Friday, March 11, 2016

The NRR Project #1: Phonautograms

A phonautogram track, in close-up -- analog recording, like waves on the shore.
Edouard-Leon Scott de Martinville
c. 1853-1861
Though not officially listed as first in the Registry, this pivotal set of recordings was brought to life 149 years after they were made. It’s only natural, I guess, that the first recorded sound would be someone singing a dirty song.

Edouard-Leon Scott de Martinville, who invented the phonautogram, was not looking to make what we think of as a recording, something you “play back.” A bookseller turned amateur physiologist, he sought to reproduce, make visual, sound waves as they affected the eardrum. To this end, he attached a boar-bristle stylus to the narrow end of a container, leaving it just touching a soot-blackened cylinder on a screw. Shouting, singing, or playing into the concave sound-catcher, the vibrations would be crudely transferred onto the paper.

Scott was aware of the archival possibilities, and saw in the tracings a kind of code too be codified and interpreted, but by the naked eye, not via mechanical reproduction. “Will one be able to preserve for the future generation some features of the diction of one of those eminent actors, those grand artists who die without leaving behind them the faintest trace of their genius?” he wrote.

Members of the incredibly helpful site conceived and spearheaded methods of digitally reading and reproducing high-resolution copies of the original lampblacked prints. The result is handful of precious voices from 1860.

Interestingly, they are readable today only because Scott included some 250-Hertz tuning-fork calibration tones, a simple but effective way to insure a constant reading rate. If the tone doesn’t waver, the recording is running at correct speed. A few earlier phonautograms he made, from 1857, can’t be interpreted because no one knows at what speed or speeds (the recording device was hand-cranked, adding more variables to the problem) they were recorded.

Audio literally "tracked" through a recording medium -- the first sounds from 1860.
 The first sound, from April 9, 1860, is Scott singing “Au clair de lune” (“By the light of the moon”), a familiar French folk song from the 18th century, the lyrics of which turn to double-entendre pretty quickly. Someone needs a pen, someone’s fire is getting lit, and a guy and a gal get it on. Nice. FirstSounds's top-notch site  houses all the recordings of Martinville's work. As acoustic scientists worked on the project, there is plenty of precise, helpful data to be found.

The key to understanding pre-digital recording technology is the imaginative leap of analogy. Sounds are waves, invisible but waves nonetheless, in evidence like the tide against the shore, or wheat rolling and rippling in the summer wind. They are palpable, capable of measurement and description. Looking at an enlarged segment of phonautogram, it resembles nothing so much as calligraphic brushstrokes on rice paper, furrows on a hillside. Like an animal's track, sound leaves its wake in a medium that captures it, freezes it forever -- or until the medium is lost.

Even from an attempt this crude, we receive a message. No wonder it once was considered akin to magic. It’s a haunting kind of time travel.

The National Recording Registry Project tracks one writer’s expedition through all the recordings in the National Recording Registry in chronological order. Up next: ‘Around the World on the Phonograph.’

No comments:

The NRR Project: Rachmaninoff and Stokowski

Piano Concerto No. 2 in C minor Composed by Sergei Rachmaninoff Performed by Sergei Rachmaninoff, piano Philadelphia Symphony Orchestr...