Perfecting Sound Forever-An Aural History of Recorded Music. By Greg Milner. New York, NY: Faber and Faber, Inc., 2009, 416 pp., $35.00.
Reviewed by Doug Preis
Music should be seen and not heard,” read a sign I saw posted decades ago in Harvard’s Music building. It was creative wordplay on the early 15th century English proverb, “Children should be seen and not heard.” This new dictum was, most likely, the reaction of a disgruntled graduate student disappointed with the obsessive study of musical scores as opposed to actually hearing music performed by visible musicians. Those of us who side with this student do welcome the opportunities to hear live musical performances, receive broadcasts of concerts, or otherwise listen to recorded versions of music. Unquestionably, technology has made significant inroads in music acquisition, modification, storage, production, and dissemination. Even to this day, one should not underestimate the extremely important and underlying role played by radio, for example.
Greg Milner delves into these different inroads with aplomb, beginning with the earliest attempts to capture sound and culminating with the current status of sound production, reinforcement, broadcast, and especially recording. Perfecting Sound Forever, subtitled An Aural History of Recorded Music, is a well-written, interesting story indeed, but in my view, and perhaps that of the author, it does not have a happy ending. Surprisingly, the reason for this is the curious interplay between technology and society.
The book clearly reveals a complex admixture of philosophy, psychology, creativity, and aesthetics as well as monetary issues associated with musicians and technologists. The reader will encounter strong opinions, even stronger personalities, ingenious contributors, outstanding music makers, and admirable archivists. Unfortunately, narcissism abounds, sonic auteurs emerge, and greed obtains.
Control of sound is the central issue. They all want it: the musician, the sound reinforcement engineer, the recording and mixing engineers, the producer, and the listener. The control of sound started innocently enough by using performance spaces having “good acoustics” like salons and concert halls that simply contained the natural sound of voices and/or musical instruments and served to direct primary sounds. These spaces neither augmented nor reinforced the original sound. The invention of microphones, amplifiers, and loudspeakers caused a technological paradigm shift. Radio City Music Hall’s opening in the mid-1930s demonstrated how sound could be captured, controlled, and reproduced throughout the hall by using electrical technology. A boastful claim was then made that there was nothing that architectural acoustics could do that could not now be done electrically. Initially, sound reinforcement met with strong disapproval by audiences and critics alike as being both unnatural and larger than life. But, over time, to the dismay of many, it became accepted, and sound reinforcement is now the norm, being both expected and demanded by audiences everywhere. Who really controls the sound in today’s performance spaces? By popular demand, the listeners in our society do.
Recording sound is the process of capturing some of those seemingly elusive propagating compressions and rarefactions of air and storing them in a different physical form in such a way that a facsimile of the original sound can be reproduced from that stored version. The process initially evolved scientifically with the simple goal that recorded sound should yield “an ever bettering approximation” of reality. Along the way a schism arose because of a new, somewhat paradoxical, view that reproduced sound, which is itself sound, could be made better than reality! Hence, there is the author’s critical subtitle, “perfecting sound forever.” The latter view now dominates, and technology has become the slave to the socalled “art of recording,” whose devotees unceasingly strive to perfect sound without actually defining what perfect sound is.
Leopold Stokowski was an unabashed fan of using electrical technology to dramatically “enhance” his recordings of certain classical works. Audiophiles would refer to them obliquely as the “Stoki thrillers.” Les Paul, the pioneering genius of the electric guitar, virtually created a new musical instrument and made fascinating original recordings consisting of multiple layers of sound. So some musicians certainly embraced the new technologies, whereas others like the eccentric pianist Glenn Gould, the legendary folkie Pete Seeger and the honest Neil Young, who now self-effacingly refers to himself as “Grandpa Granola,” eschewed them. The eclectic Bob Dylan was true to his unpredictable self. The technologists became enablers in service of the artists’ creativity.
Along the way, the numerous contributions to recording, post production, and sound reproduction technologies from various and rather colorful scientists, engineers, and inventors have been remarkable indeed. The Edison cylinder acoustic recordings of operatic tenor Enrico Caruso, which sounded poorly owing to acoustic resonances of the horns used in the recording/reproducing process, were decades later, restored digitally by Dr. Tom Stockham, using a signal processing technique called blind deconvolution, to sound modern and more realistic.
Magnetic recording is both the hero and villain in the author’s interesting story. It initially used a moving wire that was magnetized analogously to the microphone’s electrical signal. Then, as the technology developed, the wire was replaced with ferric oxide on an acetate backing. A durable permanent recording could be made, copied, and even edited (“to make recorded sound lie”). Most significant was the fact that two separate channels or tracks of sound could be magnetically recorded simultaneously, which initially gave rise to extremely spacious sounding stereophonic recordings. Early commercial examples of these are open-reel “pre-recorded” stereo magnetic tapes and long playing stereo records, especially those made by conductor/producer Enoch Light, which were transferred from the original master tapes. Using even more channels, many separate sounds could be recorded at the same or different times and then blended or mixed together afterwards. These mixed separate channels, each often consisting of a single voice or instrument, constituted the final recording. “Liveness” and a “sense of space” were therefore constructed artificially. Gertrude Stein’s famous “There is no there there” is amusingly applicable here. Now this final recording, in a somewhat contradictory sense, becomes the actual performance. Eventually, these individual tracks “laid down” by various musicians came to serve simply as a palette of recorded sounds that enabled a mixing engineer to become a sonic auteur, the arbiter of the sound of the recording, often to the dismay of the original performing musicians who clearly had lost control of their own sound.
One evening on Johnny Carson’s Tonight Show the sound reinforcement system went dead. He remarked to the TV audience that the “presence” of his voice was missing as he spoke. Mercury records had a line of classical recordings marketed as, “Living Presence.” What, in fact, is “presence”? Milner devotes more than 50 pages to this nebulous concept. Is it the aural illusion of “being there” with the musicians or a form of “constructed liveness” or even “an improvement on reality”? The quest for presence was multidirectional but one with a major discovery: the augmentation of recorded sound with added echo. Columbia Records had an echo chamber for that purpose and one only has to hear, for example, Dave Brubeck’s “Take Five” album to realize that there was more to Joe Morello’s bass drum sound than the wooden beater he used to strike it. The effect is more subtle on “Kind of Blue” by Miles Davis. Those two albums became recorded jazz best sellers. Elvis Presley’s voice on his early recordings was given an odd but interesting spacey sound by Sam Phillips of Sun Records, who simply but cleverly used a brief electronic delay to add a single echo to it. Modern enhancements to recorded sound such as, for example, slap echo, doubling, artificial reflections, and reverberation were inevitable because they ultimately serve to sell recordings.
Milner clearly discusses compression of electrical analog versions of sound. In the early history of telephony, Bell Lab engineers used a nonlinear electronic device to squeeze the range between quiet and loud sounds for voice transmission over electrically noisy telephone lines. Then a complementary device was used at the receiving end to restore the original dynamic range. This combination of compression and expansion, called “companding,” improved voice quality by effectively reducing the interfering channel noise. Compression alone is perceived as a reduction of musical dynamics. The softer sounds become louder and, in the extreme, there is little distinction between soft and loud. Everything sounds loud. Extreme compression is certainly attention-getting (witness contemporary TV/radio commercials, which, compared to earlier ones, are not only loud but also continuously and uniformly loud because they are so compressed) but it also can be annoying and fatiguing. Society is becoming more accustomed to audio compression because it is now endemic. Consider todays cell phones, television, film, compact discs, and radio broadcasts. In comparing different loudspeakers, compact discs, or radio stations, for example, the louder of the two is normally preferred. Film sound is often very loud and compressed because it has to compete head on with our visual system which normally uses about 80% of our combined perceptual abilities. Loudness always wins and loudness sells.
Radio has both a glorious and inglorious history. Recall the days of “payola” where record companies bribed radio stations to give airplay to certain of their records to create “hits” and improve their sales. Milner indicts contemporary radio broadcasting mercilessly. Radio is a commercial medium. Even National Public Radio has its too frequent “fund-raising” commercials. But it is what commercial radio has done to recorded music that is his cause for concern. In order to increase the size of its listening audience and thereby increase advertising revenue, a radio station will use virtually every technological ploy available to increase its broadcast range and modify its broadcasted sound to distinguish itself as being superior sounding in comparison to competing stations. The music is severely compressed and heavily equalized to emphasize bass and treble, musical peaks are clipped off, stereo separation is exaggerated, artificial echoes and synthetic reverberation are imposed, all with a view to increasing modulation levels and enhancing the sound. Radio is now the epitome of the loudness war and it exerts final control over the music we hear. The resulting over-processed sound has a tremendous “in-your-ear” impact and gives an almost otherworldly impression. But what we hear is actually a grossly distorted version of the music both in terms of the emphasis of high and low frequency tones, which is an alteration of the spectral balance, and diminished musical dynamics due to extreme compression and clipping or limiting of musical peaks. Furthermore, these distortions are superimposed on those already existing in the original recording. To experienced listeners the music just sounds terrible. Today’s CDs of popular music and radio broadcasts of them have been criticized as sounding like “noise with a beat.”
There is still a market for what might generously be called over-processed recorded sound because in certain ways it is excitingly different. Some people continue to buy such CDs and music downloads, but not as many as before. When I recently asked a student who was seriously interested in pursuing a career in sound recording, “What is a good recording?” He responded, “Whatever people want to buy.”
Others simply have given up listening to contemporary recorded music as well as certain radio stations and even stopped attending sound-reinforced live musical performances. Sadly, they now follow the advice given in the second part of the dictum, “Music should be seen and not heard,” posted at Harvard, namely, that music should be “not heard.” In a different interpretation of the dictum, rock concerts have actually turned out to be much more profitable than sales of a groups’ recorded music; concert goers really want to “see the music” performed and hearing its processed sound, blasted from multiple loudspeakers at near painful acoustic levels, is almost secondary. They see but do they really hear the music? The group’s recorded sound on CDs is often sold as a souvenir of the concert along with printed tee shirts and coffee mugs. Do the visual and social concert experiences outrank the auditory one?
Greg Milner has given us much to contemplate regarding the interplay between technology and society in both its service and disservice to music. It’s a good book and there is much more information about music and musicians contained in it than can be mentioned in a short review.
Doug Preis is Professor in the Department of Electrical and Computer Engineering, Tufts University, Medford, MA 02155. Email: firstname.lastname@example.org.