explicit
labelling, or by the use of appropriate rests to show what instruments are not
playing at the time. My guess is that the note would be taken by both
instruments together, given that in other bars in the same passage, they are
playing in octaves. This is a bass-line, and basically it is being played by
both instruments: they just happen to be in octaves at some times, and in unison
* p. 52, bar 347 - Fl. Picc.; p. 53, bar 350 - Fl. Picc., Fl. I.:
Error: In each case, the trill sign is too long, and should be
Reason: The instrument in question rests on the third beat of the
bar, so the trill sign should stop just before the crotchet rest.
* p. 53, bar 351 - Camp., second staff:
Error: The first note is positioned too far to the right.
Reason: It needs to be aligned above the 2nd semiquaver in
the Celesta part (immediately below), not above the 3rd semiquaver.
? p. 54, bar 354 - Fl. Picc.:
Error: Posssibly the B-nat. needs to be an octave higher.
Reason: Once again, it is a recognizable figure, similar to the
Celesta figure at bar 339, discussed above, and having the B an octave higher
would preserve the shape of it. However, it also appears possible, perhaps even
likely, that Scriabin judged this Piccolo note (the second-highest note on the
instrument) to be too shrill for use here - so this proposed correction is only
tentative. However, it might be noted that, in the previous bar, the composer
did not find the high Bb, only a semitone lower, to be too high in register for
use in an almost identical figure.
? p. 56, bar 367 - Percussion:
Error: Should there perhaps be a cymbal clash at this point?
Reason: I am sure I have heard recordings that used the cymbals at
this climactic point - but of course that may be only conductors' additions.
* p. 56, bar 369 - Cor., 1.-VIII.:
Error: All eight Horns are shown holding a dotted-minim chord, which
lasts for the full length of the bar; this chord almost certainly should be
without a dot, and a crotchet rest inserted after the chord.
Reason: On the third beat of the bar, much of the rest of the
orchestra plays a triplet figure in full harmony which clashes with the chord
held in the Horns. In spite of the complex harmonic style of this composition,
a clashing of harmonies as they change against each other is not in the least
characteristic, and it leads me to feel that almost certainly the Horns are
If it had been a full-blooded, extravagant, romantic work by someone like Arnold Bax or Karol Szymanowski (just to pick 2 obvious composers out of many I could have named), I would feel rather less certain on this point; but with Scriabin, I'm prepared to stick my neck out and assert this with certainty. Prometheus is certainly very lavish and late-romantic in style, and possibly could even be described as "full-blooded" - and certainly could be described as "extravagant". However, music by composers like Bax and Szymanowski has a certain "overgrown", chaotic feeling to it, a certain irregularity to it, so that it might be very difficult to know if something was a mistake or an intended irregularity; whereas Scriabin's music, for all its high-flown romanticism, has a very pure consistency of style to it, even a kind of disciplined regularity, an economy of texture that almost reminds me of J. S. Bach (although perhaps on a larger scale) - and thus I feel confident that there are certain things that Bax or Szymanowski might do, which Scriabin would never even dream of doing.
Accordingly, I believe that the minim in all the Horn parts should be
without a dot, and that it should be followed by a crotchet rest.
* p. 57, bar 374 - Cor., VI., VII.:
Error: The lower note, a D (concert pitch, A), should be D# (G#).
Reason: An examination of the harmony, plus a look at unison
doublings of this note elsewhere in the orchestra, makes it clear that the note
needs correcting thus.
* p. 57, bar 377 - Piano, r.h.:
Error: In the r.h. second chord, the C should be C#.
Reason: C# agrees with the harmony in the entire passage, whereas C
* p. 58, bar 386 - Piano, r.h.:
Error: The entire r.h. contents of this bar need an "8va" sign
Reason: Necessary to preserve the continued upward motion of the
figure, found in many other passages in this work. Moreover, without this
adjustment, the two hands would constantly clash in this bar, and be very
awkward to play.
? p. 58, bar 389 - Viole, Celli, C.B.:
Error: Probably all the Gb's and Ab's in this bar should be spelled
as F#'s and G#'s.
Reason: The spelling used for these notes in this bar violates the
criteria Scriabin normally uses very strictly and precisely for enharmonic
notation. (For more on this, see bar 222.) The fact of
several notes being spelled wrongly probably rules out a simple slip of
the pen or typesetter's error, and I have to accept that this is probably what
Scriabin wrote (and I say "probably" because I do not have access to a facsimile
of the composer's manuscript). But the fact is, it does violate the way he
normally spells notes.
For the misspelled notes at the beginning of the bar, it can probably be
explained by the continuity of the parts from the previous bar, in which the
notes are correctly notated in their flat form. Perhaps Scriabin felt
prepared to compromise his spelling system to allow any one instrument to avoid
having two different spellings of the same note in quick succession in a single
figure. But there are passages elsewhere in his music where he does not make
this concession, and will quite happily write (for example) Ab followed
immediately by G#.
Less explicable is the continued use of the misspellings later in the bar -
well into the area where a new area of harmony based on E is established, in
which F# and G# are the correct notation (and used in other instruments at this
* p. 59, system 1, bar 390 - Viole:
Error: The Cb marked in the second-last chord should be C-nat. (with
no natural sign needed because the C would be natural anyway).
Reason: The Cb is wrong harmonically, and clashes with the other
strings which play C-nat. in the same octave in close proximity.
* p. 60, system 1, bar 407 - Piano, l.h.:
Error: The last l.h. quaver is misaligned, and should be moved a
little to the left.
Reason: The note is played before the last r.h. note; therefore it
is wrong for it to be lined directly beneath it.
* p. 60, system 2, bar 419 - Piano, l.h.:
Error: The F in the first chord should be F#.
Reason: F# fits the harmony, whereas F-nat. does not.
* p. 62, system 2, bar 434 - Piano:
Error: In the second beamed group of notes, the first two notes
probably should be G and A, not A and B.
Reason: The B clashes with the prevailing harmony, and would not be
acceptable even as a passing note. (The only non-harmonic passing note that
occurs in this composition seems to be the note a semitone below a major 9th
above the root of the harmony. But Scriabin never uses a minor 13 as a passing
note in places where the harmony is, or includes, the major 13th above
In addition, G and A allow the figure more precisely to conform to the
shape and intervallic make-up that is very characteristic of this figure which
is used so intensively in this passage. Comparison with some of the surrounding
figures should make this clear.
? p. 63, bar 445 - Fag. III:
Error: The C# shown here is presumably what the composer wrote; but,
strictly, according to the composer's own criteria for enharmonic spelling, it
should be Db.
Reason: The low strings use Db, and this obviously fits in with the
prevailing harmony based on the Eb-Prometheus scale. See bar
389 for more about this. However, in this case, unlike
the one in bar 389, there are no considerations of continuity in a single part
that might indicate that the correct spelling of the note should be overridden.
* p. 65, bar 454 - Fag. III:
Error: The Bb on the second beat should be Db.
Reason: Agrees with other instruments doubling it; Db fits
the prevailing harmony, but Bb does not.
* p. 70, bar 488 - Organo, l.h.:
Error: The lower Bb, the minim, should have a tie leading into
it from the previous page.
Reason: The corresponding note at the end of the previous page has a
tie leading out of it, obviously to indicate a tie to the next system. The
organ part is mostly sustained harmony, and, in this tied chord, there is no
reason to think that one of the Bb's should be played again at this
* p. 76, system 1, bar 556 - Piano, r.h.:
Error: The last semiquaver in the first beamed group, Gx, should be Fx.
Reason: Fx fits the harmony; Gx does not.
* p. 76, system 1, bar 559 - Piano, l.h.:
Error: The fourth note in the demisemiquaver group, D#, should be D-nat.
Reason: D-nat. fits the harmony, but D# does not.
* p. 76, system 2, bar 560 - Luce:
Error: The Db in the lower part should be tied into the next
Reason: The lower line in the Luce part matches the root note of the
prevailing harmony, which is Db in both these bars. In the Luce part, Scriabin
usually ties the notes together when they are the same in two adjacent bars -
and the absence of the flat sign in front of the second D also indicates that
the tie should be included.
* p. 78, bar 596; p. 79, bar 599 - Organo, r.h.:
Error: The F in the first of these two bars should be tied into the
next bar; the Bb in the second of them should be similarly treated.
Reason: The Organ is playing sustained background harmony here,
and all the notes are tied from one bar to the next when they are the same.
There is no reason to suppose an exception to this was intended here.
It is true that the Organ chord in bar 600 is not tied to bar 601, even
though the harmony stays exactly the same; but it appears in this case that the
chord as a whole was intended to be articulated again for emphasis, so it does
not affect my assessment of the missing ties in bars 596 and 599.
* p. 79, bar 602 - Cl. II., III.:
Error: The two Clarinets trill on the notes Ab and Eb (F#, C#,
concert pitch). The two trill signs for the two notes both have flat signs,
thus indicating the auxiliary notes to be Bb and Fb (G# and D). The flat sign
for the Eb trill should be omitted, so that the auxiliary note is F-nat. (D#).
Reason: Much of the orchestra is trilling at this time, most of the
instruments being non-transposing and thus written at concert pitch. Those
instruments trilling on C# all use D# as the auxiliary note, and there is
nothing about the passage that indicates that some instruments should have
different inflections for the auxiliary notes.
To do so, or to have all the instruments trilling on C# use D-nat. would
not be in keeping with the style of this work. To the contrary, the harmony is
an F#-major triad: the conclusion of the work, and the only triad in the entire
piece, which finally gives resolution to the harmonic tension that has built up
previously - and, in accordance with this, the D# (F-nat. for the Clarinet) is
simpler and more in keeping with the pure F#-major feel of this final chord.
? p. 79, bar 606 - Percussion:
Error: I am sure I have heard recordings of this work where there
was a second Bass-Drum stroke in the last bar of the piece...
Reason: ... but of course it could merely be something the conductor
decided to add.
The Poem of Ecstasy and Prometheus: The Poem of Fire,
although rather different in form and duration from the Symphonies 1 - 3,
clearly can be seen as successors to them. I have very occasionally seen them
referred to as Symphony no. 4 and Symphony no. 5 (perhaps more often as a
subtitle to the well-known titles than as primary titles), although this is very
rare, and has no official status whatever that I am aware of. Certainly my
copies of the scores make no mention whatever of these possible alternative
But the third work in this series of five orchestral works is clearly
midway in status between the early symphonies and the late symphonic poems: it
is called "Symphony no. 3", and is without a doubt a symphony as far as form
goes - but its title "The Divine Poem" clearly indicates it also to be a
symphonic poem at the same time, clearly with some kinship with the two late
symphonic poems. All that had to happen was for the three movements of the 3rd
Symphony to be telescoped into a single condensed movement, as in the last two
orchestral works. The midway status of The Divine Poem is, in my view,
clear evidence that the symphonies and symphonic poems can very validly be
regarded as a continuous series of related works, which appear to be different
in form merely because the symphony/symphonic poem genre Scriabin used itself
changed as he developed as a composer.
A similar process can be seen in the ten wonderful Piano Sonatas (every one
of them an absolute gem of piano writing, in my opinion, with not a single dud
amongst them): the early ones are multi-movement works (4, 2, 4, and 2 movements
respectively in Sonatas 1 to 4), and the later ones (from no. 5 onwards) are in
the same highly-condensed single-movement form as the two last symphonic poems.
The difference here is that they are all called Sonatas, making
completely clear their kinship and sequential relationship. I would argue that
the three symphonies and two symphonic poems (or two symphonies and three
symphonic poems, depending on how you classify the Symphony no. 3 / The
Divine Poem) should clearly be regarded as a single series of works in
exactly the same way as the Sonatas are, even though their titles do not, unlike
the Sonatas, follow a single format.
And what both of these remarkable series of works have in common is that,
even as the length of the works get shorter and shorter and more condensed, the
scale of the conceptions behind them got ever bigger and bigger: you can listen
to these late works - a mere 10 minutes or even less in length in the case of
the Sonatas, and less than 20 minutes in the last two symphonic poems - and get
the impression you've heard something far longer. In spite of their brevity,
these works are truly as grand in scope as anything Wagner or Richard Strauss
[ Back to: The Poem of Ecstasy -
Prometheus: The Poem of Fire ]
I hope this list comes one day to the notice of anyone who might be
involved in producing future editions or reprints of Scriabin's orchestral
music. If any such people read this, please e-mail me if you think the above
list will be of assistance in correcting the editions, or if you think, on the
above showing, I have the capacity to be of assistance in any other way -
perhaps doing more error-hunting, or perhaps even proof-reading of a new
engraving of these works.
If any such people who read this really feel I have nothing useful to
contribute, and that I am merely duplicating discoveries others have already
made, I would appreciate being told this so that I do not waste time searching
the scores for mistakes.
But I would very much appreciate feedback or suggestions from anyone who
reads this who is familiar with these scores, or who might one day be involved
in producing future editions of these works.
Thursday, 3 January, 2002.
E-mail me about this music.
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.
Introduction - Front page, which leads to Contents
Web Site of Michael Edwards - Contents
( Alexander Scriabin )
Errors in orchestral works by Alexander Scriabin (this page)
This page created on Thursday, 3 January, 2002;
last modified on Tuesday, 17 February, 2004.