oboe by Howarth
Back to The Oboe 1903-53 CD details   Back to Links and Surprises
Recordings to celebrate the world of the oboe


The Oboist as the Custodian of Orchestral Pitch

A discussion between Geoffrey Burgess and James Brown, concerning the history of the Oboe “A” and its fluctuating
pitch in the years that followed 1800, together with some ongoing problems of pitch in the Recording Era.


J.B.
Having been an orchestral musician all my musical career, I was intrigued by what you wrote in your remarkable book The Premier Oboist of Europe: A Portrait of Gustave Vogt about the origins of what is one of our important responsibilities: tuning the orchestra.

watercolour of Gustave Vogt (BNF) GB:
It was exciting to discover those documents in Paris, and I still haven’t found evidence of earlier references to the oboe tuning the orchestra. This is a précis of the passage from the book:
The earliest explanation for the practice of using the oboe as orchestral pitch reference, appears to be a letter from François Sallantin (principal oboist 1770-1816) to the administration of the Paris Opéra dated 1802, in which the oboist complained of having to arrive at the theatre early in order to give the A to the first violinist, who then made the rounds of the other players before the orchestra entered the pit. Sallantin’s successor, Gustave Vogt, explained in his Méthode (c.1816-1825) that the oboe was the logical choice because it warms up more readily and stays longer at a constant pitch than the other woodwinds. Later in 1833 Fétis also pointed out that “the most common practice is to take the A from the oboe, because this instrument is less impressionable to variations in temperature than the flute or clarinet,” but he added that considerable variation was still possible due to irregularities amongst reeds and temperature fluctuations. In A travers chants (1859) Berlioz reported that occasionally the flute was used in the place of the oboe to give the orchestral pitch. This does not seem to have been the standard practice in Paris, but may have been a custom that Berlioz met on his travels.

JB:
What pitch would Sallantin and Vogt have given to the Opéra orchestra?

GB:
‘Pitch’? You mean ‘pitches’! The answer is not straightforward. Over the course of the nineteenth century, the establishment of a pitch standard was a fiercely debated topic. The copious writings arguing for a standard only emphasize the reality: namely, that pitch was far from fixed.
Throughout the eighteenth and nineteenth centuries, the general trend was for pitch to rise. In some centres it shot up rapidly, while in others the climb was more gradual. Even in the one city pitches could vary. This was no truer than in the case of Paris where each of the theatre orchestras, the Conservatoire, the churches and military bands maintained a slightly different pitch level. This could restrict the movement of musicians amongst the different organizations. In the eighteenth century the Opéra more than any other musical organization in Paris resisted the upwards trend in pitch, and at the beginning of the nineteenth century it still played lower than other Parisian theatres.
A tuning fork owned by Pascal Taskin dated 1783 and said to be tuned to Sallantin’s oboe, gives the pitch A=409Hz. In the last years of the eighteenth century, a standard was set by the Conservatoire called ton d’orchestre which was a semitone higher than the pitch at the Opéra. If this was based on Opéra pitch at c.A=409, this would place the ton d’orchestre at A=435. Meanwhile, by 1811 the pitch of the Opéra had risen to around A=425. This is almost exactly the pitch of a pair of pitch pipes made by the pre-eminent Parisian wind-instrument maker Delusse, who was active in the last decades of the eighteenth century and who supplied instruments to the most important oboists including Sallantin. So it seems reasonable to assume that Delusse’s oboes could be played at 425Hz.

Oboe by Delusse, formerly owned by Gustave Vogt, with three alternate top joints (Paris: Musee de la Musique E387/C480) GB:
But what of the difference between the pitches at the Opéra and Conservatoire?

JB:
At the time, the oboists at the Opéra and the professors at the Conservatoire were one and the same. The rise in pitch at the Opéra resulted in a rapprochement, but around 1810 there would still have been a difference of about 10Hz. Delusse customarily supplied oboes with three alternate top joints, each approximately 5Hz apart in pitch, thus accommodating the two standards 10Hz apart. His oboes continued to be used into the 1820s. In addition to the written record of oboists like Brod and Vény from the 1820s praising them, a number of Delusse oboes survive with keywork added in the 1820s, equipping them to play more modern music. It comes as no surprise that the oboes and joints chosen for this treatment are amongst the shortest surviving from Delusse.
In 1824 the upwards rise in pitch at the Opéra was temporarily checked under exceptional circumstances. At that time the pitch was A=432Hz. Several singers, including the famous soprano Alexandrine-Caroline Branchu (1780-1850), protested that the rising pitch was affecting their voices and insisted that it be lowered. A committee was assembled to review the situation, and the decision was taken to lower the pitch to around 415, not far from Taskin’s fork.

JB:
Were the wind players expected to purchase new instruments who made the instruments for the oboe section?

GB:
Fortunately there was a precedent. In 1821 the Opéra had agreed to foot the bill for expenses incurred for the upkeep of instruments at Opéra pitch, and again on this occasion, the administration agreed to purchase new instruments for the wind section, at a total cost of 6,092 F. They may not have been “new” oboes at all: in some instances the oboists probably got by with the longer top joints of their valued Delusse oboes. This lower pitch was maintained only for five years until Mme Branchu retired in 1829, at which point A=432 was resumed.
In the 1830s, pitch at the Opéra climbed to A=440, and rose gradually still higher until 1854, where it was reported to have reached at least A=446. But this was well after Sallantin and Vogt had retired. The commission responsible for lowering the pitch in 1826 had been concerned by the effect that the sudden change of pitch would have on audiences, and also blamed the wind players for the rising pitch. The younger players were the likeliest ‘offenders,’ as they would have been conditioned to the sharper Conservatoire pitch, and were more likely have new instruments built to sharper pitch. During the 1840s and 1850s when the pitch of the Opéra rose at least another 5Hz to A=446, younger pupils of Vogt — August Bruyant, Charles Colin and Louis-Stanislas-Xavier Verroust — comprised the oboe section.
The pitches at the other Parisian theatres were close enough not to prevent circulation of players. Not surprisingly, orchestras with the most overlap of personnel — such as the Opéra and the Société des Concerts du Conservatoire — had the most similar pitch levels.

JB:
What of instrumentalists required to stand in when the regular players were ill or on leave?

GB:
When Vogt and Brod took leave from the Opéra in 1828 and 1829, they were replaced by Jean-Jacques Fouquet, the first oboe of the Théâtre Italien, where pitch was within 1Hz of the Opéra and therefore well within the tolerance of a single player and instrument.

photo of oboist James Brown
JB:
Differences were obviously greater outside the same centre. The lack of an international pitch standard made it hard for travelling virtuosi. I discovered a number of instances where soloists were not heard to their best advantage in foreign centres where their oboes would not accommodate local pitch.

GB:
Those crtics’ reports in your article The Musical Press and the Nineteenth Century Oboist, at Home and on Tour make fascinating reading, and show that pitch was a pressing concern. Again having a number of different top joints, or the tuning slides found on Viennese oboes by Koch, Uhlmann, Küss etc. could help, but only so far.

JB:
During the nineteenth century there was a good deal of traffic in oboists between London and Paris. There must have been consistency of pitch there to allow Vogt, Barret and Lavigne to develop their trans-channel careers.

GB:
From what we know about the pitch used in the early years of the Philharmonic Society in London, it must have been close to that of the Paris Opéra (A=c.425). So when Vogt visited London to play with the group in 1825, pitch was probably not an issue. Three years later in 1828, pitch at the Philharmonic rose to approximately A=430-433 and during the next season, the pitch at the Paris Opéra also rose to 430-435Hz. Was this almost simultaneous shift on both sides of the Channel connected with Vogt’s presence at both Opéra and Philharmonic Society with a reconditioned Delusse oboe equipped with seven keys? Unfortunately, there is no way to verify this, but at the least Vogt seems to have been able to cope with the changes.
Robert Schumann, after a portrait of 1844 On the other hand, the other famous French oboist to exercise considerable influence in nineteenth-century England — A.-M.-R. Barret — championed his own oboe design, built by Triébert but designed to play sharper than other French-made oboes of the day. A third Frenchman, Antoine-Joseph Lavigne, who played in England in the latter part of the nineteenth century, was famous for playing Boehm-system oboes that also played at Old Philharmonic pitch. Most of us cringe at the thought of those high pitches around 452Hz, but the reality is that for the generation of Mendelssohn, Schumann and Brahms it was the norm. I recently acquired an oboe by Mahillon made for the British market, and was hopeful that it would be at 440, but it turned out to date from the 1870s at Old Philharmonic Pitch. Still, it’s a fine instrument in rosewood with neat keywork, and hopefully will dispel my apprehension about making music a quarter tone above 440!

JB:
When was it that we see an effort to standardize pitch?

GB:
I would see the instigation of the so called diapason normal as the first move towards an international standard. On 17 July 1858 a French law was passed, to take effect six months later, setting the national standard at A=435Hz. This was an across-the-board measure and attempted to protect the voices of young singers from damage due to the continual rise. But with the pitch of most French orchestras and theatres being at least A=445Hz, the law was a radical solution. It must have caused a flurry of instrument building; this was, of course, right around the time that Triébert brought out système 6, which with small modifications would become the Conservatoire oboe.
The 435Hz standard had lasting and widespread impact and continued to be the standard for over half a century. The Austrian empire, where pitch had risen to 454Hz, adopted the French standard in 1862. However it does not seem to have stuck, as records indicate that by 1878 some Viennese orchestras were back to 447. The 1885 international pitch conference in Vienna re-affirmed 435 as the standard. In England Henry Wood called for a compromise between Old Philharmonic Pitch (452Hz), common among British orchestra, and diapason normal. In 1896 new wind instruments were ordered for the Queen’s Hall Orchestra, made by the Belgian firm Mahillon, to play at 439Hz, which was readily rounded up to 440. Over the next decades other British orchestras followed suit, but wind bands remained at the high pitch for some time. On the Continent 435Hz was the official pitch until 1939, when the International Standards Association adopted 440. If the diapason normal became the most widespread pitch of the latter half of the nineteenth century, 440 has had a similar status in the twentieth. In the US, it was adopted by the American Federation of Musicians already in 1917, but there was still considerable variation, particularly as orchestras still relied heavily on wind European players.

JB:
Despite its official status, we all know that, like any pitch standard in the past, 440 has had a checkered career. The setting of international standards has done little to abate the pervasive upwards trend. It is a truism that the ear will accept a certain amount of sharpness but not one iota of flatness. For some of the upper woodwind soloists, this meant that they were able to sit on top of the rest of the orchestra by playing marginally sharper than the rest, thus leaving their egos intact. This meant that other players in the orchestra would be pushing upwards a little harder! And so it went on.

GB:
In short, we should not pretend that the twentieth is so different from earlier centuries with regard to pitch standardization. Theory and practice rarely coincide: there has probably been no time in history when 440Hz has been the reality at concerts around the globe. Even with the virtual monopoly of the French oboe, pitch has continued to vary, and until recently, players have not stopped complaining of having to force their instruments to play at unfamiliar pitches.

JB:
Surely, recordings must give an accurate picture of the pitch of orchestras and oboists over the course of the twentieth century?

GB:
One imagines that it would be easy to establish the pitch at which specific musicians performed, from their recordings. But recordings present their own special problems, so rather than clarify the situation, they only muddy the waters. The reality is that with all pre-digital forms of recorded sound — the cylinder, discs at 78, 45 and 33.3rpm, and the various tape formats — pitch is dependent on the speed at which the medium was recorded and, equally importantly, the speed at which it is played back.
If we consider the recorded legacy of Léon Goossens, for instance, we will find that he is represented on re-releases with pitches ranging from A=435 to 450Hz. Now, we can be pretty sure that, even if Goossens did not always conform to exactly the same pitch, this range is quite a bit wider than reality.

JB:
I agree completely. The most important and constant factor was the fact that he played on the SAME Lorée Oboe from the age of 14 — 1911 thus — until he was into his eighties, all except for a gap of two years in the middle somewhere when it had been stolen. There was only one real instrument repairer in those days, so when it needed some sort of repair or adjustment, it went to this repairer who recognised it instantly, resulting in its return to the indisputable owner. Goossens had absolute pitch, and I never heard anybody question that his A was not the correct one. You probably know the story with Goossens and Beecham? The latter asked him to give an A, and when it was given with rather a wide vibrato, Beecham said "take your pick, gentlemen". The basic and correct A would have been at the core of the vibrato. But as for playing at 435 or 450…Well, it was never like that in my experience, although after he was in his sixties, he did in moments of stress play a little bit sharp at the top. But that must be an age thing, because many players get this happening to them, without apparently noticing it. Even when he played orchestral stuff with big solos, Goossens remained at 440 -- certainly in the older, LPO days. I would say that in his 30's to 50's, when he felt absolutely comfortable in himself, he was supreme. Remember how Sir Adrian Boult once said “it is better to be sharp than out of tune” – but he certainly wasn’t referring to Goossens.

photo of Leon Goossens thinking GB:
But how much sharpness would Goossens himself have tolerated? If we go by those recordings, there are some performances that must have been completely off the radar for him. Take his first recording of the Mozart Quartet recorded with Spencer Dyke, violin, Ernest Tomlinson viola, P. Patterson Parker, cello in January and February, 1926. A wonderful performance, and a fascinating comparison with his later performance with the Léner Quartet. But when you play back the four sides of the 1926 recording at 78rpm, the first movement comes out at 440-443Hz on average, but the second and third are at 435. I say on average, because there is variation across the sides of the discs. The end of the last movement sounds flatter than the beginning. Surely not intended by the musicians! — much more likely an anomaly with the motor turning the table at the recording sessions. The National Gramophone Society recording was made on the cusp between acoustic and electric recording, but from the narrow pitch response it’s more likely to be acoustic. Electricity was used to drive the turntable, but the electricity supply in England at that time was notoriously variable, hence the pitch variation. To hear the recording at 440Hz, the second movement has to be played back at 78.89rpm and the third movement at 79.07. This can be done digitally, of course with very little effort. (Unlike in the analogue days where speed and pitch were integrally linked, pitch and speed can be modified independently with digital processing.) Nevertheless not all digital re-masterings have recovered the original performance pitch. More often digital editors assumed 78rpm, rather than checking the pitch, and where pitch was taken into account, 440 was assumed out of ignorance. This means that a large number of European recordings made up to the 30s have been re-released sharper than the original pitch of 435Hz.
Jumping ahead twenty years to Goossens’s recording of the Strauss Concerto with the Philharmonia Orchestra conducted by Walter Susskind recorded September 1947, played back at 78rpm this sounds at A=446-452Hz. Establishing the pitch of this recording is also tricky as Goossens is sometimes above the pitch of the orchestra, and once again the pitch varies from side to side. The first movement is noticeably sharper than the rest.

JB:
Do you mean to say that the pitch is not consistent within the same piece?

GB:
That’s right, and furthermore, the pitch drops from the initial 450-52 on side 1, to 446-48 Hz where side 2 picks up in the middle of the movement — figure 13 in the Boosey Edition. (I give pitches with 2Hz tolerance, based on averages of samplings from across each side). My guess is that the engineers were worried that the 4’40” up to figure 13 stretched the limit of how much music a single side could hold, so they ran their equipment 2 cycles under speed (76rpm) so as to fit as much as possible on one side. The change in speed trimmed about 10 seconds off the timing. Neither did they retain the same pitch for the other sides, nor did they return to 440Hz (that would have been an embarrassing drop!). There were other long sides coming up: side 4, the second half of the slow movement and cadenza is 4’28”, and side 5 is 4’00”. The engineers probably split the difference and recorded at 77rpm, giving the resulting pitch of around 446Hz. In the absence of the studio logs, I would guess that these adjustments were made on the spot during the recording process rather than at a later transfer stage. This recording was made a year or so after Columbia had begun experimenting with 30cm discs spun at 33.3rpm. These discs could hold 24 minutes of music and so could have fitted the entire concerto on one face. (This would have been later transferred to 78rpm format for commercial release.) But this would have meant a single take of the entire work. There are a couple of reasons why this is unlikely: firstly it would have been exceedingly grueling to expect a note-perfect take of 24 minutes duration from soloist and orchestra; secondly the breaks for the change of sides sound musical, as if they were played live rather the result of later splicing.
When they re-mastered this recording to LP, HMV assumed 78rpm, and did not check the pitches. The CD version (Testament, SBT1009, 1992) established A=approx. 435Hz. Both of these seem to be off the mark. Other re-released recordings of Goossens give less extreme pitching errors. In his recent anthology, Jeremy Polmear presented Goossens in an assortment of tracks with pitches from 440 to 445Hz. This is a more reasonable range, particularly given that the original recordings were drawn from a period of over fifteen years—1931-47. [This credit should really go to the transcriber, Malcolm McMillan. His article can be found here.]
There’s one other anomaly that confuses me. His recording of Elgar’s Soliloquy is high -A= c. 445Hz. The original LP recording dates from 1975 when there can be little doubt that the technology was at fault.

JB:
I would say that the Soliloquy — recorded when he was 78 at least — might be because of the ageing process if it was at 445, although I am a little surprised. But let me see if I’m following. We seem to agree that Léon played at 440 throughout his life, but how can it be that the recordings make him play at these different pitches?

GB:
A record can only reproduce the tempo and pitch of the recorded performance if spun at the same speed at which it was recorded. The term “78rpm” is at best a catch-all for recordings made to be played back at speeds anywhere from 70 to 90rpm. Speeds varied not only from company to company, but even from disc to disc. And we can be sure that, nine times out of ten, domestic gramophones did not play back at the correct speed.
Because of the paucity of accurate information, both on the pitches of recording artists and the exact speed at which recordings were made, we’re left with something of a Catch 22 — in order to find the correct speed of a recording we need to know the pitch of the performance, but in order to ascertain the pitch, we need to know the speed at which to play the recording.
That’s why Goossens is such a convenient case study. We know that he played — or at least aimed to play — at 440 over a wide span of time, so we can calculate the correct speed to playback his recordings. Obviously, we want to hear the entire Mozart Quartet at one pitch, and bring the Strauss down to where it was actually played in the studio, at 440Hz.

JB:
That makes sense. I also discussed this with my friend Anthony Burton, a former oboist, turned writer, broadcaster, presenter and a very perceptive reviewer of recordings of all sorts, mostly for the BBC. His “Building a Library” programmes are a delight. I remember him saying that he had a similar experience when doing the Mozart Bassoon Concerto for Building a Library: “I discovered that my transfer of the old Gwydion Brooke recording with Beecham was very sharp, which of course made it sound not just brilliant (which it is) but ridiculously brilliant. I was going to say something in the script, but when we checked the newest reissue we discovered it had been corrected. Of course, with a 1958 recording it can't have been a case of 78rpm masters: it must have been an unreliable tape machine, or something in the mastering process. You'd think it would be safer to make every transfer on the basis of A=440: it might not always be right but it would be a lot safer than assuming 78 rpm.”
Does all this mean that Léon’s rather energetic tempo of the first movement of the Strauss may not have been as fast as we know it?

GB:
Bringing the pitch down to 440 will take the edge off the tempo. Not only that, when it is restored to the correct speed we hear the fullness of his tone. I still remember putting that World Record Club disc on for the first time and wincing at the first notes: did the famous Goossens really have such a bright and edgy sound? He recorded another two performances of the Strauss in 1947, now held in the National Sound Archive. These were recorded for radio, probably on the large acetate discs that they used mid-century, and the one that I have been able to access so far (BBC Theatre Orchestra, conducted by Walter Goehr) sounds at 438-440, which restored my faith in Goossens’s sense of pitch!
Sometimes written records can help to resolve the Catch 22. For instance, Myrtile Morel (1889-1979) explained that when he was beginning his career the pitch in Paris was 438Hz. That is a bit sharper than diapason normal which set the scaling that French oboe builders had been using and continued to use virtually unchanged up to the 1970s (hence the perennial problem of the flat lower end).
Certainly players continued to push the pitch up, and firms like Lorée did their best to tune oboes according to different pitch preferences. Up to the 1970s, it was essentially the same bore and scaling of the Conservatoire oboe design that was pressed into service for pitches between 435Hz, in the most conservative cases, to 446Hz in the most elevated. Reed and staple design, along with blowing technique accounted for much of the pitch variation… but that is the subject for another article.

© Geoffrey Burgess 2005 on www.oboeclassics.com

Back to The Oboe 1903-53 CD details   Back to Links and Surprises