Recording »The Gothic« in Bratislava

David J Brown

David J Brown Our first inkling of a possible interest in HB by the Marco Polo label came on 12 November 1987, when I received a letter from a British music enthusiast named Kevin Mandry (who, I am pleased to say, subsequently joined the HBS) enclosing a copy of a very positive reply from Klaus Heymann, Managing Director of Pacific Music Co Inc of Hong Kong, the Marco Polo parent company, to his suggestion that the company might consider adding works by Boughton, Bantock and Holbrooke to its roster. So encouraging was this reply that Mr Mandry felt the HBS might well find a similar approach concerning Brian fruitful, and in due course I wrote to Mr Heymann.

The response dated 1 March 1988 stopped us in our tracks: "We have actually scheduled two works by Havergal Brian for recording later this year… the violin concerto with a suitable filler and the Gothic symphony. At a later date we may schedule other symphonies of Brian for recording… we would appreciate some advice concerning specific priorities as we have not been able to listen to any of the works other than the ‘Gothic’ and the Violin Concerto. We have been in touch with Ole Schmidt and hope that he will conduct The Gothic."

A s you might imagine, such a reply elicited elation and disquiet in about equal proportion, and my letter back devoted a good deal of acreage to The Gothic’s incredible scale and enormous difficulties (with which, his response in turn duly assured us, he was quite familiar). One particular point I made concerned the desirability of engaging two full normal-sized symphony orchestras to give the necessary string strength, and one of the things we must have to thank Mr Heymann for is his picking this up and holding to it right through to the actual recording. Another suggestion I made early on was that, if necessary, the Symphony could be recorded more manageably in two stages — part one for orchestra alone followed in due course by the vast Te Deum setting that forms part two, in which the full choral and orchestral forces are used. In the event, this was what had to happen.

It would be wearisome to go on through the lengthy three-way correspondence that ensued (United Music Publishers naturally became involved as soon as it proved clear that this was a really serious proposal). The Gothic recording was first scheduled for December 1988 with the Slovak forces in Bratislava, in sessions also to include the violin concerto with Tokoko Nishizaki, Klaus Heymann’s wife, as soloist (plus the 11th Symphony, which I had by now suggested as the fill-up).

For various reasons this proved to be impossible: delays began to pile up; first to January or February for The Gothic, then to a firm rescheduling on 20-31 March 1989 with the violin concerto/Symphony 11 postponed indefinitely. At this time Ole Schmidt was to conduct the recording, but he had to withdraw and Pacific Music Co derided to let the Music Director of the Czech Radio Symphony Orchestra, Ondrej Lenard, conduct in his stead. Some time after thus the dates were again set back slightly, to 27 March - 2 April.

Meanwhile, the potential for misunderstanding and delay of the Hong Kong-Bratislava-London communications triangle was making itself felt over the supply of the performing material, and by early 1989 it became clear that, due to the lack of enough rehearsal time for the choirs, the whole Symphony could not be recorded between these latest dates. Part one went ahead at the end of March (with, as I was delighted to learn when I arrived in October, the full strength of both the Slovak Philharmonic and the CSR Orchestras), with part two now rescheduled, yet again, for the end of May.

A worrying gap in communications with Hong Kong now ensued (we had had no direct contact with anyone in Bratislava), and it was not until mid-June that we learned part two had once more been postponed, for the same reason, and was now rescheduled for 15-18 October. This seemed a rather short period. especially compared with the 12 days for the whole Symphony in the first March schedule, and misgivings on this score were once again justified.

Early in our correspondence. Mr Heymann had invited me to attend the Gothic sessions and, as it became increasingly likely with the approach of October that at last, this time, the dates were going to stick, I decided to accept. Dr Ivan Marton, the Bratislava representative of HNH International (Pacific Music Co had undergone a change of name in the interim) organized my visa and hotel, and so on 14 October I found myself crossing the Austrian border only a few miles from Vienna, and just outside Bratislava, and passing through the rigorous and intimidating Czech border security — the brevity of whose remaining existence no-one then could possibly have foreseen. With so much media attention to Czechoslovakia in recent weeks, it’s probably much more common knowledge in the West now than then that the country consists of two linked republics, the more southerly of which, Slovakia, has its capital in Bratislava.

The Slovak Philharmonic is, therefore, virtually a national orchestra, second only in standing to the Czech Philharmonic, having been finally established as a State professional body in 1949. Czech Radio has its headquarters in the city, and its orchestra, the aforementioned CSR, is the second of the three professional symphony orchestras based in Bratislava. (It was founded in 1926, making it the oldest professional orchestra in Slovakia.) A considerable rivalry exists between the SPO and the CSR, and I later learned from the producer of the Gothic recording, Günter Appenheimer, that when the two bands came together at the beginning of the part one sessions in March, he had had to rearrange the seating to eliminate apartheid between the two cello sections!

That recording had been made, as part two was to be, in the new Concert Hall of Czech Radio, some twenty minutes’ walk from the centre of the city (northwards — the whole centre lies on the north bank of the Danube). The hall lies in the shadow of the organisation’s fiercely modernistic, inverted-pyramid office building. The hall has an extremely wide and deep platform, as well as the additional benefit of being thoroughly sound-insulated. This was in marked contrast to the SPO’s Redoute Concert Hall in the city centre — an ornately beautiful and rather art nouveau building with good acoustics (but far too small for The Gothic) which has the misfortune to be entirely surrounded by all-too-audible tram routes.

That same evening of my arrival I met both Günter Appenheimer and Ivan Marton for the first time. The latter wanted to go through few points in the score — mostly confirming all the passages that were purely a capella, those where the children were involved, and those requiring the extra brass. These facts would determine the order of recording to a large extent. Some details about the organization of the sessions also became clear. They would be on afternoons only, from 2pm until roughly 6pm, though the absence of strict union supervision meant that they might go on longer.

Secondly, the Sunday session was to be for the choirs alone — the first time, in fact, that all of them had been together — with the orchestras joining them only for the Monday, Tuesday and Wednesday sessions. And then the SPO were going off to Italy for four days (all that for one concert — by coach!), so if the symphony were not finished on Wednesday it could not be continued that week. I was not reassured by some of this, though I should have foreseen the reasonable enough point that the non-professionals amongst the singers could only assemble for part of each day.

I spent the next morning (Sunday) walking round a good deal of the surprisingly deserted city centre, and along the banks of the Danube, with a brief early lunch to follow before arriving at the hall around 1pm. Members of the 100-or-so strong Bratislava Children’s Choir were already arriving, and soon began rehearsing under their choirmistress as the adult choristers started to trickle in, and Günter and I thought about the best way of seating them. The orchestra would fill most of the stage — that was obvious. Ondrej Lenard wanted the children in two groups behind them and facing him, packed extreme left and right (they sing either unison or in two parts for a good deal of movement IV, not at all in V and briefly in the middle of VI, where the boys, for five bars only at fig 342, divide in two against the girls’ single line).

This left the adult choirs. They would have to occupy the audience space facing the platform — long, straight rows of comfortable upholstered seats extending the entire width of the hall, accessed by aisles by the walls. Günter wanted to group them across the hall sopranos-altos-tenors-basses, not realising the extent to which they subdivided into separate choirs. I persuaded him to change this so the four groups. choirs IA, IB, IIA and IIB (as they are designated in the score) were seated left to right facing the platform, with the SATB sections of each in that order, front to back. After some thought about balances, he decided to reverse this last point only, so that the basses were nearest the platform. Thus the seating remained throughout the recording. The final deployment of numbers was as follows:

Choir I: Slovak Philharmonic Choir (100 professionals) + Slovak Folk Ensemble Chorus (50 pros)
Choir II: Slovak National Theatre Opera Chorus (60 pros) + Lucnica Chorus (50 amateurs) + Bratislava Chamber Choir (50 ams).
The further level of subdivision to A & B in each was a mingling of voices from the choirs concerned.

Ondrej Lenard arrived at about 2pm, and after a good deal of conversation — unintelligible to me because mostly in Slovak, with a larding of German when Günter was involved — rehearsals started around 3pm. It was already clear that I might be useful in the control room rather than just sitting in the hall listening to the music as a guest, so I took myself off there (it was situated in the centre of the rear wall of the hall just under ceiling level, with a large window giving visual access both ways) to sample the sound balance over the monitors. There I stayed for the next four days! By the end I had not heard a single note of The Gothic live in the hall itself.

A word about the organizational set-up: Winter Appenheimer is an employee neither of HNH from Hong Kong. nor of Slovart, the Bratislava concern that looks after HNH’s recording interests there. He comes from an independent studio in Heidelberg, and spends several weeks at a stretch in Bratislava, making as many recordings — usually with the CSR Orchestra but sometimes with the SPO — as can be fitted in. It is thus something of a production-line — immediately before these Gothic sessions had come Glazunov’s eighth symphony for Marco Polo, and immediately afterwards, when the SPO had departed for Italy, his Chopiniana Suite as fill-up to a complete Nutcracker already in the can for the bargain Naxos label.

The Gothic, of course, is the last work in the world to fit into this kind of production-line, and it became clear that it had been repeated attempts to make it that had resulted in the delays and re-scheduling. Günter had heard no recording of it, and knew very little of its nature and background apart from its being a large and complex choral work. He normally acts as both producer and engineer, but having some inkling of its complexity. he had asked for technical assistance from Heidelberg. This proved to be less than satisfactory, and as the days passed I became increasingly involved in production.

We began, appropriately enough, with the very opening of IV (that is, of course, apart from the introductory orchestral chord) up to the side drum roll at fig.6. It was immediately obvious that Günter had an extremely acute ear for pitch, and many times throughout the recording stopped takes on the grounds of poor intonation from both orchestra and choirs at points where I had not detected any waver or slippage. He was also extremely concerned about the clarity of the words, to a point where I thought, and said, that too much emphasis on this might be counter-productive. My concern was dynamics, and I must say now that at many places an optimum detailed observation of gradations was not achieved. For example, I suspect that the children’s entry (five bars before fig 1) will not be as fff as it might be, and I know that, despite requests from me, resulting in long and unintelligible exchanges between Günter and Lenard, the choirs could not manage what I think is the proper dramatic fpp effect on the first "Ju—" syllable as they enter in turn at the opening of V.

But I am getting ahead of myself — and perhaps starting with an unduly negative impression. What was apparent straight away — and continued throughout — was the attack, weight, power and security of tone of the choirs. This will undoubtedly be the best-sung Gothic yet, and I would go so far as to say that it would be difficult to imagine it better done in this imperfect and rushed world. Initial doubts about the numbers being rather low were swept away, though I reserve judgement about the cumulative effect of the whole work on the ear in that hall’s acoustic, which seemed a little small and tight for it.

Günter was at first a little unwilling to record any of the a capella music without the orchestra present, on the grounds of the change of acoustic from an empty platform to a full one, but I persuaded him that a little extra reverberation would probably be a good thing, and that practically speaking there was too much a capella to waste orchestra time while it was recorded. He readily agreed, and after about 40 minutes rehearsal, these opening 35 bars were duly taped (yes, the recording is digital, but video tape is used). This opening also gave the first indication of the excellent qualities of the four soloists — Eva Jenisova (soprano), Dagmar Peckova (contralto), Vladimir Dolezal (tenor), and Peter Mikulas (bass), all enjoying successful operatic careers — at one of the very few points in the score where they sing as a quartet.

After a break, we moved on to the opening of V — much longer than the earlier passage and containing some of the most fantastically elaborate choral writing in the entire symphony. Lenard’s tempo for this seemed slower than any previous performance — a true Adagio molto solenne e religioso that tested the choirs’ abilities to the limit but over which, to my ears, they triumphed magnificently. My one residual worry about this passage is the positioning of the soprano solo who interjects twice "in the distance". In the distance she was not, and I was unable to persuade Lenard or Günter to have her moved away from front centre stage. Despite the latter’s assurances that she could be made to sound distant electronically, my doubts remain until we hear the result. This whole long passage of 98 bars to the distant trumpets’ entry at fig 177 was recorded, and the remainder of the afternoon spent rehearsing music later in IV. A good start — but 133 bars out of 1529 did not inspire much confidence in a four-day finish!

Monday — and the first appearance of the orchestra. Not quite the full strings of both the SPO and CSRO were used, but I don’t think anyone could complain about 24.20.18.16.11 (cf the score’s recommended 20.20.16.14.12 — I think one bass player was ill). Elsewhere there were compromises: the pedal clarinet and (surprisingly) the basset horn are unknown in Czechoslovakia, so bass clarinets were substituted. I’ll note other points as they arise naturally, but I hope I’m forgiven for not querying some things that, in retrospect, I might have done. This was partly due to language but more because of the sheer magnitude of the work’s practical difficulties, which perhaps only Robert Simpson and its four previous conductors would understand truly. Just getting through it, bit by bit, absorbed every scrap of energy and time. Lenard, for one, was grey with exhaustion at the end of every day’s work.

After a little trouble sorting out the horn parts in the first fanfare (figs 6-8), the whole 157-bar section to "Pleni sunt coeli" was recorded before the break, as well as the very first chord before the unaccompanied choral entry at fig l. Numerous takes were necessary to get the trumpets’ intonation right from figs 8—10. Other things which exercised us were making the children’s line as audible as possible from fig 13 onwards, getting a satisfactory balance for the soloists from fig 47, and (one of mine, this) the bells from fig 48 — I hope you can hear them!

This whole passage gave a good insight into Günter’s production style — a forward, well-spread orchestral picture based on a prominent string sound: good for The Gothic, I think. Something else also became clear: that the recording facilities were not really adequate for the wholly exceptional demands of the work. Not being in the business, I have no idea whether the set-up there would be considered good, bad, or indifferent by a Western professional (for the latter amongst you, it was a Studer console with a maximum input of 24 channels; microphones: Neumann LI87/U47/KM86, Schoeps /AKG; recorded on Sony U.Matic 1630/7BL — apologies for the mistranscription if any of that doesn’t make sense), but in order to pick up points of detail in the score, some very prompt adjustments of microphone levels frequently had to be made.

Now Günter had his work mostly cut out just listening, assessing, and noting the takes as the music went past, so it fell to me as the only person present who knew the symphony as an aural experience to try to ensure that as little as possible of its myriad changes of texture, timbre and dynamic were lost. We quickly worked out a system where, after a brief "tutorial" on each section as it came along, I would call (or yell!) a warning to Günter a couple of bars before the event and he would adjust the level from the nearest microphone(s) in order to catch it. It does sound rather crude and worrying, but we were continually listening to the results, which I hope and trust don’t come out anything like the excesses of the old Decca Phase Four system.

Following the break, we went back to the big "Stately and Majestic" tutti from fig 48 to get a good overlap, and then on to fig 104 before a second break was called, omitting only the a capella figs 69-76. To my ears, the descent from the Largamente climax at fig 94 through the ensuing horn and trumpet fanfares was particularly beautiful and affecting.

Again we jumped — to fig 126, to take in the remaining part of IV (to the conclusion) requiring the children’s voices. With this safely in the can — or computer — they departed, and we returned to fig 104-119 to complete the orchestra’s work in this movement. With them gone, the adult choirs remained to tackle one of the two remaining unaccompanied sections of IV. Figs 119-125 don’t look particularly formidable on paper beside the torrents of notes elsewhere — a mere 1½ pages of score — but it proved one of the most troublesome of all passages. Again and again the choirs had to be stopped because of intonation problems before it was safely completed, but not before Lenard’s very slow treatment of the chanted "Tu de victo" at fig 124 had registered. I mentioned this and he was quite adamant that it was "quasi Monteverdi". The last thing of all on the Monday was some rehearsal with piano of figs 203-210 in V, the section where each of the four choirs in turn come in, supported by the four small brass orchestras. All of IV was now recorded, apart from the 27 bars of figs 69-76: a tremendous amount of work, but still much less than necessary to finish in the time.

Tuesday afternoon -. first appearance of the extra brass. Here again was a compromise, but a not unexpected one. Brian’s ideal of 4 × (2 horns, 2 trumpets, 2 trombones, 2 tubas, and 1 timpanist) was reduced to a single group of 4.4.2.2, placed at back left where part of the children’s choir had been the previous day. I cannot now recall, and my notes do not help, where the extra timpanist was, if any. I think this part was taken by one of the orchestra regulars, playing the main set. We began with figs 203-210 in V, the whole band being used to back each of the four choirs in turn, moved electronically each time so that it appeared to emanate from the appropriate quarter of the stereo spectrum.

From here we moved to fig 227 in the same movement, at the point where the choirs re-enter pp section by section, after the second of the movement’s two long, purely orchestral passages. This builds through 57 bars to the apocalyptic climax with full orchestra and choruses and extra bands — everyone in fact except the children (who were not called on this day) — and on this final page of the fifth movement two queries arose. Firstly, "What is scare crow?". No problems with that one. We went through a jolly little pantomime with me swinging my arm round above my head like any loyal United supporter (football, for our overseas’ members), while one percussionist countered with a passable impression of a silent movie cameraman. A brief hunt in the percussion store yielded both types of rattle, wielded by two players, and I don’t think anyone will complain about inaudibility in the final result.

The other question, though, was a facer. After the first take of the final section (with rattles) came the query from Lenard — why did the lower half of the Choir I tenors have C# in their final chord when everyone else had B, E, or G#? Surely this was an error? I had nothing to offer apart from mentioning Brian’s habit of putting a dissonance in one single part into an otherwise pure consonance, and it wasn’t until much later that I noticed what anyone who really knows the work would have pointed out instantly: that the C# is prepared by its occurrence twice in the bass line that rises through the preceding three bars. I don’t think Lenard was convinced by this, though.

The choirs were dismissed, and a break for the orchestra ensued. I could fill a lot of space with these breaks — small glasses of black coffee and not much smaller ones of brandy being handed round the tiny, crowded kitchen through a smog of cigarette smoke, amidst a deafening babble of guttural conversation — with outside, of course, one of the most rigid regimes in the eastern bloc passing through what no-one could possibly have suspected was its penultimate month of existence.

The remainder of Tuesday took care of the two big vocal solos in VI: the tenor at the outset (but preceded by the beautiful oboe d’amore solo — no problem here with the instrument) and the bass towards the close. More time was needed for the former. It is no criticism to say that Vladimir Dolezal’s voice was a little light for what Brian throws against it near the close, and Günter had some trouble in adjusting his microphone balance so that he could still be heard while not seeming unnaturally close. No such problems, however, with the huge bass tones of Peter Mikulas. So ended Tuesday, and with some music from both IV and V, and most of the huge movement VI still to be done, clearly there was no chance of finishing on the Wednesday.

With this in mind, Ivan Marton quickly organized a further day’s work: such rescheduling seems to be far easier to arrange there than here, probably because the orchestras seem to do virtually all their recording work for this one company anyway, so from that point of view it was presumably simply a matter of shunting later projects a little further into the future to make way for the monster in hand. Even so, I was still surprised that it could be fixed for as early as the following Monday — though that was right out of the reach of my return ticket.

Wednesday did, however, see a great deal more completed: the finish of movement IV, for a start, with the taping of the remaining a capella section for adults’ and children’s choirs from figs 69-76. Then came the only remaining portion of the symphony that needed the children — figs 334-345 from VI, wherein at fig 341 Brian at the only point in the work narrows his vocal resources right down to a single girl’s voice, against a solo clarinet, muted violas, and repeated octaves in harp, cellos, and basses. After trying one voice, and then two, we settled for three in unison here - one more small, acceptable compromise.

Then it was on to the very end of the work, the 119 bars beginning with the molto espress. e teneramente for wind band at fig 408; the ensuing few bars for bass solo; the glorious fugal "In te Domine speravi" for unaccompanied choirs; and the final "Day of Judgement" cataclysm and resolution. Here we had the one serious organizational muddle (at least, that I was aware of). The extra horns, trumpets and timpani had been called, but not the trombones and tubas. The whole section was recorded without the extra heavy brass, and did sound wonderfully impressive; only those thoroughly familiar with the work would have been aware of any lack of weight. However Ivan Marton told me that it would be retaken with the extras on the following Monday; at the time of writing, I do not know whether this happened.

So we came to the final part of the sessions — the last of The Gothic that I was to hear, though far from the end of the work, in either sense. The choirs departed, and the orchestra went back to the first purely orchestral passage in V, beginning at fig.177 with the protracted and appallingly difficult fanfare for trumpets that follows the soprano solo "gradually groning [sic] fainter" (as the Cranz edition memorably puts it) — already safely in the can from the previous Sunday. Another compromise here: Brian’s eight trumpets playing in pairs became four playing singly. Purists will mutter at this, justifiably, but it was a case of working within the capabilities of the players. A solo horn in the orchestra follows and then the stormy Andante ma pesante that judders to a halt at fig 203 in the face of the interjections for bands and choirs.

I listened to a preliminary runthrough. What was wrong? The thin sound of one side drum — that was what was wrong. There is really remarkably little of this instrument in Part Two of The Gothic — its most extensive use by far is in the famous clarinet march in VI, where a single drum is perfectly adequate. That, however, was a section still to be recorded, and the whole matter of Brian’s side drums à3 had simply not occurred to me until this moment. I got them to stop and put the point that, though there was no indication in the score, the texture of the music at fig 186 really needs extra weight in this department. A seemingly endless pause ensued while another instrument was hunted out (fortunately, there were spare percussion players — in retrospect I should have asked for three!) and then the recording went ahead with much more satisfactory results.

And that was that. The SPO went off the next day for their short Italian visit and, for want of anything better to do, I returned to the hall for Chopiniana with a very unaugmented CSRO — small beer indeed! Three large passages of The Gothic remained to be recorded: the second orchestral section in V, from figs 210-226, and two big chunks of VI, figs 267-334 and 345-391, — a grand total of 480 bars, still approaching a third of the whole of part two. One more day didn’t seem enough, and we subsequently learned that a further day beyond the Monday had been needed to complete; exactly when this was I don’t know.

I spent my last evening in Bratislava going through these remaining sections and making detailed notes about all (as seemed to me) important changes of instrumentation, texture and entries to aid Günter in my absence. As I complete this report on the last day of the l980s the first legitimate studio recording of "the world’s largest symphony" has been completed; we have drafted the CD cues; but the editing has not yet been done, consequently there is no release date scheduled yet: don’t hold your breath, therefore, but don’t worry either.

On re-reading this unconscionably long account, I’m aware of many omissions: nothing about the sound of the orchestras (not bothersomely alien to my ears, but certainly different in many timbral details from the sound of English orchestras in the Royal Albert Hall that forms the only aural experience nearly all of us have of this work); nothing about the organ (a fairly uncharacterful, four-manual affair that I don’t think does justice to Brian’s very extensive part for the instrument): nothing about musicians’ reactions to the work (everyone I spoke to said they liked it, perhaps "after getting used to it"; they couldn’t all have just been humouring the foreigner: and there were far more guts and commitment in the playing and singing than one could expect in the UK); nothing, above all, about Lenard’s interpretation as such.

Here, I will wholly reserve judgement. Tempi seemed judiciously, sometimes imaginatively, chosen, but with such a disjointed schedule and with much of the symphony missing [from my personal experience], including the whole of part one, I was not able to form any opinion about his conception of the work’s totality. What is not at issue is his thoroughgoing professionalism, and to this, plus Günter Appenheimer’s acute ear, rapport with the performers, and wonderfully improvisatory make-do-and-mend use of his limited technical resources, must be owed whatever degree of success this first recording of Brian’s first symphony is deemed to have achieved.

To sum up, therefore: it has to be at least a partial success, if only because of the glorious choral contribution. There remains the carrot for some infinitely wealthy and altruistic company of the first recorded Gothic with the full specified forces. For the present, though, it couldn’t have been done anywhere but where it was: nine days’ work, in all, with never less than c140 professional musicians (in part one) and often nearly 400 involved (plus the amateur adult choristers and the children): think of all that in terms of Musicians’ Union rates and then grab the smelling salts.

Two aural impressions remain with me, both of the choirs: firstly, the electrifying express-train crescendo from p to fff on the final "—turus" syllables one bar from the end of V — perhaps the single most exciting choral moment I have ever experienced; secondly, the very end of the work, "non confundar in aeternum" fading into infinity on the granite foundation of 80 superb Slavic basses. Not even Brian’s mind’s-eye splendour of his beloved pre-World War 1 Potteries choirs could surely have surpassed them.

NL86 © 1989 David J Brown

David J Brown was the Secretary/Editor of the Havergal Brian Society 1975-1992 and its Chairman 1994-1998


Newsletter, NL 86, 1989