‘I look upon the invention of the metronome as a welcome means of assuring the performance of my compositions everywhere in the tempi conceived by me, which to my regret have so often been misunderstood.’
- Ludwig van Beethoven
The general public may be surprised to learn that a work as great and popular as Beethoven’s Symphony No.9 has rarely received a performance that realises Beethoven’s stated wishes as to how the music should be played.
Performances of the piece have been dominated for the past hundred years or more by interpretations by great conductors of the Romantic repertoire, who have chosen to disregard the metronome marks Beethoven left for the work. In order to achieve effects of grandeur and significance, they dramatically altered the speed of the music, establishing a tradition of performance that is far removed from what Beethoven seems to have intended.
In recent years several recordings have been released that claim to follow Beethoven’s indications, but which, to varying degrees, offer compromise realisations. This recording is the first to attempt to present the Ninth according to Beethoven’s marked tempi in each section of the work, including two sections that have perplexed musicians and scholars for decades: the Trio section of the second movement and the ‘alia Marcia’ section in the finale, which are played here in the light of recently discovered information.
Over the past twenty years there has been ever increasing attention given to the problem of ‘authenticity’ – the problem, that is, of exactly what we must do if we are to achieve performances of earlier music that are as close as possible to the performances that the composer had in mind and that would have been provided by his contemporaries. The value of the authenticity movement is, of course, not in question: we have all become aware of hitherto unsuspected beauties attainable on obsolete instruments correctly played; our ears have been opened, our sensibilities honed.
However, most discussion of authenticity have failed to make clear the vital distinction between matters of sonority that are largely cosmetic – what instruments are used, how they are placed on the stage, and so on – and the far more fundamental matter of tempo, which is as central to a piece of music as the actual notes to be played. It is, of course, no insignificant matter whether a piece is played on the piano or the harpsichord, on valved or valveless brass instruments, on stringed instruments with steel or gut strings. But the tempo at which a piece is to be played – a question often lumped together with these others in discussions of authentic performance practice – is of a different dimension of significance.
If we hear the opening of Beethoven’s Fifth taken as slowly as it was by such great conductors as Furtwangler and Klemperer, the music speaks with majesty, force, power, ‘Fate knocking on the door’. If we hear it at the tempo indicated both by Beethoven’s Italian direction ‘Allegro molto vivace’ and by his metronome marking ♩= 108, it seems driving, violent, impetuous, headlong, as though a gauntlet were being thrown down – similar to the openings of the String Quartet Op. 95 or the Ghost Trio Op.70 No.l
When Beethoven composed the opening of the Symphony No.5, he must have had some particular ‘meaning’ in mind, he must have heard the music in some particular way. He cannot possibly have heard it both at the Furtwangler-Klemperer tempo and the one he wrote in the score. It is unlikely that he was indifferent about the matter – just as unlikely, in fact, as that he would have been indifferent as to which notes were played. For Beethoven cared so deeply about the tempi at which his works were performed that, according to his friend Anton Schindler, whenever he heard about a performance of one of them, ‘his first question invariably was: “How were the tempi?” Every other consideration seemed to be of secondary importance to him’.
Or take the example of the so-called ‘Moonlight’ Sonata: largely because of its nickname, generations of pianists have played the first movement at a very slow tempo (approximately♩= 52). But once we realise that the name ‘Moonlight’ was not Beethoven’s own but an addition by his publisher, we may take a new look at the evidence of the printed page: it is a Sonata quasi una Fantasia with an Adagio sostenuto tempo (i.e. slow and sustained) There are two beats in a bar 𝇍♩ though, interestingly, in some editions it has been changed to four beats per bar, presumably because the editor thought that that would produce a more suitably contemplative mood for a moonlight-suffused evening. In fact, a closer look reveals that the true motion of the piece is not two but one large, slow beat per bar – as in the motion of the bass – and thus the tempo will be much faster (maybe half-note = 44, or nearly twice as fast as the traditional tempo) and the triplets will be moving, free and rhapsodic, as one might expect in a fantasy.
But, if we feel the opening of the ‘Moonlight’ Sonata one beat to each bar, then, though the triplets now move much faster than we are accustomed to hear them, there is a sense in which the tempo is extremely slow – i.e. whole-note = 22! What we must do then is distinguish between two senses or applications of the term ‘tempo’. The first sense is the one seemingly dictated by the time-signature, and by what has been called the tyranny of the bar-line.
When a person looks at the score of a piece of music, the most obvious means of organising the otherwise seamless (and hence chaotic) horizontal flow of notes is afforded by those regularly placed vertical lines that separate one bar from another. So there is a natural tendency to think of the tempo of a piece as fast if the beats in the bars go by comparatively quickly, and slow if they go by slowly. But to do this is to turn music into what the Germans call ‘Augenmusik’ – music as experienced by the eye.
If, on the other hand, we consider music as experienced by the ear – which is the way it is heard in practice – how fast the notes in the bars go by need have nothing to do with how fast or slow the tempo of the piece feels. For example, a fast trill on a note in an Adagio tempo does not affect the sense of Adagio.
Everything depends on which series of notes is perceived to express the motion of the music and on how many impulses are felt. And if, as we saw to be the case with the ‘Moonlight’ Sonata, the movement is determined by whole-notes, each lasting a bar and registered by the ear as occupying just one beat, then each four bars will be heard as one gigantic four-beat bar and the tempo will feel (and hence will be) very slow. Bar lines are merely visual and intellectual conveniences; often the music flows across them, creating a tempo quite different from that which they seem to stake out so neatly. This is the second sense of the term tempo.
Because Beethoven cared so deeply about such issues of tempo, he left more detailed instructions on the subject than did most other composers. When he headed each movement of his symphonies, and each section of each movement, both with an Italian descriptive phrase (such as ‘Allegro molto vivace’ or ‘Adagio’) and also with a metronome marking, he thought he was leaving future performers not only precise indications of the speeds (and hence the characters) of the various movements and sections, but also the key to the successful realisation of the work as a whole. In a letter to his publishers, Schott and Sons, Beethoven wrote: ‘I have received letters from Berlin informing me that the first performance of the Ninth Symphony was received with enthusiastic applause, which I attribute largely to the metronome markings.’ (My italics)
Yet few performances of the Ninth Symphony have completely followed his instructions as to tempo. In view of Beethoven’s well-known concern about matters of tempo, why should his tempo indications for the Ninth Symphony have been so rarely observed in performance? Received opinion has been that the tempi indicated by Beethoven in his scores are much too fast. There has been much controversy over the validity of his metronome markings and even over the accuracy of his metronome itself. Could the machine have been faulty? Or could Beethoven’s deafness have prevented him from hearing it properly? One answer to the first question is that Beethoven’s letters make it clear that he took great pains to have his metronome in good working order; the answer to the second question is that the metronome had a visible pendulum, so that Beethoven did not need to hear it ticking.
Yet another speculation has been that the ethereal instruments of the inner ear may move more fleetly than those of the real acoustical world. Many musicians have continued to resist the notion that a genius such as Beethoven’s could (or should) be fettered by the ticking of a mere machine. That Beethoven himself had no such apprehensions, however, we can see clearly from a comment published in the Wiener Vaterldndische Blatter of 13 October 1813:
‘I look upon the invention of the metronome as a welcome means of assuring the performance of my compositions everywhere in the tempi conceived by me, which to my regret have so often been misunderstood.’
But, to return to our question: why should the tempo indications of the Ninth have been so rarely observed in performance?
As is well known, Beethoven had an enormous influence upon the Romantic composers and interpreters who followed him. And the Ninth Symphony, more than any other single piece of music, fired the imaginations of the Romantics. ‘The opening of the Ninth Symphony,’ wrote Sir Donald Tovey, ‘has been a radiating point for all subsequent experiments for enlarging the time-scale of music … no later composer has escaped its influence.’ But in all such cross-stylistic transportations, when a work of one age is interpreted according to and, in this case, even plays a significant role in defining the aesthetics of a later age, something of the work’s original spirit is lost. Romantic interpreters, much influenced by Wagner and Liszt, favoured extremes of tempo and frequent, even violent fluctuations between extremes of tempo. They tended to equate slow tempi with profundity – thus the slowing up of ‘the hammer blows of Fate’ at the beginning of Beethoven’s Fifth. Since they saw the beginning of the Ninth in extra-musical, even mystical terms, as an evocation of primeval chaos or formlessness, the Romantics usually played it so slowly that it became separated from the rest of the movement. Wilhelm Furtwangler, one of the most revered post-Romantic interpreters of the Ninth, took the Adagio third movement at less than half the speed indicated by Beethoven, in order to attain a mood of profundity and spirituality! Yet, at the other extreme, in a memorable 1951 performance of the Ninth, Furtwangler urged the concluding music of the finale into a frenzy of religious ecstasy by taking a tempo 70 points faster on the metronome than the one indicated by Beethoven!
These interpretative decisions have come down to later generations, often in somewhat modified form, as powerful performance traditions that the present-day performer defies at some risk. In the past few years, however, some conductors prominent in the ‘authenticity’ movement, have taken that risk and have performed and recorded Beethoven’s symphonies at, or close to, the tempi he indicated. And predictably they have, while causing much excitement in many quarters, drawn the fire of many musicians and conservative critics. Samuel Lipman, for example, commenting on some ‘authentic’ Beethoven performances, speaks scathingly of tempi ‘proudly fulfilling Beethoven’s metronome markings,’ that ‘seem most strikingly to be chosen from a range restricted to fast, faster and fastest,’ resulting in ‘a lack of breadth and space’, and of performances founded, ‘in a way Beethoven could hardly have desired or even imagined on the tick-tock of the metronome’…‘Because the music does not breathe, this quintessentially passionate music conveys no passion.’
There is a danger here for performers who are bent on realising Beethoven’s professed intentions in regard to the tempi which none of us has escaped entirely. Tempo is, after all, not an end in itself but a medium that allows different expressive forms, just as water allows for coral reefs, fish and anemones, and air makes possible pine forests, deer and human beings. Those that inhabit these elements do not notice water or air: the tempo is never the subject of a successful performance. Perhaps, if we can hear this music free of the bar lines, fidelity to the metronomic indications need not necessarily result in a sense of mechanical regularity or a lack of breathing space or passion. It all depends, after all, on what is done with and within the chosen tempo.
What then can one learn about the first movement of the Ninth by playing it at the tempo indicated by Beethoven’s metronome marking – a tempo almost never observed in performance? At first the Italian verbal description ‘Allegro non troppo’, un poco maestoso (‘Fast but not too much so, somewhat majestically’) seems in direct conflict with the metronome marking ♩= 88, which is a good 20 points faster on the metronome than most performances of the opening and as much as 40 points faster than the slowest performances. Even Roger Norrington’s ground-breaking 1987 recording and the recording of Charles Mackeras from 1990 fall 10-12 points short of Beethoven’s metronome marking in this movement.
But ♩= 88 is not an unusually fast marking for an ‘Allegro ma non troppo’ in Beethoven. The last movement of the so-called ‘Archduke’ Trio (Op.97), which is also marked ‘Allegro ma non troppo’, 2/4 is invariably taken that fast or even faster. Also, while no conductor on record has taken the opening of the Ninth at = 88, many conductors take the fugato section that begins at bar 218  at that tempo. Furtwangler, for example, pushes the tempo up past 90 at this point, as does Toscanini. Thus, when people say that Beethoven’s marked tempo for the first movement of the Ninth is ‘too fast’, what they may really be saying is: ‘It seems too fast for the opening although it is fine for the fugato section.’
Of course, no sensible interpreter would insist on there being no tempo deviation between the sections of a classical piece or movement, but deviations as wide as those often heard between the opening of the Ninth and the first movement fugato section – deviations of 20 to 40 points on the metronome – can hardly be considered as belonging to the same tempo category.
The Romantics thought of the opening of the Ninth as representing the awakening of the primordial forces of the universe. This is why it’had so powerful an influence on Romantic composers such as Bruckner, who began his fourth, seventh, eighth and ninth symphonies with clear allusions to the opening of Beethoven’s Ninth. But there is a profound difference between the beginnings of those symphonies and the model that inspired them.
In each Bruckner instance, the strings play an amorphous, rhythmless tremolando, whereas
Beethoven gives the second violins and cellos a precisely measured sextuplet rhythm. (Interestingly enough, Furtwangler always insisted that his string-players play a rhythmless tremolando in the opening of the Ninth rather than Beethoven’s sextuplets.) Moreover, the sextuplet rhythm recurs at several points in the movement – most strikingly in the second violins at bar 240, about 20 bars (or 30 seconds) into the fugato section , lending validity to the conviction that Beethoven conceived the entire movement as being in one and the same tempo.
Recall our earlier distinction between the different applicable meanings of the term ‘tempo’. If we cease to succumb to the tyranny of the bar-line, marking off the quarter notes at 88 per minute and instead start to hear the music first in one-beat bars (i.e. half-note = 44) and then in still slower one-beat two-bar ‘bars’ (i.e. whole-note = 22), it has all the calm, timelessly suspended mystery we could want.
Moreover, when the violins enter at the end of the second bar with their descending motive made up of pairs of notes – short-long, short-long, short-long -the music sounds like a majestic French Overture if we hear it one beat to each bar. Thus the ‘poco maestoso’ effect that Beethoven asked for is obtained not by taking the music at a slower tempo in the within-the-bar sense, i.e. by adopting a slower metronome marking, but rather by taking it as a fast enough metronome marking so that the music can flow across the bar-lines, each bar (or even each two bars) being heard as one beat.
To make this concept fully clear, it may be helpful to actually conduct along with the music, beating each bar, or even every other bar (i.e. whole-note = 22 instead of each quarter note = 88). It turns out therefore that in the first movement of the Ninth there is no conflict between Beethoven’s Italian direction ‘Allegro ma non troppo, un poco maestoso’ and his metronome marking ♩= 88. (NB: Beethoven could not have marked it half-note = 44 even if he had wanted to because 50 was the lowest number on his metronome.)
The latter specifies the within-the-bar tempo while the former evokes the effect that the music makes when heard across the barlines. This difference in function of the two tempo indications, the Italian verbal description and the metronome marking, is frequently found in Beethoven’s music. The two indications are used to indicate tempo in two different senses of that term.
Here we have the rhythmic secret not only of the first movement of the Ninth but also of much of Beethoven’s other music. By taking the ‘fast’ tempo indicated by the metronome marking but then reducing or even eliminating the impulse at the beginning of each bar, one can get the music to sound as ‘slow’ as the Italian direction suggests. This vast, slow, spacious motion-one beat or impulse per bar or even one every two bars – is attainable not only at the movement’s beginning but also at the beginnings of the development , the recapitulation , and the coda . At other times the music can be given a gentle Allegretto feel, two beats to the bar in the calm and joyful second theme , for example, or the transition theme  with its ecstatically rising thirds. And at still other times the music can take on an overwrought obsessive impetuosity, four beats to the bar, owing to the sheer speed of the notes, as, for example, at bar 132  or in the ferociously powerful fugato section , which, as I mentioned earlier, is taken at Beethoven’s metronome marking by so many conductors, regardless of how slowly they may have begun the movement.
In contrast to the opening tempo of the first movement, that of the second movement Scherzo has presented little problem to interpreters of the Ninth. Conductors as different as Furtwanagler and Toscanini take it at exactly half-note = 116, just as marked. No one ever deviates more than a few points on the metronome from this tempo. (It is surely no coincidence that the Scherzo of the ‘Eroica’ Symphony, in many ways so similar to that of the Ninth, is also marked ‘Molto vivace’, half-note =116.) Moreover, whenever there is a slight deviation from Beethoven’s metronome marking in this Scherzo, it is almost always an increase in speed, an interesting fact when we recall that most people profess to consider Beethoven’s metronome marking too fast for this music.
However, while the Scherzo of the second movement meets with general agreement, the tempo of the Trio is one of the most hotly debated in Beethoven’s entire output. This section and the ‘alia Marcia’ section of the finale of the Ninth have vexed both scholars and performers for generations.
Beethoven marks the 3/4 Scherzo ‘Molto vivace’ dotted half-note = 116, and the Trio 2/2 (i.e. 𝇍)’Presto’ half-note = 116). In other words the two sections of the movement are to be played at the same tempo in the sense that the basic beat of each section – a dotted half-note for the one-beat-to-a-bar 3/4 of the Scherzo and a half-note for the two beats-to-a-bar 2/2 of the Trio will register 116 times per minute. But the two sections are joined by an eight-bar passage marked ‘stringendo il tempo’ (‘accelerate the tempo’). What sense can we make of an injunction to get faster that joins two sections that are to be played at the same speed?
It has often been pointed out, moreover, that half-note =116 is absurdly slow for a ‘Presto’. Stravinsky, that most astute of music critics, was quite unequivocal about the matter: ‘It is always wrongly played. A measure of duple time should approximate a measure of triple time. And if this were not already obvious from the stringendo lead-in, it would be from the ‘Presto’, which is a more reliable marking than the metronome and which, unlike it, could not be a misprint. Clearly the relationship is roughly the same as the one obtaining between the duple and triple meters in the scherzo of the Eroica.’ [Themes and Conclusions, p. 168.]
Stravinsky is therefore suggesting that since half-note =116 is obviously too slow a tempo for a ‘Presto’, and since Beethoven marks his lead-in passage ‘stringendo’, we should take the Trio not at half-note =116 but rather at whole-note =116, i.e. just twice as fast as Beethoven apparently directs.
As it happens, evidence has recently come to light that offers convincing substantiation for what Stravinsky, acting purely as an intuitive musician rather than a scholar, suggested. The metronome markings for the Ninth, as they appear in published scores, are based on a letter of October 1826 sent by Beethoven to his publisher, Schott and Sons. In that letter the metronome marking for the Presto Trio of the second movement is quite clearly half-note =116. But we now know that this letter, while signed by Beethoven, was actually written by his nephew Karl, and Karl was getting his information in part from a conversation book in which Beethoven had written his desired metronome markings. In the conversation book the tempo for the Trio reads as follows:
116 half-note half-note 𝄀 Not one half-note but two, and followed by a bar-line. This appears to mean that the metronome marking 116 is to apply not simply to a half-note but to a whole bar exactly as Stravinsky had maintained!
It is, of course, not surprising that Karl, who was not a professional musician and who had had little experience with the metronome, should have made this error in transcription. Now that we realise it was indeed an error, we can not only understand the Italian direction ‘Presto’ – for whole-note = 116, unlike half-note = 116, is a genuine ‘Presto’ – we can also see why Beethoven should have marked the eight-bar lead-in passage ‘stringendo’.  For now the four quarter notes that make up a bar of the Trio take up exactly the same amount of time, viz., 1/116 of a minute, as the three quarter notes that make up a bar of the Scherzo. What the term ‘stringendo’ means, therefore is simply that the quarter note will have to come faster if we are to get four into the time-space formerly occupied by three.
This recording at last gives listeners a chance to judge for themselves whether or not this tempo of whole-note = 116 feels right for the Trio .
It produces music that is far removed from the gentle, bucolic mood of traditional performances. This is a true ‘Presto’, not only extremely fast, but actually at the borderline of payability – electrifying and, I believe, completely Beethovenian in spirit.
Incidentally, another great late Beethoven piece, the Scherzo of the C# minor String Quartet (Op. 131), is not only similar in character but also has the same Italian tempo direction ‘Presto, alia breve’, and is usually performed at approximately Whole-note = 116.
It is also of interest that Beethoven originally wrote not ‘Presto’ but, ‘Prestissimo’ for the Trio of the second movement of the Ninth. The word is still faintly visible-in the manuscript, and if he had allowed it to stand, the tempo used in this recording surely would have been accepted as the correct one. How close Beethoven came to resolving this issue forever!
The third movement of the Ninth Symphony offers the quintessential example of a controversial Beethoven metronome marking, and is the section of the work most often cited to prove that Beethoven’s metronome is a false god. The Italian direction for the movement is ‘Adagio molto’, a very slow tempo, but the metronome marking is ♩= 60, which is moderately fast, certainly at least an Andante rather than an Adagio – let alone an ‘Adagio molto’. Most performers, in their eagerness to create an emotional effect that answers to the direction ‘Adagio molto’, have simply disregarded the metronome marking and opted for a much slower tempo. Romantic conductors like Furtwangler and Bernstein in fact took this movement at half the speed indicated, creating a mood of reverent, ineffable stillness.
One might be inclined to give in to the Romantics, were it not for a singular coincidence: of the eight Adagio movements in 4/4 that Beethoven wrote, only one, the second movement of the String Quartet Op.59, No.2, has a metronome marking, and it too is♩=60! Can Beethoven have made the same mistake twice? A quick comparison of the themes of the two Adagio movements reveals that they are strikingly similar: both are chorale types in minim motion.
Suddenly everything becomes clear: both movements are to be felt and heard as being in half-notes, not quarter note, i.e. at half-note = 30, which is a perfectly fine Adagio tempo but a marking too slow to appear on either Beethoven’s metronome or those of our own day. (This is surely the reason Beethoven marked the movement with the quarter note indication, but how much easier for us as interpreters – and for the case of the metronome – had Beethoven indicated ’30’ in parenthesis or marked the movement in cut time 𝇍 as he did in the slow movement of the Piano Concerto No. 1)
If we listen to the opening phrase of the violins  and feel not the four impulses per bar that the metronome marking ♩= 60 leads us to expect but rather only two or even just one or (as the slur suggests) one every two bars, then the slow spaciousness appropriate to an Adagio molto is perfectly captured. Instead of a static, quasi-hypnotic trance, we have a buoyant set of variations full of grace and delicacy, consoling and human because eminently singable. It is my-greatest hope that this way of reconceiving the third movement, not at all difficult to grasp intellectually but far from easy to realise in practice, comes through clearly enough in the recording that listeners can experience it for themselves. For the whole shape of the Ninth Symphony is thereby altered. If the slow movement takes nearly half as long as before, it will obviously play a very different role in the work’s overall structure.
Another remarkable result of this new conception of the third movement’s initial Adagio is the light that it sheds on the Andante moderate, ♩= 63.  It has often been pointed out how absurd it is to have, back to back, two sections, one of which is marked very slow, ♩= 60, and the other is marked ‘moving moderately’,♩= 63! But we can now see what Beethoven intended. The ‘Adagio molto’ section is to be heard in a very slow two beats to the bar or, still better, one to the bar or even occasionally, one impulse to each two bars! At the point of change to ‘Andante moderato’, the music moves imperceptibly into triple time. Suddenly what was slow and timeless in the earlier pulse-frame becomes transformed into a lilting three to the bar. Moreover, Beethoven heightens this magical shift of mood by pushing the tempo up very slightly, from ♩= 60 to ♩= 63, a scarcely noticeable change, but one that helps to accentuate the lilt of the new 3/4 meter. Such was the subtlety of mind of a composer so often accused of confusion and bluntness of sensibility in the area of tempo notation.
But what are we to make of the 12/8 variation?  Did Beethoven really intend that the same tempo should prevail even through this section of the piece? This involves a complete transformation of our perception of this whole section. In traditional slow performances, the sixteenth note figuration in the first violins is treated as the principal melody. But at Beethoven’s indicated tempo the real melody (the main theme in half-notes, Ex. 2) can be clearly heard in the winds behind the first violins’ filigree decoration. As this theme, so simply presented by the violins at the beginning of the movement, becomes more confident, bravura embellishment, as in a baroque aria, leads to an ecstatic, virtuosic display. Perhaps Beethoven intended that the tempo should slacken when the notes became so fast, in order to maintain the calm mood, but it is also possible that he cared more about the structural connection of the different sections of this set of variations. We will never know, but this performance gives us a chance to hear how it might sound.
For those who still find it hard to accept what may seem a distressingly fast tempo for the 12/8 variation, it may be helpful to recall the finale of the ‘Pastorale’ Symphony, with its very similar, gently undulating semiquaver figurations in the violins and its melody in the winds. The metronome marking for that movement is identical to that of the third movement of the Ninth,♩= 60!
If this does not seem an apt analogy, it is probably because of the widely differing associations that most of us have with the two symphonies, and the two movements, in question. The ‘Pastorale’ Symphony deals with light and pleasing matters: arriving in the country, sitting by a brook, watching peasants dance, getting caught in a storm, and (in the finale) hearing the shepherds’ song of thanksgiving after the storm. The Ninth, on the other hand, as we all know, is a profound work of Beethoven’s last period. Since slow movements are considered by some to be more profound than fast ones, the most profound movement in this work will be its slow third movement. While a relatively quick and gracious tune in the winds accompanied by sixteenth note figurations in the violins may be perfectly appropriate to a shepherd’ song of thanksgiving, it surely, so the thinking goes, has no place in the rapt, intense meditation that the slow movement of the Ninth must be. But to think this way is to make the same mistake, on a slightly higher intellectual level, as pianists who seize upon the word ‘moonlight’ in the nickname of Beethoven’s sonata, imagine the languorous romance of a moonlit night, and choose their tempo accordingly. We must learn to trust Beethoven more.
There is no argument amongst musicians about the tempo for the opening of the fourth and final movement, a passage that has sometimes been referred to as a ‘cry of terror’ . Indeed, it is hard to imagine it at any other tempo than the Presto dotted half-note = 66 that Beethoven assigns to it. But there has been considerable disagreement about the cello and doublebass solo passages starting at bar eight. Beethoven indicates no change in tempo: on the contrary, he asks that the passage be played ‘in the character of a recitative, but in tempo’, i.e. Presto. Each of the quotations from earlier movements -from the first movement , the second , and the third  – is given the same tempo indication as the corresponding passage at its first appearance. It may, however, be worth pointing out that many conductors take the third movement citation considerably faster than they did when it appeared in its original context. Once again, we see the effect of the powerful associations that can lead a performer to equate late Beethoven with profundity and slowness. These associations do not, however, seem so pressing in the context of the finale, with its initial ‘cry of terror’. The ‘Freude’ theme  is marked half-note = 80, which is a good deal faster than we have grown accustomed to hearing it. But when taken at that tempo, relieved of its usual burden of Romantic significance, it regains an ecstatic flow and confidence wholly appropriate to its text and context.
The most controversial section of the fourth movement and, along with the Trio of the second movement, one of the two most controversial sections in the whole of Beethoven’s ceuvre, is the notorious ‘alia Marcia’ section . It is marked ‘Allegro assai vivace’ (‘fast, rather lively’) dotted quarter-note = 84, which gives us another apparent conflict between Italian verbal direction and metronome marking.
For = 84 is ludicrously slow for an ‘Allegro assai vivace’. But once again the fault almost certainly lies with Beethoven’s nephew Karl. Karl’s mistake here, however, is slightly different from his earlier one. For in this case what Beethoven wrote in the conversation book was simply “84 6/8.” He did not draw a note of any kind. Karl translated this admittedly rather cryptic notation into dotted quarter-note = 84. But what Beethoven probably meant was that a whole bar of 6/8 would register 84 times per minute. Thus the correct translation of what he wrote in the conversation book would not be dotted quarter-note = 84 but rather dotted half-note = 84, a metronome marking that yields an exhilaratingly fast but perfectly playable tempo that is a true ‘Allegro assai vivace’.
Moreover, the ecstatically heroic text set by the music of this section is well fitted to a very fast tempo: ‘Joyously as his suns fly across heaven’s magnificent expanse, brothers, run your race, joyously as a hero [runs] to victory.’ At the ultra-slow tempo = dotted crotchet = 84, these words seem comical, even grotesque. Also, it is clearly impossible to maintain the slow tempo throughout the fugato section without having the music plod ridiculously. Almost all conductors, therefore, speed up a good deal. If, however, the whole passage is taken at dotted half-note =84, there is no need to increase the tempo (and, indeed, no possibility of doing so) .
The moment at which the faster tempo is revealed as blazingly, irrefutably right is the overwhelming moment at which the chorus enters with the ‘Freude’ theme, now just slightly faster than when it was first heard (dotted half-note = 84 as against the earlier half-note = 80) but with far greater excitement and with the whole string section in pursuit . At the traditional tempo for this section, an arbitrary dotted crotchet = 120, the full effect of this ecstatic moment is lost and the structural relationship between the two appearances of the ‘Freude’ theme destroyed. Moreover, when the ‘Freude’ theme next enters, in the great double fugue in 6/4 section , it is marked at precisely dotted minim = 84! (another tempo usually performed faster than Beethoven’s marked tempo). This is no coincidence but further evidence of Beethoven’s uncanny sense of tempo relationships that plays a crucial role in binding together the disparate elements of the great set of variations that make up the finale of the Ninth.
Of course, none of us will ever know with absolute certainty exactly how Beethoven wanted these two most controversial sections of his Ninth, the Trio of the second movement and the ‘alia Marcia’ of the finale, to sound. But this recording gives listeners for the first time the opportunity to hear how they sound at Beethoven’s apparently intended tempi. Certainly these tempi accord with the English organist George Smart’s report that Beethoven emphatically insisted to him that the entire symphony should take no more than 45 minutes.
At Beethoven’s tempo, ‘Andante maestoso’ half-note = 72, the choral section ‘Seid umschlungen, Millionen’ (Embrace, oh ye millions)  emerges as a noble and compelling expression of universal love. At the slower tempo of Romantic tradition we have grown accustomed to hearing it not as loving but rather as a heavy-laden, emphatic call to arms. On the other hand, the charmingly elegant, almost Mozartian ‘Allegro ma non tanto’ section that follows  presents no problems at the half-note = 120 tempo that Beethoven prescribes, and it has, therefore, traditionally been taken at that tempo.
The work’s one remaining controversial section is the final Prestissimo . When taken at the steady half-note = 132 that Beethoven indicates, it may seem tame in comparison to the wild rush to the end that we have come to expect. As usual, conductors have opted for the Italian verbal direction (‘Prestissimo’) and have ignored the seemingly contradictory metronome marking. But when this section is taken at the usual half-note =152 or even faster (Furtwangler clocks in at an unbelievable = 200!) audiences find it difficult to decipher the words and orchestral players have trouble articulating the notes clearly. At half-note = 132 not only are both words and notes perfectly clear, but the magnificent triplet figure in the timpani in the final bars is also perceived as a genuine rhythm and, indeed, an entirely new rhythm – introduced by Beethoven only at the very last moment – rather than as mere noise.
Most important of all, the final section of the Ninth is, like so many of its other sections, a march, a fact that is completely obscured by too fast a tempo. At half-note = 132 it is, to be sure, a Prestissimo march that moves at the very fastest pace a human marcher could reasonably be expected to maintain. But it is a real march nonetheless. Beethoven was writing of love, of friendship and joy; for him the march was a natural mode in which to express his message, and it makes a fitting conclusion, transcendently joyous and yet firmly earthbound, to his grandest testimony to the brotherhood of man.