Romancing the Horn: Opera Stars Record Like It’s 1900

The soprano Susanna Phillips, left, and the tenor Piotr Beczala, right, listen to a recording they made with early-20th-century technology.

Susanna Phillips, a star soprano at the Metropolitan Opera, was listening to a recording of herself one recent afternoon, as she had done so many times before. This time, though, something wasn’t quite right.

“I can’t tell it’s me,” she said. Her rich tone sounded thin; her usually steady vibrato was strangely shaky.

The difference was the way her voice had been captured: that is, the same way they used to do it more than a century ago. It was a method that offered a pale — if, back then, magical — approximation of opera’s greats.

Whenever Luciano Pavarotti was asked to name the greatest tenor ever, he always answered Enrico Caruso, who became a household name from his recordings, made from 1902 until his death in 1921.

But how did Pavarotti know? Especially on Caruso’s breakthrough records, the sound is scratchy, wiry and wobbly. The same holds true for early recordings of Nellie Melba, Luisa Tetrazzini and other luminaries of that era. While there are entrancing hints of astonishing voices, it’s hard to tell what they were really like. If only we could record a singer today on the equipment used back then and compare the playbacks to modern recordings.

Well, that precise experiment took place earlier this month at the New York Public Library for the Performing Arts at Lincoln Center, thanks to the curiosity of Piotr Beczala, a leading Met tenor.

Touring the Met’s archives a couple of years ago, Mr. Beczala mentioned that his dream was to record some arias under early-20th-century conditions. He wanted to learn firsthand how faithful — or far-off — the results would be.

Peter Clark, the company’s archivist, mentioned Mr. Beczala’s fantasy to Jonathan Hiam, the curator of the performing arts library’s Rodgers and Hammerstein Archives of Recorded Sound; Mr. Hiam then contacted Jerry Fabris, from the Thomas Edison National Historical Park in New Jersey, who knows a collector in Illinois who makes wax cylinders like those Edison once produced.

So on his day off from the Met’s revival of Verdi’s “Luisa Miller,” Mr. Beczala got together with Ms. Phillips and the technical team to try out the vintage operation.

Verdi, the old way

Piotr Beczala sings an aria from “Luisa Miller,” recorded using wax cylinder technology.

Verdi, the new way

Mr. Beczala, as heard on a broadcast from the Met this year.

The library owns Edison cylinder machines, as well as an early Berliner gramophone — a competing technology that used flat discs. In 1912, Mr. Fabris explained, flat-disc phonographs finally outsold the cylinder ones and before long took over the market.

Mr. Fabris had brought similar equipment from New Jersey: an Edison Home Phonograph with a large black bell horn, a rotating holder for the wax cylinders and a hand-crank device to wind up the internal springs; and a similar-looking Edison Fireside Phonograph to play back the recordings. Both machines date from around 1909.

The material surrounding the wax cylinders is not really wax, he said, but something called metallic soap. Before using the cylinders, he had to warm them up under a light to make the material soft enough for the stylus to cut grooves as the disc spun.

“You want it to be like butter,” Mr. Fabris explained.

Mozart, the old way

“I can’t tell it’s me,” Susanna Phillips said of her wax-cylinder recording of “Per pietà” from “Così Fan Tutte.”

Mozart, the new way

Ms. Phillips’s “Per pietà,” on a modern broadcast from the Met.

The process is better at recording midrange sounds and has trouble with high frequencies. (Ms. Phillips was warned that it tends to favor tenors over sopranos.) Wide dynamic variables also test the machine’s capacity: Not knowing this, Ms. Phillips had prepared “Per pietà,” an aria from Mozart’s “Così Fan Tutte” that moves through extremes of high and low, loud and soft.

But Mr. Beczala was first up, singing “Quando le sere al placido” from “Luisa Miller,” accompanied by Gerald Moore, who played on a small upright piano so as not to compete with the voices. Putting the cylinder in place, Mr. Fabris was careful not to touch the surface: Even a slight thumbprint can create an impression. While Mr. Beczala sang, Mr. Fabris held a small brush in one hand and a little squeezable air bag in the other to disperse the dustlike shards of wax that are created when the stylus cuts into the cylinders.

Since the machine has no meter to check levels, Mr. Beczala tried out the opening of the aria twice, the second time moving closer to the machine. Both times, the ringing, virile quality of his sound came through fairly well, though dynamic variations essentially disappeared. Mr. Beczala was most rattled that his intonation sounded off — though this was a flaw of the equipment, not of his solid technique.

Finally, it was time to record the aria — or at least the first half or so, since each cylinder can hold only a little more than two minutes of music. “It’s like a black hole,” Mr. Beczala said, staring at the bell horn. “It takes you in.”

Listening to the playback, he commented that the resonance was not bad and that the high notes were O.K. But his softer singing sounded faint and distant, and the consonants, he said, “are nonexisting,” though in the room his diction was excellent.

‘Carmen,’ the old way

Mr. Beczala’s wax-cylinder “Flower Song.”

“The Flower Song,” from Bizet’s “Carmen,” came through more clearly. “I tried to sing more crisp than usual,” Mr. Beczala said.

When Ms. Phillips tried out the faster section of “Per pietà,” full of florid runs and roulades, she proved a quick study at the skill of leaning forward for soft passages and way back for louder ones, standard practice during Caruso’s era.

“You have to romance the horn,” she said.

To end the session, the two singers tried out some of the Act I love duet from Donizetti’s “Lucia di Lammermoor.” The machine is “not forgiving,” Ms. Phillips said, adding: “The tone quality changes, but not the dynamic. That’s infuriating to me.”

The contrast between their big, healthy voices and the crackly, thin recorded playbacks was stark. It proved just how difficult — indeed, impossible — it was to capture the sounds of the legendary singers a century ago.

Yet context is everything. For opera lovers in the 1910s, it must have seemed simply miraculous that the great voices could appear at will, however flawed their sound, in your living room.

In Other News

© 2020 US News. All Rights Reserved.