Last Sunday afternoon a sold-out crowd eagerly listened as Claremont Trio played in the new Calderwood Hall at the Gardner Museum. This article relates my experience listening to the performance from the first balcony in front of the musicians. My experience there was good – not great. Listeners in other areas of the first balcony had a much more variable experience, and on the whole they were disappointed. Balance between the strings and the piano was poor, with the strings often inaudible. Comments from people I trust who listened from seats on the floor were uniformly very good. I was unable to determine how people perceived the balance in the two higher balconies. The purpose of this article is to at least partly explain why the sound was so variable, and what might or might not be possible to do about it.

The Claremont is a treat to hear, and their recording of the Mendelssohn Trio in D minor Op. 49 is a family favorite. I was very eager to hear them play it live. To make the performance even more interesting, this was the first performance I have heard with their new pianist, Andrea Lam. (Donna Kwong is taking a break for a stint as a new mother.) BMInt had arranged for reviewer’s tickets, and I was assigned by the museum to the first balcony. My first problem – where to sit? The piano (with the cover removed) was at a 45° angle to the walls of the cube, facing the right-hand entrance door of the balcony.  There were several seats on the entrance wall looking down at the front of the group, just where I guessed the best sound for this trio would be, and I took one.

As you probably know, the balcony seats are surrounded by a thin wooden rail, angled to give the least possible interference with the view of the stage. Below the rail is a clear glass panel, angled almost imperceptibly downwards, presumably to prevent flutter echoes. (These would not have been problematic in a hall of this width.) Interestingly, my neighbors on either side mentioned that they were quite uncomfortable sitting just in front of the clear window. There was an irrepressible fear of falling through to the floor below, and the fear made them sink back into their seats. One of them said the fear of falling could be conquered by placing a foot against the window. The physical contact somehow convinced the body that there was an obstruction. I don’t know how many others in the balconies felt this way – but I suspect they were numerous.

When the trio arrived on stage I found I was looking directly at the top of the violin. (This is good.) The cello faced away from me, at about a 45° angle from my view. If I sat back in the seat the cello was behind the glass window. (This is not good.)

The set opened with the Piano Trio in C Major K 548 by Mozart. To my ears while sitting back in my seat, the piano was unusually clear, with a great, well balanced, and rather too strong sound. I could hear the violin fine throughout the concert, but when I sat back in my seat the cello was nearly always inaudible. So I sat forward, in order to see the body of the cello over the balcony rail. This was much better. Whenever the cello had a line not doubled by the piano, it was audible, although a bit weak. For the second movement, I sat further forward on the chair, so I could see not only the body of the cello, but also the (invisible) reflection of the body on the floor. (The floor is nearly perfectly reflective for sound, but not for light.) The reflected energy comes soon enough after the direct sound from the cello to add about three decibels to the loudness without decreasing clarity. Sure enough, the cello balance got quite noticeably better, but the strings were still weaker than I would have liked. Fortunately my back is strong enough to sit upright on the edge of my seat for the whole concert, and this is what I did.

The Claremont played the Mozart with perfect ensemble and a light gay touch. If the balance had been a bit better and my sitting position more comfortable, I would have been thoroughly delighted.  The next piece on the program was “Trio” by Sean Shepherd, commissioned for the Claremont in honor of the opening of the Calderwood Hall. It was in three movements, the first and last were interesting and full of humor, the middle movement, titled “Calderwood” was lyrical, starting and ending with a lovely solo for cello. The piece was fun and not too long. What more could anyone want! There was a lot of solo string work in the piece, and I did not notice any serious balance problems.

The last piece was the Piano Trio in D minor Op. 49 by Mendelssohn, very familiar to me from the Claremont recording. I again sat upright on the edge of the seat looking down for the whole performance without complaint, although somewhere in the Scherzo my neck got quite stiff, and I had to move my head around. I looked around the balconies to see who was leaning on the rail or sitting as close as possible. About 20 (of ~200) were leaning on the rail, perhaps as many more sitting (possibly uncomfortably) close. On the whole the performance was very good. Andrea Lam is a spectacular pianist, and the Bruskin twins played with their usual precision and sense of line. I was happy, if a bit uncomfortable physically.

After the concert I spoke with as many people as I could about what they had heard. One friend in the first balcony moved from a seat directly across from mine at the side and partially behind the musicians to an empty seat in front of the group. She said the sound in the first seat was very poor, with both strings nearly inaudible. The frontal seat was much better. Another friend in a similar position also said he could seldom hear the strings and added the additional information that leaning over the rail made very little difference. I was able to confirm that when sitting back in his position the instruments could be seen above the balcony rail, so leaning forward would not have made much difference. I did not find someone from the area of the balcony further behind the musicians, but I doubt his experience was better.

When the concert ended I was able to talk to people who had listened from the floor. They uniformly thought the sound was perfectly balanced. One of them – looking at me in sitting in the first balcony – thought that next time he should try to get a ticket there. Maybe he should – but probably not for a concert involving a piano and strings.

I have great respect for architect Renzo Piano and acoustician Yasuhisa Toyota, the designers of the hall. They have dared to do something original without having had the opportunity to test it beforehand. Engineering and Physics do not work like that. The sudden inspiration that solves a problem only comes to a mind prepared by months and sometimes years of partial success and failure. Tools for testing architectural acoustics are still crude and unreliable. The data you would need even to crudely simulate the sound of the Calderwood Hall are unavailable or suspect. With hindsight we can see at least a few of the details that could be improved. The first of these, of course, are the glass fronts of the balcony rails. Could someone without a degree of vertigo have predicted that many would find the vision of the floor beneath their feet uncomfortable? I am probably not one of them. Piano made the rest of the building out of rectilinear metal and glass. How could he not use the same material for the balcony fronts? And Scott Nickrenz wanted a venue where the audience could easily see each other, making a concert a community experience as well as a musical one.

Alternatives exist, of course. You need something acoustically and visually transparent that looks solid enough to feel absolutely safe. The cast-iron balcony fronts of Boston Symphony Hall and Jordan Hall fill most, but not all, of these requirements. With hindsight a rectilinear version of the Boston Symphony balcony fronts made from thinner metal might have done the job.

But even leaning over the balcony rails did not solve the balance problem for me in the Claremont concert. We must always remember that instruments do not radiate sound equally in all directions, and that both chamber music and orchestral music performance practice has evolved to mitigate these directional properties to some degree. Balance problems still remain. The frequencies from 1000Hz to 4000Hz that define the timbre of instruments occupy only a two-octave segment of the ten or so octaves that we can hear. String instruments radiate these frequencies chiefly from the top plate and direct them perpendicular to that surface. The sides of the instruments are made of thicker wood, and radiate very little. The back radiates some – but not as much as the top. The conventional string quartet arrangement with the viola on the audience right favors the violins, as they point their tops to the audience. The arrangement the Borromeo uses, with the second violin on the right, brings to viola to the fore, leaving the second violin to try to remember to play louder than he or she might think appropriate, given what they are hearing from the other instruments.

Orchestras typically play in large halls, and large halls are reverberant. Sound that is not directed to the audience typically hits surfaces that eventually direct it into the audience. The loudness of sound in a large hall comes chiefly from these reflections, since there are a great many of them. Clarity suffers, but loudness and balance become less dependent on the directionality of the instrument. Calderwood Hall is not reverberant. As I noted in the previous review, the corner reflectors formed by the under-balcony surfaces reflect sound back down to the floor, creating a kind of reverberation and improving the balance between instruments. But in the balconies the majority of the loudness comes from the sound that travels directly upward. As a consequence clarity is unusually good, but balance becomes uneven.

When I heard a rehearsal of A Far Cry, the balance (although not the clarity) was good all the way around the third balcony, at least if one looked over the balcony rail. This is because all the instruments had similar directionality. I did not hear the combination of piano and strings, where the difference in directionality is huge.

A piano radiates the frequencies of 1000Hz to 4000Hz from the top of the sound board. In a conventional hall with a piano cover at full stick, these frequencies are bounced prominently into the audience, where they are enjoyed. The top blocks these frequencies from filling the stage house and the upper reaches of the hall, increasing the relative strength of the direct sound and improving clarity.

But in the Calderwood a piano cover at full stick will direct all the brightness to one side, and much of the balcony will be in shadow. So the top is removed. Then what? Suddenly we are outside a hundred years of performance practice. How can we proceed?  Musicians can learn to make the balance good for some people – but for all?

Pianists look down on the sounding board and get some of the brightness and loudness of their playing. But most of it goes up to the balconies. The sound in seats on the floor is largely the sound from the bottom of the piano, which lacks brightness and sounds muddy. String players, accustomed to the sound bouncing off the cover, are suddenly uncomfortable, uncertain as to how loud they should play. A friend told me that the Claremont players struggled to be comfortable with the cover off. Their success at achieving a perfect balance for the people on the floor indicates that they succeeded – but only for the people on the floor. To do it, the strings probably had to play a bit softer than they were used to, and the piano a bit louder. The folks in the balcony – at least those not in front of the top surfaces of the instruments – ended up not being able to hear the strings.

A string quartet, on the other hand, is all strings, and the players can all hear each other. There is sufficient reverberation on the floor that they will feel in familiar territory and are very likely to give a great performance. I don’t think balance will be a problem anywhere in the hall. The same appears to be at least partly true for A Far Cry.

I wish I could pull out a magical solution for the combination of piano and strings. We need an acoustically semi-transparent cover for the piano that directs sound both upward and laterally in all directions. In addition it must be visually transparent and sufficiently traditional-looking that we will actually employ it. I suspect something like this could be designed. It won’t work the first time. Patience is required, but we now have a hall we can use for testing. Don’t go away. The Calderwood Hall is very promising – and still a work in progress.

Note: This is a follow-up to David Griesinger’s previous BMInt article here.

David Griesinger is a Harvard-trained physicist who is eminent in the field of sound and music. His website is here.