Sound recording, barely a century old as a commercial and accessible product, has democratised classical music, created new kinds of artists and affected the way music's performed. It's been responsible for the rapid spread across the planet of genres which otherwise would have hardly been known outside the places that gave them birth. At a deeper level, it's changed music from being an occasional listening pleasure to the point where it's a continuing theme in our lives, like a Ennio Morricone tune in a Spaghetti Western. And yet the origins of sound recording lie with a surprisingly simple device. If you'd lived before the principle was known, would you have guessed that the combined sounds of a choir and orchestra could be captured in, and reproduced from, a single wavy groove in a wax surface?
The first sound recorder was a table-top gadget that used existing technology and was just waiting to be invented. It took a long time, though, for the penny to drop. In 1806 an English scientist, Thomas Young, had shown how to capture the vibrations of a tuning fork as a pattern on a wax-surfaced revolving drum. Then in 1857 a Frenchman of mechanical bent, Leon Scott, made what he called a "phonautograph". Scott's device was the first of many to convert sounds into visual patterns which could be studied. It consisted of a large horn to catch the voice, tapering down to where a taut membrane - typically a bit of pig's bladder - was stretched over the end. Attached to the membrane was a single pig's bristle. The bristle rested on a revolving (hand-cranked) drum coated with "lamp black".
You spoke into the horn, the membrane vibrated, the vibrations were converted into a shaky line scratched on the revolving drum (see talkingmachine.orgphonautograph.html) Eventually, there were those who realised that to make a phonautograph convert its tracings back into sound, the trick would be to inscribe the pattern of vibrations - to replace the bristle with a needle capable of cutting a groove in the revolving drum, rather than just draw a line. The inscribed groove could repeat its vibrations back through the taut membrane and the horn if the drum was rotated. That very idea dawned on another Frenchman, Charles Cros, who, on April 30, 1877, presented a paper on the subject to the French Academy of Sciences. Crucially, although Cros described a device, he didn't get round to making one, and, true to the mores of fin-de-siecle Paris, he died from too much absinthe a few years later (see history.sandiego.edugenrecordingcros.html) Edison's breakthrough
The person who famously did first make a recording and playback machine was Thomas Edison. Working in his laboratory in New Jersey, US, he made a device that traced sound on a cylinder. You spoke into a horn, the vibrations were picked up by a delicate diaphragm (various materials were tried, including layered rice paper, very thin glass, or mica till eventually thin metal became the standard material), and transferred to a needle that inscribed its line on a rotating cylinder covered with tinfoil into which the vibrating needle dug a continuous groove. This meant you could reverse the process, so the groove made the needle and the diaphragm vibrate, and lo, your voice came back at you from the horn. In 1877 - reputedly August 12, though that's not certain - he made what's claimed as the first sound recording: his own rendition of "Mary had a little lamb."
The original no longer exists, but you can hear Edison briefly telling the story and repeating the verse, in a 1927 recording on www.nps.govedisedisoniadocumentary.htm
Edison soon dropped tinfoil for wax. Cylinders (played on phonographs) gave way to discs (played on gramophones). Alternatives to the steel needle were tried, such as the carved thorn needle and the stylus of diamond, ruby or sapphire. Wax was replaced by the more robust shellac. Direct linkage from the voice to the stylus was replaced by electrical impulses. The search for new materials and longer playing times eventually produced the vinyl long-player (LP) which arrived commercially in the UK in 1950. The ability to encode more information in the groove gave rise to stereo recordings.
(The groove has "hills and dales" and so the stylus is driven minutely up and down as well as side to side. The combined movement produces two channels of sound. "Quadraphonic", with the sound distributed to four speakers, also appeared, but never really caught on.) Electronic digital recording spread from the studio where the masters were made into the consumer product, and gave us the compact disc. Now the CD is already fading in importance, eclipsed by the ease with which digital recordings can be stored and shared between computers via the internet.
Steady progress, yes - but all technology progresses by revolutionary leaps as well as evolutionary development. The story of sound recording is like that. Evolution was seen in the the careful honing of studio techniques, the perfection of materials, the increased understanding of electronics and acoustics. Revolution existed in a handful of step changes that consigned an earlier system if not to the dustbin then to the attics of our old friends the diehard traditionalists (they're the ones who preferred cylinders to discs, acoustic recordings to electric, 78s to LPs, vinyl to CDs).
Acoustic to electrical
The earliest recordings were acoustic - the sound went directly to the vibrating diaphragm and stylus. Enormous amounts of the sound spectrum were lost in the process. Strings sounded feeble and special "Stroh" violins were developed (named for their inventor) with horns to focus the sound.
See them at historywired.si.edudetail.cfm?ID=46
Players and singers had to cluster near the huge horns that gathered the sound, which meant that big orchestras and choirs couldn't be recorded. An early acoustic recording of Elgar's great choral work The Dream of Gerontius was made with an orchestra of nine, a choir of eight and, instead of the organ, a bass concertina. In the very early days there was no way of producing copies, or "pressings" of an original record. So you'd put maybe 10 recording machines around your players, who would perform a two-minute piece, say, 30 times over three hours. This gave you 300 wax cylinders to sell. By 1901, things had improved and cylinders could be mass duplicated by a moulding process that produced up to 150 copies from the original.
Meanwhile, Emil Berliner, also an American, was developing his "gramophone"
playing flat discs. He made many attempts, but his patents of 1887 and 1888 confirm him as inventor of the disc record "gramophone". Discs and cylinders co-existed for some time, both making rapid strides in quality as materials and techniques improved. In 1913 however, Edison, whose company had made and marketed many kinds of cylinder machine, including the luxury Victrola - a cylinder player in a big cabinet - moved to discs with its Diamond Disc gramophone, and the cylinder era was effectively over. At best, what you had from an acoustic recording (often called "pre-electric") was a sort of sketch of the real thing. If it was well done, though, and the music was right for the medium, you got something that's still worth listening to today. That's true, for example, of the first jazz recording ever made - the Original Dixieland Jass Band playing "Livery Stable Blues"
and "Dixie Jass Band One Step" on February 26, 1917 - you can hear it at www.redhotjazz.comodjb.html
Then in 1925 came electrical recording, developed by Western Electric, in which the collected vibrations were converted to electrical impulses that could be amplified. It was an enormous leap, supported by developments with microphones, speakers and amplifiers that enabled full orchestras to be recorded and reproduced with a realism that began to approach what we're familiar with today. It was with the advent of electrical recording that the notion of "hi-fi" (High Fidelity sound) was born.
Shellac to vinyl
The black shellac record, played at 78 rpm - usually 10 inches in diameter, though 12 inch for many classical recordings - held sway for all of the inter-war years and well beyond. Its great disadvantage was that it played only four to five minutes a side. Performers and listeners made the best of it. Popular pieces, after all, could easily be tailored to the right length. Classical music was more difficult, and was often trimmed to fit.
If you wanted a long work, you bought a set of records. These were usually issued in large book-like holders - the original "albums" - with individual pockets for the records and some pages of notes and cost serious money. In 1937 a complete EMI recording of Sir Thomas Beecham's Glyndebourne performance of Mozart's Don Giovanni came on 23 78rpm records at pound;6 18- (about Pounds 250 today).
During the 1930s, however, work was going on in the recording companies that would eventually change everything. It culminated at the Waldorf Astoria Hotel in New York on June 20, 1948 when Columbia records announced the first commercially successful long-playing record. It's worth quoting here from the man masterminding the presentation, Edward Wallerstein, then a Columbia senior executive: "As I stepped up to the podium to address the 50-odd representatives of the press, on one side of me was a stack of conventional 78-rpm records measuring about eight feet in height and another stack about 15 inches high of the same recordings on LP."
Wallerstein played a comparison test, showing the superiority of the new format. "The critics were struck not only by the length of the record, but by the quietness of its surfaces and its greatly increased fidelity. They were convinced that a new era had come to the record business."
The vinyl LP, supported in the studio by the growing use of magnetic tape instead of wax for the master recording, went on to rule the industry for some 40 years, changing forever the way we listen to music. The pop "album"
appeared, for example, as record producers made themed collections of successful 78s and re-sold them as LPs (Columbia's first pop LP was The Voice of Frank Sinatra). The first vinyl LPs, played at 331Z3rpm went on sale in in the UK in 1950, followed the following year by 7 inch discs, chiefly for popular music, played at 45 rpm. Stereo arrived quickly, by 1958, at which time the 78 was effectively dead. If you have a Beatles 78, you have a very rare object. Read about them at users.telerama.comagp discography.html
Analogue to digital
Digital recording is based on sampling the stream of sound at extremely frequent intervals (typically over 44,000 times a second) and encoding each sample electronically as a number. The principle was understood for some years before the invention of the digital compact disc, usually credited to the American William T Russell in the late 1960s. The real breakthrough came when electronics giants Sony and Philips combined to solve the technical problems of commercial CD production. So according to Philips, CD was "invented collectively by a large group of people working as a team".
When CDs went on the market in 1982, there were grumbles from the record shops about designing new shelves, but they soon realised that people were buying their music all over again. Some say that the arrival of CDs made record companies lazy, because all they had to do was sit back and resell their existing catalogue at higher prices. The real difficulty for the industry came later, with the realisation that the recorded digital product sitting on a CD could have a life of its own as a digital audio file that can be stored and passed around as a file in a computer. Once music exists as a digital file, you can do anything at all with it, just as you can manipulate, store and share a digital photograph or a word-processed file.
The major casualty has been the CD single - the bottom has dropped out of this market although sales of the actual tunes, as legal downloads, have increased enormously. The process came of age at the start of this month when, for the first time, a single, Gnarls Barkely's "Crazy", reached number one on the basis of legal internet downloads, before a CD was released.
Perfected through the 1930s and 1940s, magnetic tape's (or wire) main importance for years lay in the way it made the studio process so much easier. Singer Bing Crosby, anxious to be freed from his relentless schedule of live radio broadcasts, put a lot of money into the development of high quality tape recording during the 1940s. The story's told, with a super photo of Bing with a tape recorder, at www.tvhandbook.comHistoryHistory_tape.htm Magnetic tape appeared seriously in the home with the inventions in the 1960s of the compact cassette by electronics firm Phillips, and the Stereo 8 cartridge, largely targeted on the in-car market, by Motorola. Stereo 8 died in 1980, at about the time that cassette sales outsold LPs for the first time.
The experience of music was fundamentally changed as recording developed.
In particular, where a soloist was concerned, a sense of intimacy could develop - a feeling that the person was with you in the room. This began with singers such as Enrico Caruso and built steadily until now there's a galaxy of favourite stars across all styles. With the advent of MP3 players and iPods, users can also create their own playlists and categories of styles and moods. This would have turned a 1950s teenager green with envy.
The Science Museum in London has lots of early recording equipment, but there's a limit to what's on show at any one time. www.sciencemuseum.org.uk
The National Sound Archive, part of the British Library, has some equipment, but its main purpose is to preserve historic and representative recordings. www.bl.ukcollectionssound-archivensa.html
The National Museum of Scotland has an extensive collection of gramophones and phonographs. www.nms.ac.uk
Edison and contemporaries at www.nps.govedisedisoniadocumentary.htm
I found some real gems, including a very early pre-electric version of Hoagy Carmichael's "Stardust", at the website of the Starr-Gennet Foundation www.waynet.orgnonprofitgennett.htm
Lots of pictures of early equipment, and sound samples at www.inkyfingers.com
Early recordings of American variety acts from the Library of Congress collection rs6.loc.govammemvshtmlvssnde.html