Berlioz: Symphonie fantastique, op.14

2014-09-04 — Original posting (on Blogger)
2014-09-13 — Added reference to review for a concert on 2014-09-11
2014-11-13 — Re-posting as is (WordPress)
2015-07-04 — Added link for music score
2016-07-24 — Brushed up for better readability

Introduction / The Recordings:

This posting is about the Symphonie fantastique, op.14 by Hector Berlioz (1803 – 1869), of which I currently have 2 recordings on CD:

  • Pierre Boulez, Cleveland Orchestra (1996)
  • Paavo Järvi, Cincinnati Symphony Orchestra (2000)

In this case I did not try “resurrecting” the recordings in my LP collection, where I have recordings by Igor Markevitch (L’Orchestre des Concerts Lamoureux, Paris) and Sir George Solti (Chicago Symphony Orchestra). None of these was added by systematic search / collecting, or out of a specific preference: I was just following mere curiosity. So at that point I did not feel a real need to add digital versions of these older recordings.

Background, About the Composition:

Hector Berlioz (1803 – 1869) wrote his Symphonie fantastique, op.14 before he turned 30. It was first performed in Paris in December 1930. As Alain Pâris points out in the liner notes to the first recording below, this was before Beethoven’s ninth symphony was introduced in France (March 1931). Both works revolutionized the genre of symphony, though in totally different ways. Indeed, the work we are discussing here marks the invention of a new genre, the symphonic poem. That term hadn’t been coined yet, though. Superficially, Beethoven’s sixth symphony (“Pastoral”) appears to do the same. But there, the descriptive titles given to every movement mostly just describes a generic, static scenery.

Events in Berlioz’ Biography that Led to this Composition

Berlioz’ work first carried the title “Episode de la vie d’un artiste” (episode from the life of an artist), and the composer wrote a detailed, two-page “libretto” that he described as being indispensable for understanding the dramatic contents of the symphony. Instead of reproducing that program here, I have prepared PDF versions of the text (taken from the Eulenburg score EE 6665, with minor typographical adjustments), both in French and in English. One should note that the story in that “libretto” originates from events in Berlioz’ personal biography, namely the encounter with the Irish actress Harriet Smithson in 1827 (and the despair caused by her cold responses towards the composer).

This was compounded by the discovery of Shakespeare, Beethoven’s symphonies, and Goethe’s Faust; all this led to a plan to take revenge towards the actress by making a big splash in the concert halls through his upcoming symphonic composition. And indeed, the composition was a huge success already in the first concert, in 1830. Interestingly, Berlioz revised the score two years later, then also composed a sequel, “Le retour à la vie“, later entitled “Lélio“, describing the artist’s recovery through the healing power of music. Life then followed the score! After these two compositions were presented in the same concert, in 1832, in the presence of Harriet Smithson, Berlioz and the actress were introduced. They married within a year (a marriage that didn’t last very long, though).

Is the “Libretto” Needed?

Back to op.14: in the early 19th century it may indeed have been necessary to provide the “libretto”. However, I think that Berlioz underestimated the power of his music, at least for the succeeding generations: today’s audiences are undoubtedly able to enjoy this piece with merely a vague idea about the underlying poetic program, or even without ever looking at the “script” at all.


The underlying dramatic concept was not the only area where this piece was revolutionary: the symphony also marks the first use of a number of instruments, techniques and sounds/sonorities in a symphonic composition, such as

  • two pairs of harps
  • the cor anglais
  • a pair of bells
  • clarinet in E flat
  • four tuned timpani
  • multiple groups of double basses
  • string instruments playing col legno (beating the strings with the wooden part of the bow)

and more (his instrumentation in general), plus, the extensive use of syncopated rhythms (that we now would call jazzy). On top of that, in many parts it is a very virtuosic composition, if not a “show piece” for orchestras (ideally!) to demonstrate their precision, perfect coordination, as well as orchestral balance etc.; Berlioz is extremely detailed in his score, and he also specifies the minimum strength of the string section (15 + 15 violins, 10 violas, 11 celli, 9 double basses). Overall, it’s a fascinating, enthralling composition that rarely fails to impress audiences.

The idée fixe

The so-called idée fixe that Berlioz refers to in the program is a single theme that appears in and unifies all five movements. It first appears in bar 72 of the first movement (only flute part shown here):
Berlioz, Symphonie fantastique, op.14: idée fixe

The Movements:

As outlined in the composer’s program, the symphony has five movements, all of which are carefully annotated with metronome numbers.

I. Rêveries — Passions

Largo, 4/4 (1/4 = 56) — Allegro agitato e appassionato assai, 2/2 (1/2 = 132) — Religiosamente
Interesting detail: in bar 17, Berlioz felt compelled to add the (now somewhat patronizing) remark

Les onze mesures qui suivent sont d’une extrême difficulté; je ne saurais trop recommander aux chefs d’orchestre de les faire répéter plusieurs fois et avec le plus grand soin, en commençant au changement de mouvement (plus vite) et finissant à la rentrée du thème (Tempo I). Il sera bon de faire étudier leur trait aux premiers et deuxièmes violons séparément d’abord, puis avec le reste de l’orchestre, jusqu’à ce qu’ils soient parfaitement sûrs de toutes les nuances de mouvement, qui me paraissent ce qu’il y a de plus difficile à obtenir de la masse, avec l’ensemble et la précision convenables. [HB]

or, in my translation:

The eleven bars that follow are of extreme difficulty; I couldn’t recommend more to the conductors to have these repeated several times and with greatest care, beginning at the tempo change (plus vite) and ending with the return of the theme (Tempo I). It will be best initially to have the first and second violin sections study their parts separately, then with the rest of the orchestra, until they are perfectly firm about all nuances of the tempo, which to me appear to be hardest to obtain from the entire body, with all voices together, and with the necessary precision. [HB]“.

II. Un bal

Valse, Allegro non troppo, 3/8 (3/8 = 60)

III. Scène aux champs

Adagio, 6/8 (1/8 = 84)
This is the one movement that resembles the equivalent (also third) movement in Beethoven’s 6th symphony, the “Pastorale”. In that composition, the third movement “Lustiges Zusammensein der Landleute” (joyful gathering of the country people) describes a countryside idyl that in the end gets disrupted by an upcoming thunderstorm. There is thundering at the end of Berlioz’ third movement as well. However, in this case, this is the anticipation of the march towards the scaffold that follows in the subsequent movement.

IV. Marche au supplice

Allegretto non troppo, 2/2 (1/2 = 72)

V. Songe d’une nuit du Sabbat

Larghetto, 4/4 (1/4 = 63) — Allegro, 6/8 (3/8 = 112) — Allegro assai, 2/2 (1/1 = 76) — Allegro, 6/8 (3/8 = 104) — Dies irae — Ronde du Sabbat: un peu retenu 6/8 (3/8 = 104) — Dies irae et Ronde du Sabbat ensemble
The un peu retenu (holding back a little) in Ronde du Sabbat is explicitly stated to be the same tempo as in bar 40 (Allegro, 6/8, 3/8 = 104).

The Interpretations, Overview:

In order to provide a rating overview, as well as an idea about tempo relations both within an interpretation, as well as between the two recordings, I have prepared the little table below. Note that the color coding for the tempo (blue = slower, green = faster) refers to the metronome markings in the score (column #5), not to the average between the recordings:

Berlioz: Symphonie fantastique, op.14: timing/rating comparison table

Note that measured metronome values are estimates (just consider the amount of rubato used in this music!). The metronome numbers refer to the time units specified by Berlioz — e.g.: for the 6/8 bars in the third movement, the unit is 1/8.

The Interpretations, Detail:

I have partly relied upon some Amazon reviews for the selection of the two recordings below, even though I don’t usually rate such reviews very high. For the first recording: I felt that even though Pierre Boulez (1925 – 2016) is himself a modern composer and somewhat specialized in music from the Second Vienna School and beyond, as a French conductor (even though conducting an American orchestra in this recording from 1996) he should most likely be able to deliver a good, if not excellent interpretation of this music. In order to have a contrast, I selected one of the popular younger conductors of current times, the Estonian Paavo Järvi (*1962), also directing an American orchestra (even though not one of the “big 5”, as Boulez) in a recording from 2000.

Pierre Boulez, The Cleveland Orchestra

Berlioz: Symphonie fantastique, op.14 — Boulez; CD coverBerlioz: Symphonie fantastique, op.14; Tristia

Pierre Boulez, The Cleveland Orchestra & Chorus

DG 4553 432-2 (CD, stereo); ℗ / © 1997
Booklet: 26 pp. e/d/f
Berlioz: Symphonie fantastique, op.14 — Boulez; CD, UPC-A barcode
—Find CD(s) on—

spacerWith the Cleveland Orchestra, Pierre Boulez (1925 – 2016) was conducting one of the “top five” orchestras in the US, as a guest conductor (he had previously been the orchestra’s Musical advisor in 1970 – 72). As mentioned above, this recording was made in 1996. An American Orchestra would not necessarily be my first choice for this composition. But my guess that the conductor is the key factor here turned out to be correct, see my comments below.

The orchestra features a compact, balanced, “integrated” sound, excellent string bodies, and overall a very good performance. An excellent, solid tool in Boulez’ hands, I think! The one thing I’m missing in this performance is the “show effect”, which, I believe, is also inherent to this music. The dynamics are not quite as expansive as in Järvi’s recording, but this may partially be attributed to the sound engineers.

Notes on the Movements

Some details on the movements:

I. Rêveries — Passions

Duration: 13’58”
In the tricky passage starting in bar 17 (see above), Boulez “played safe”, taking a moderate tempo, ensuring that that sextuples are played cleanly and in all detail. It is all clear and clean, however, it also misses the virtuosic challenge (the inherent “circus effect”, so to say). In the Allegro agitato e appassionato assai it becomes clear what Boulez was aiming for: his focus is on expression, emotions, and on the atmosphere, as given in Berlioz’ program, all exemplified by the wonderfully singing violins (without excessive vibrato) and woodwinds!

II. Un bal

Duration: 6’10”
Much better than Järvi / CSO! The waltz feeling is there, the portamento is noticeable, but a nice feature (and used in the specified places only) rather than a dominant show effect. Even the Cleveland Orchestra has some coordination issues in bars 242 ff., but not nearly to the extent shown by their contenders. The stretta is enthralling, and played very well, congrats!

III. Scène aux champs

Duration: 15’23”
With 1/8 = 76, the tempo here is more fluent than Järvi’s, but still fits the meaning of Adagio, i.e., calm (Berlioz specifies 1/8 = 84). Very nice, these singing violins & flutes, the agogics! Boulez paid attention to the emotions in the music; he does not aim for overall perfection, but he plays the program! Yes, the dynamics are less impressive than with the Cincinnati’s (e.g., the oboe “behind the scenes” is much closer here), but that is not the key to this movement.

IV. Marche au supplice

Duration: 7’03”
This interpretation may sound less virtuosic than Järvi’s, but it follows the score, including the repeat, the instrumentation and the playing instructions. Also the tempo is closer to Berlioz’. The only “hair in the soup” is some nasty tapping noise in the first part. It sounds like either the conductor or one of the instrumentalists. This could easily have been avoided. I’m not asking for perfect audio, but there should not be such distractions.

V. Songe d’une nuit du Sabbat

Duration: 10’04”
An excellent interpretation, with the tempo pretty much following the score (only the last part is faster). Also technically, the interpretation is excellent. The only minor incident that I found is in bars 399 – 402, where the trumpet is out of sync for a while. The fast passages in the strings in the last part (Dies irae et Ronde du Sabbat ensemble) are clean. In the entire movement, the articulation in the strings (in particular) is excellent.

Overall Duration: 52’38”
Rating (see above for details): 4.2 — Among the two recordings that I compared, I don’t hesitate to recommend this one over Järvi’s, by a long stretch!

Paavo Järvi, Cincinnati Symphony Orchestra

Berlioz: Symphonie fantastique, op.14 — Järvi; CD coverBerlioz: Symphonie fantastique, op.14; Love Scene from “Romeo et Juliette”

Paavo Järvi, Cincinnati Symphony Orchestra

Telarc SACD-60578 (DSD/SACD, stereo/multi-channel surround); ℗ / © 2001
Booklet: 12 pp. English
Berlioz: Symphonie fantastique, op.14 — Järvi; CD, UPC-A barcode
—Find CD(s) on—

Even though it is not an unknown entity, the Cincinnati Symphony Orchestra is not one of the top five in the US. From listening to the two recordings in this comparison, that seems to apply to both orchestras here. The weakest part is in the strings, particularly the violins, which use an excess of vibrato, to a degree that often affects the intonation. Also, the strings often sound non-uniform, even thin at times. The strength of the orchestra is in the brass section. However, in Berlioz’ music, the brass section is often not acting in solo parts, but just adding specific colors. It’s understandable that the recording (and the conductor?) wants to highlight the best part of the orchestra. But that doesn’t necessarily help this music.

The key quality in this recording is in the recording technique: an SACD with 5-channel surround sound. However, this again only helps a minor part of the audience that has the necessary equipment. I’m listening to this in ordinary stereo. But even there, the transparency and the spatial plasticity of the sound are excellent, especially highlighting the quality of the brass and percussion sections. Unfortunately it also exposes the weaknesses in the string section. To me, the musical quality is far more important than perfect sound. This recording was made in 2000, with the Estonian conductor Paavo Järvi (*1962). Järvi is a popular name in today’s recording scene. Maybe he is not the optimum choice for this music? See the following details on the various movements.

Notes on the Movements

I. Rêveries — Passions

Duration: 15’28”
There are several weaknesses in this recording, primarily in the strings, as outlined above. The intonation is sometimes fairly bad, but there is also a lack of agogic play. Often, the music is too metric, lacking the rhythmic differentiation between weak and strong parts within the bar. I miss some rhythmic elasticity, if not even the “soul” of this music. It is not enough to play the notes and let the libretto do the rest. To the contrary:

I believe that primarily, the orchestra and the conductor must understand and “digest” the program. Then, the listener will be able to understand this music even without the libretto. The horns (and the brass section in general) are excellent, but these don’t act as soloists, they merely add color to the overall sound. The virtuosity of the brass section and the overall volume may impress the audience, but are listeners really able to understand the emotional content of the composition? I doubt it.

II. Un bal

Duration: 6’24”
The vibrato in the violins is fairly awful. Berlioz specifies portamento in specific intervals in the waltz melody. Here, it is exaggerated, especially in combination with the strong vibrato. Also, the fact that Berlioz wants portamento does not mean that one should apply it ad libitum in arbitrary other places as well. I also miss the “waltz swing”. The agogics (where applied) appear badly coordinated within the orchestra. Coordination in general is fairly marginal, bars 240 – 252 a total chaos, the last bars superficial. It is possible that in order to help the plasticity of the recording, the orchestra was spread over a large area. This of course would not help the coordination.

III. Scène aux champs

Duration: 17’17”
Berlioz specifies 1/8 = 84 — Järvi uses 1/8 = 68, which is definitely too slow, making the music static, and it probably is also the reason why the movement lacks tension and emotion in this interpretation. The movement would also profit from more agogics. The one positive feature here is the nice sound of the woodwind instruments. But it is not enough to play this music formally correctly and accurately.

IV. Marche au supplice

Duration: 4’47”
Berlioz explicitly writes that the 1-bar injections in bars 2, 4, ff. are to be played by the horns, muted with a hand, not with the regular mute. Järvi not only ignores this, he also has this played by trumpets with mutes. I suspect he was after the grotesque effect that this creates. It’s definitely not Berlioz’ intent. Maybe also the tempo is a tad fast. Overall, this feels more like a “happy walk towards the scaffold”, or like a moody Scherzo, perhaps. The repeat of the first part is not observed. That’s a pity in this short movement! Still, technically, this movement is well-played.

V. Songe d’une nuit du Sabbat

Duration: 10’08”
The movement describes a ghastly, weird scene. I can hardly feel this. The beginning is too slow, the fast sections are mere show effect. The last part (Dies irae et Ronde du Sabbat ensemble) isn’t even played very carefully, but rather superficial in the strings. Is it too fast for this orchestra?

Overall Duration: 54’01”
Rating (see above for details): 2.6 — The recording may be a technical marvel, but the music is not Berlioz, i.e., misses the spirit of this music. The orchestra tries to show off, but clearly runs into technical limitations. No, I really can’t recommend this.


So much for Amazon user ratings! On, Pierre Boulez received a rating of 4.5 (out of 5), and Paavo Järvi a clean 5.0
Are these people just watching out for spectacular sound?


Berlioz’ Symphonie fantastique was featured in the opening concerts for the 2014/15 season at the Tonhalle in Zurich, with the Tonhalle-Orchestra Zurich and its new conductor, Lionel Bringuier. See my concert review in the posting “Bringuier & Wang in Zurich, 2014-09-11“.


Share:Share on Facebook0Tweet about this on TwitterShare on Tumblr0Share on Google+0Share on LinkedIn0Share on StumbleUpon0Pin on Pinterest0Digg thisShare on Reddit0Email this to someone

Leave a Reply

Translate »
%d bloggers like this: