“It doesn’t quite have the atmosphere,” producer Annabel Connellan said, poised over her copy of the score. Howard Shelley, the pianist, agreed with her: “It’s almost too honest.”
Annabel and Howard were seated in a back room of St. Silas Church in London, along with Ben Connellan, Annabel’s husband and the session engineer. They were in the early stages of setting the balance for three days of recording piano sonatas by Munzio Clementi—sessions that would produce what would be the fifth pair of discs in Shelley’s recordings of Clementi’s complete sonatas. (The Clementi series is released on Hyperion Records; you can buy it here.) Annabel and Howard had both perceived an imbalance in the recording: a bit too close to the piano, which resulted in a strong presence of the action of the instrument’s pedals and hammers. Ben adjusted the microphones so that they were a bit farther from the piano and captured a bit more ambience and resonance in the church. Everyone was happy with the new sound, and they were then forced to take a break while the church held its noontime service.
Howard Shelley is nothing if not a highly accomplished pianist. Over lunch one day he estimated that he had made 120 records during his career, and his experience in front of the microphones was evident throughout. He rarely needed Annabel to point out mistakes he made; to the contrary, when he flubbed a note or missed a chord, he often stopped himself, backed up several bars, and then carried on through the passage in question. He placed great value in recording full movements and works without a break—what in the classical recording industry are called “long takes”—but he also understood that when recording, the game is to make sure that the producer and engineer capture enough material to assemble a complete, mistake-free record.
So Howard’s typical way of working was to play through each movement twice, which was not such a strain, as Clementi’s movements are generally quite short. After those run-throughs, he would consult with Annabel about passages that needed to be “covered”: places where he had made a mistake in one or both run-throughs, for instance, and areas where there were some interpretive questions to resolve. They decided on areas to retake—very often just another run-through of the movement, perhaps with some of Howard’s spontaneous decisions to back up and try again.
The relationship between Howard and the Connellans was easy. They had been working together for years, and they had a comfortable rapport. Annabel and Ben understood Howard’s working style quite well: they knew that they always had to be on their toes when he was at the keyboard, as he was liable to stop and begin a new “take” at any time. Howard was relatively laid back in the studio, but this casual attitude belied his intense approach to the music and the recording process. He took away copies of the session recordings in the evenings and returned the following mornings with notes about what he wanted to rerecord, adjustments he wanted Ben to make, and so on. Even with some entire works that Howard wanted to rerecord, the sessions were concluded a full day ahead of schedule.
“It sounds wetter. I want it more rich,” Jack said to William across the control room. (All names in this story have been changed.) The session was for solo viola (Jack) and piano (Michael), and was being recorded by Tom (the producer) and William (the engineer). Jack wanted to hear more of the room’s resonance in the recording, to have a more “churchy” sound and to hear less of the string.
In a way, Jack’s complaint about how William was capturing him reminds me of Howard Shelley’s initial impression of the recording being “too honest.” Both musicians wanted more room resonance in the recording. But while Ben Connellan successfully captured a sound that Howard liked, in this instance, William was unable ever to fully please Jack. According to William, it simply wasn’t possible to do what Jack was asking. The viola sound was not thick and voluminous enough on its own; it was “sonically behind” the piano, and in order to balance the two instruments, William needed to have the microphone close to the viola.
William recognized the contradiction, but was apparently unable to make Jack understand the problem, as the resonance (or lack thereof) in the recording was a recurring theme throughout the two days of recording work. The best they could do was to tell Jack that they would be able to add some artificial resonance during the editing process. Jack worried that this would be “cheating,” but Tom assured him that “it’s more common than you know.”
The resonance question, though it occupied almost the entire first morning of recording, proved to be almost irrelevant in the face of the session’s other difficulties. Tom and William both like to make records using a series of long takes, editing when necessary but generally avoiding the kind of minute patching that other recordists use. But with some of Jack’s performance difficulties, the long take approach proved impossible. Tom quickly grew exasperated with Jack’s apparent inability to hear the tuning on his own instrument; he was constantly point out pitch problems that the violist had apparently not noticed.
Part of the problem is that there was no real leader in these sessions. Tom sometimes tried to set the pace, but that seemed to irritate Jack. But when Jack and Michael took the lead, it immediately became apparent that they had no idea how to pace themselves, move through the pieces, and capture everything that needed capturing. By the end of the very long first day, everyone was tired and frustrated, and they had not yet covered half of the music scheduled for these two days.
Everyone was frustrated and chippy. While Jack repeatedly described his concept of one piece as “visionary,” Tom continued to point out small note-level mistakes he was making. After a take, Jack cheerfully said, “I thought the dynamics were good that time!” Tom replied over the talkback, “They were,” although he added just to William and myself, “You just have to play the right notes.” Eventually, they completed the sessions and the record was even professionally released. (It’s even pretty good at times, but for anonymity’s sake, I won’t link to it here.) But this session is a cautionary tale about the many ways things can go very wrong in the studio.