Support us financially by purchasing from

Ludwig van BEETHOVEN (1770-1827)
Symphony No 3 in E-flat major Op 55 ‘Eroica’ (1804) [46:53]
Etienne-Nicolas MÉHUL (1763-1817)
Les Amazones, ou La Fondation de Thèbes: Overture (1811) [7:19]
Les Siècles/François-Xavier Roth
rec. March 2020, MC2, Grenoble, France; September 2020, Théâtre municipal, Tourcoing France and Maison de l’ONDIF, Alfortville France (Beethoven); La Seine Musicale, February 2020, Boulogne-Billancourt, France (Méhul).
Reviewed as a digital download with pdf booklet from eclassical.com (24-bit initially at same price as 16-bit)
HARMONIA MUNDI HMM902421 [54:12]

It has been a terrific twelve months for iconoclasm here on MusicWeb. It began last year with Bohdan Syroyid’s review of Bernhard Ruchti’s À Tempo project and his latest disc of Schumann piano music. In a nutshell, Ruchti’s compelling argument is that we have been reading Schumann’s metronome markings wrong and as a result playing him much too fast. At first, I thought this idea grabbed me because I much prefer playing Schumann on the piano much slower than any recording and that that was a just an excuse  for my technical limitations at the keyboard! But listening to Ruchti’s performance of the Schumann Fantasy in C, I heard a great masterpiece made even greater. The implications extend far beyond Schumann and, of relevance to this review, touch on the vexed issue of tempi in Beethoven.

The second blow for iconoclasm was delivered in April 2021 in Chris Salocks’ thorough and provocative review of the Faust, Queyras, Melnikov and Heras-Casado recording of the Beethoven Triple Concerto. Salocks limited himself to the thorny subject of vibrato and to the issue of using fortepianos in this repertoire. Apart from regular harrumphing from Pinchas Zukerman, which mostly makes him come across as a bit of an old fart, I haven’t heard any serious discussion about the issue of what used to be called Period Instrument performances since the 80s. My recollection is that this particular culture war was quite bitter but that for decades now it has largely been resolved in favour of what we now call Historically Informed Performance or HIP.

My issue with this whole business is somewhat different from Salocks’ but I do have a great deal of sympathy with what he says. It seems to me that a lot of performance practice in the area of the classical era and early Romantic period has become rather homogenised. The worst culprits in this regard seem to me to be conventional orchestras attempting period style rather than specialist HiP ensembles. By way of illustration, I would cite the way every accented note has to involve an explosion, especially if underlined by the timpani. Every phrase must feature an identical emotional bulge. Every tempo must be fast.

Taking these in order: massively exaggerating every accent robs the sforzandi, so crucial in Beethoven of any real meaning as well as disrupting the rhythm of the piece. I am of the view that if Beethoven wants to make a point, he will write it so. But it isn’t just in Beethoven that this interpretative tic has become virtually universal. I found the recent reissue of the Mackerras recordings of the Mozart symphonies rather marred by this effect, good though it was overall.

Equally I have no doubt that “bulging” phrases are historically accurate but am I really supposed to believe that every single melodic phrase in Haydn, Mozart and Beethoven is meant to be identical? Such identikit phrasing seems to me to smack of laziness rather than historical research.

As for tempi, I have a sneaking feeling that a whole generation of recordings of Beethoven’s 9^th will be recognised as having been blighted by an excessively quick tempo for the slow movement. Critics fall over themselves to praise such performances as up to date but never seem to address the crucial issue of the relationship between the two main tempi in that movement. The second tempo is clearly indicated as faster than that of the opening. Most recordings adopting a faster tempo for the opening make very little distinction between the two speeds for the simple reason that to go faster still upon reaching the andante section reduces the music to a gabble no-one seems prepared to contemplate. As a result, the contrast Beethoven was clearly looking for is lost. Abbado’s 2012 recording with the Berlin Philharmonic is just the most prominent of many.

The early zealots of the Period Instrument movement seemed of the view that historical research could replace interpretation. Of course, it is not at all clear to me that this is not, in fact, a gross distortion of their view put out by their opponents. Harnoncourt, as one of the leading lights of that movement, never held this view. I have immense affection for Harnoncourt’s set of the Haydn Paris symphonies but I don’t really think I am listening to them as performed in Paris all those years ago. I am listening to a distinctly early 21^st century way of playing Haydn, and an outrageous over the top pleasure it is, too.

The danger is that, like a kind of weird sect, we fall into the trap of thinking that anyone can get at the real Beethoven or Haydn or whoever. In relation to ubiquitous fast tempi, I often find myself thinking of the issue of the acoustic in the Thomaskirche in Leipzig. The church, where the bulk of Bach’s religious music was first performed, possesses an echo that precludes the now standard extremely rapid speeds. My point is that musicians tend to be practical people and what will work in one hall may not work in another.

I have nothing against fast tempi or expressive bulges. It is their homogenised application that troubles me. Take Fritz Kreisler’s 1926 recording of the Beethoven violin concerto and marvel at the wide range of vibrato he deploys and the extraordinary taste and subtlety with which he uses it. Contrast this with the blanket application of vibrato in the string sections of large orchestras by the 1970s. If nothing else, HiP has shaken up how we think about such issues. A consummate artist like Isabelle Faust varies her tone for deep musical reasons, not because of a prevailing dogma. In this she stands in a line that includes Kreisler, Huberman, Milstein, Menuhin and basically any great violinist worth listening to.

The point is that all of these are creative decisions, not matters of dogma. If Andras Schiff chooses to record Beethoven or Schubert on a fortepiano, it is because it makes an artistic point that is audible to me, the listener, and adds something to my appreciation of Beethoven or whomever. Heras-Casado’s recent recording of the Beethoven 9^th (HMM90243132, with Choral Fantasia) seemed to me to provide a wonderful solution to interpretative problems that have always dogged that work’s opening movement. It did so by the use of a faster than usual speed and the more transparent textures of a period band. This doesn’t make this the only way to play this music, and I will not be throwing out my Furtwangler CDs any time soon, but it did add something to my knowledge of the work, which is what the art of the interpreter is all about.

This brings me by an extremely circuitous route to this new recording of the Eroica. Roth, of course, conducts both regular orchestras and specialist outfits like Les Siècles, recorded here. To begin with tempi, I can say with some relief that there isn’t anything excessive here in any way. The opening movement is on the fast side but so are those new kids on the block, Toscanini and Karajan! Indeed, with bulges and explosions kept to a minimum, this is a performance that sounds frankly quite conventional.

The nature of the Eroica, with lots of motor rhythms and busy passage-work rather than long melodic lines, means that the issue of vibrato is less obtrusive. The exception is the Funeral March which highlights another interesting trend in contemporary HiP. Increasingly, performers are exploiting the sometimes rough and ready nature of period instruments, particularly Salocks’ beloved vibrato-less strings, into ever more modernist-sounding territory. This is a case in point. The Eroica still sounds, in places, like an astonishingly forward-looking work and nowhere more so than in the slow movement. The proto expressionism Roth finds in it was always there, but period instruments allow him to really push that aspect of the music in a way that the plush sounds of a conventional orchestra cannot. Let’s not pretend that this is a historical reconstruction – this is one of the distinctive sounds of the early 21^st not the early 19^th century. Much more to the point, it is thrillingly intense. In its own way, however, it is as interventionist as Furtwängler.

If this is the case, how does Roth’s interpretation stand up? This is as fine an Eroica as I have heard in a long time. Roth really understands what makes the symphony tick. He knows, for example, that there is no point weighing the finale down with portentousness in an attempt to match the weight of the first two movements. Instead, he follows Toscanini and Furtwangler and goes for broke.

The Covid-decimated celebrations of Beethoven’s 250^th birthday have been very good for Beethoven’s lesser-known contemporaries. It seems the fashionable thing to do is to shoehorn in one of their works to accompany a Beethoven big beast. To my ears the function of these works is to demonstrate why there is only one Beethoven. The Méhul included here is a pleasant enough piece that looks forward to Mendelssohn without ever really being distinctive. The performers are clearly much more convinced by it than I was.

I have deliberately left my own personal favourite Eroica to the end of this review. In a way, I hope it illustrates the irrelevance of a dogmatic approach to any music. It is the recording by Fran Brüggen and the Orchestra of the 18^th Century and, as a result, it is HiP to the tips of its toes (Glossa GCD921116, Symphonies 1-9, 5 SACDs). That isn’t the point. Brüggen’s grand dark vision of the opening movement opened my ears to hearing music I thought I knew back to front in a new and deeply stimulating way. By comparison with it, Roth is bright, clean and contemporary in the best sense but only in the slow movement does he really access the deeper layers of this music. For others, bored of more subjective interpretations, a lighter hand on the tiller may be extremely attractive.

I won’t be replacing Klemperer, Furtwängler or especially Brüggen with this new recording because ultimately I think the HIP argument is a bit of a red herring – what matters is whether a performance has something to say and no amount of historical research into performance practice can replace that. It is noteworthy that Brüggen’s tempo for the first movement is significantly slower than Roth’s. One fast speed fits all is clearly not as historical as we have been led to believe!

David McDade


Previous review: John Quinn