Who Owns Your Listening Preferences?

By Jorge Caballero

In my days as a college student the Internet was fairly new. In order to have it, you needed a computer new enough that you could plug a telephone jack to it. Also, you needed software, and for that, you made your way to an electronic goods store or a record store. There, in some display bin made of cardboard you’d pick up a single CD wrapped in plastic: Try America Online For Free. Upon returning home, you’d turn on your PC by pressing a big, soft button nestled almost in the middle of the CPU tower, and after some time, you’d insert the CD into the PC. A window appeared on the monitor. You’d then follow the instructions: select your service, agree to the terms, and connect. After a few seconds of computer magic, the America Online interface would present you the World Wide Web. “Welcome."

Because of the limited capability of the dial-up connection, the internet was not rich in graphics, and instead, written content was a significant portion. This created the perfect conditions for the advent of chat rooms and online forums, and thus, the first guitar forums appeared. In those days, I would make my way into classical guitar forums online, hoping to interact and perhaps share useful ideas and information with other guitar enthusiasts around the world. However, the artificiality of this interaction was prone to decay. Cloaked by hyperreal anonymity, people appeared comfortable writing things they most likely wouldn’t utter out loud. One time, Manuel Barrueco was the topic of discussion. Someone proferred negative comments about his playing not worth repeating here, and it bothered me to the point that for the first time I felt compelled to comment. I typed a message in defence of Barrueco. The previous poster wrote back. I wrote back, and he wrote again. Our cyberspace dance went on two more times, with our messages becoming longer and longer to the point of rhetorical futility. By the fifth exchange my fingers refused to type a reply, choosing to instead tap rhythmically at the desk where the monitor sat. What had started as a fun idea was quickly turning into an unwanted chore. Unwilling to drive my fingers to the keyboard, I concluded it would be best not to stoke the rhetoric fire, and in any event, it was time for me to focus my attention once again on more useful and edifying endeavours –like studying. I visited the forums less and less frequently, and in time, my interest in guitar forums was all but extinguished.

Before that happened, though, something I read in a thread caught my attention. It was a simple announcement of a guitarist playing a recital somewhere. As would be expected, the tone of the message was friendly and positive, the kind you can imagine if you’ve ever seen or read one before, something like: “Hey guys! Just to let you know that x guitarist is playing at x location. He is so-and-so, and has won this-and- that. The program is great. Hope you can make it.” However, if good vibes and cheer were intended by the poster, the implied effect on the one and only person to reply was anything but that. If anything, this reply seemed to inhabit the opposite emotional space, like a nemesis, an evil twin. It went more or less like this: “Definitely not interested. I heard one of his CDs once, and didn’t like it one bit.”

As an aspiring professional guitarist at the time, I felt this answer was unduly dismissive, and harsh. Unlike today, making recordings then was not as simple, primarily due to logistics. Self-produced CDs did not really exist. In order to make one, a record label had to believe in you enough to sign you and make an investment in you. That implied at least a measure of individual quality above the average, irrespective of the recording itself. Also, unless you were already well established (and I mean, well established), a record label played a significant role in deciding your repertoire. For my first album, I submitted 3 different ideas for repertoire, and none of them were chosen. The record label didn’t like them. Having run out of ideas, I then proposed 3 Bach Cello Suites offhandedly, and to my surprise, they wanted that. Thus, for my first album I recorded 3 Bach Cello Suites.

Record labels also determined the quality aspects of the production, from sound engineer and producers, where and when a recording would be made, and how many days you had to complete the recording. For my Bach Cello Suites album, I was fortunate to be given three days for the recording and to work with an excellent producer. Around that time, Eduardo Fernández, one of the great master guitarists in our history and my first role model as a guitarist, told me that he was often required to complete a full-length albums in six hours.

I also knew from personal experience that a lot of guitarists sounded far better live than in their recordings. Eduardo, once again, came to mind. The quality of his tone was often the subject of criticism by those who only knew of him by his records. But if you heard him live, you would know such criticism was misguided.

But what truly concerned me about the forum post was not the opinion given in it, but whether a recording ought to be a definitive parameter by which to form such an opinion. It would be strange to say to someone who encourages you to visit the Grand Canyon: “Definitely not interested. I saw a picture of it once, and didn’t like it one bit.” But, alas, we seem to accept such statements when it comes to music without a second thought.

---

Let’s begin with the obvious: recorded sound is only a sample representation of actual sound, more or less in the same way a picture is a sample representation of an object. Think of a recording as a computer whose processor creates images that speakers translate as sounds. Certain parameters, like sampling rates for the analog-to-digital conversion and bit depth, are set at different stages of the process in order to render sound as faithfully as possible. These sound images are then retranslated by speakers or headphones for rendering to the human ear. In a sense, recording sound is like a highly precise game of Telephone.

But just like in this game of Telephone, while the factual sound information of a recording may present its content as accurately as sampling rates will allow, the result it produces diverts meaningfully from the human experience of listening. To begin, humans inhabit a world where sound is analog, the sampling rate of which is theoretically infinite. Digital sampling rates, on the other hand, are not. Fortunately, there is a way we work around this difference. Thanks to the Nyquist Theorem, it has been established that audio waveforms sampled at double the rate of human hearing (between 20-20,000 Hz) convey all of the information a listener can use, so the sampling rate standard for recording has been set at 44.1 kHz. Technology has further expanded sampling rates for recording, so it’s now common to record audio at much higher rates than the standard; or in other words, at rates higher than a human’s ability to hear them.

While all of this is true, it would be hard to imagine that many humans would be fooled into thinking recorded sound is “real,” and while perhaps our inability to extract useful information from frequencies beyond 20 kHz can be demonstrated, it may be conceivable that the infinite sampling rate of real sound can be perceived by humans in ways that have not thus far been measured, perhaps because a human isn’t simply a listening machine, but a complex system of various entities that collectively gather information about their experiences. For instance, when we look at a photograph of the sun, we usually squint. Why? No immediate harm will come to our optical nerve from pixels printed on photograph paper. Yet, this self-protective reaction to seeing the sun even as an artifice lies so deep in our psyche that we will act upon it when induced. Reactions like this belong to the realm of visual perception, which differs from visual acuity. The latter measures our ability to discern shapes, while the former refers to our neurological responses from visual stimuli. And just like you couldn’t evaluate the former by testing the latter, measuring our perception of sound simply from the perspective of Hertz frequencies is, although perhaps an accurate measure, vastly incomplete.

And this is the point where recordings betray their illusive exterior: hard as they may try, they cannot convey the additional factors of sensorial experience brought by the viewer. We don’t just hear a concert; we see a concert. Factors such as where we are sitting, the sound of the room, the lights on stage, the movements of the performer, all coalesce to form a unique experience for every person, as unique as their position with respect to the performer. The great Russian pianist Sviatoslav Richter is said to have demanded that stage lights be dimmed as much as possible so as to make it difficult for the audience to see him. He reasoned that hearing music was the experience, not seeing him play. Alas, as much as Richter may have wished otherwise, the near darkness he imposed upon the audience turned out to be a brilliant extra musical contribution that enhanced the listening experience.

Yet, a recording can also be turned into a multi sensorial experience. As a fun experiment, here is something that you should try with any recording you own: find the right volume and distance from where you should listen to it. For example, listen to a Segovia record standing one meter away from the speaker. You may find the sound of his playing to be awkward or somewhat incoherent. But, stand 5 meters away and you’ll experience a different aural reality, where Segovia’s sound becomes precise and his playing flows beautifully. Now try the same with a more recent production and you’ll typically find the opposite to be true, where the farther away you move from the sound source, the less coherent and enjoyable the recording becomes.

There are reasons for this. When Segovia began making recordings in the late 1920s, the aim of his record producers would have been (aside from selling records) to provide a window of experience to a public unfamiliar with Segovia; to give them a glimpse of what it may be like to see him in concert. Therefore, the focal point of perception was not the speaker from whence the sound of the record came, but rather an imaginary, more distant listening point. This was also useful for another reason: it disguised the noise inherent to recording from the listener by blending it with the natural background. As decades passed, however, advancements in recording technology greatly improved the signal to noise ratio, which allowed a different kind of sound to be extracted from recordings, and thus listeners could be closer to the speaker and not detect recording noise as easily. Along with that, and with the aid of advertisement, the public became more comfortable with recorded sound, too. Consumers were told that recordings were capable of producing sound “as good” and eventually “even better” than what the human ear can perceive, and thus, a preferential shift to recordings over concerts began . Professional musicians also shifted to create interpretations of music that better suited recordings, or in the case of the best musicians, learned to adapt them to the medium they employed. By the time Glenn Gould retired from public performances in 1964, he stated that “analytical dissection by microphone” allowed him to present music from a “strongly biased conceptual viewpoint,” and presciently believed that the function of concerts had been taken over by the recording medium. (1)

Glenn Gould’s views, however, were not so eagerly shared by his contemporaries, and audiences may not have been as eager to supplant live performances for recordings…yet. During the decades of Segovia’s illustrious career, it is not likely that one would purchase his records because they were seeking a strongly biased conceptual viewpoint analytically dissected by a microphone, but instead, because they wanted to be reminded, albeit feebly, of the experience of having seen him in concert, hoping for the next time he arrived in town; or perhaps because they lived in a place Segovia was not likely to ever perform and wanted to imagine the experience. Today, however, things are different. Someone’s choice to see a musician in concert is more strongly biased depending on whether they previously enjoyed an album that musician recorded. Like Glenn Gould half a century ago, audiences have shifted their priority values, so live performances exist relative to recorded sound, unlike the way it was when recording technology was an infant, so by the time I read that comment on the internet forum in the late nineties, an individual felt comfortable defining a musician’s quality from that viewpoint.

At one level of analysis, this is a good thing. Thanks to the high fidelity and accessibility of recording technology, musicians reached their audiences as quarantines and lockdowns were imposed last year because of COVID-19. That would have been unthinkable decades ago. Another positive aspect is that it has allowed accessibility of means to what was hitherto a large, albeit somewhat marginalized group of musicians, who no longer need to hope for record labels to sign them, since making your own recordings isn’t as expensive and complicated as it would have been in prior generations. With some knowledge and a relatively small investment, any musician can obtain the recording equipment they need and control the quality of their albums. This also allows the consumer easier access to any music they wish to hear. In Segovia’s time, the only recording available of Joan Manen’s Fantasia - Sonata was his. Today, you can find 11 different versions on Spotify alone, as easy to access as tapping your finger on a smartphone. These improvements to our quality of experience with music are thanks to the wonderful efforts of science and recording technology.

But, what has been lost in this transformation?

Of course, while consumers now have a myriad of options, they now face a new problem: it becomes harder to know whether the recording they listen to is good from a musical perspective, as opposed to a recording quality perspective. In other words, it is often difficult to differentiate between a good recording and good music in a recording, and that leads to problems discerning interpretations. In a separate article, I pointed to the problems of interpretation inherent in music, and how it is impossible to know from heuristic experience whether we are listening to an interpretation that aims to present musical content faithfully. If recordings can cloud our judgement of music and interpretation by the suspension of disbelief of high fidelity, then, how can consumers discern between good and bad music, or appropriate levels of interpretation? Are they limited to choose between recordings by how good the audio quality of the recording is (music and interpretation aside), or by its cover? And if so, it may be that the quality of the artist becomes irrelevant and relegated to extramusical factors of impression by this process. For if consumers are driven to listen to recordings based on visual or sound preferences –instead of the music in that recording and its interpretation, what meaningful differences could any individual performer really make? A quick survey through recording labels shows this. With notable exceptions, the average recording label features its classical music albums based on repertoire rather than artist.

And furthermore, wouldn’t it be possible then for anyone, with or without years of education in music, to take as many parts of several recordings of the same piece and splice them together to produce a strongly-biased-conceptual-viewpoint-analytically-dissected-by-computer-software without ever spending the lifetime it takes to play a musical instrument well? And couldn’t we also create any music we wish simply by splicing or sampling the millions of sounds available to us from what already exists? If so, who needs trained musicians anymore? Perhaps Glenn Gould was right all along to prefer recordings, but I am left to wonder if he understood the irony of his predilection which, like Sophocles’ oracle, announces what could be the ineluctable end of the concert performer.

In conclusion, while advancements in technology have improved recordings beyond the capacity of the human ear by some parameters, they are but a small representation of the totality of our experience with sound. If we listened to the recording of a tornado as it passes through a town, we may have a sense of the awesome power of nature at work and lament the devastation it “appears” to be leaving in its wake. But no matter how much power this recording would evoke, and how faithfully technology recorded it, it would never recreate our listening experience of the same event if we were there. In that case, we wouldn’t just get a sense of the power of nature. We would sense it. We would see the devastation it leaves in its wake and understand how small we are in the midst of it. Fear would set. With a recording, however, there is no fear. We are in a controlled environment and there is no direct threat to our lives. There is only the illusion of fear. Our subconscious knows it. The recording is only an illusion, though perhaps outstanding beyond our ears.

And in that sense, the analogy with photography I made earlier rings true. While cameras can detect light and render images that far exceed the ability of the human eye, comparing a picture of the Grand Canyon to the experience of being at the Grand Canyon is preposterous. In music, the quality of an artist is best experienced in their live performances. Recordings could never render this, but with consumers’ values shifted, a self evident fact of experience can be easily turned on its head.

Notes:

1. Cott, J., Glenn Gould: the Rolling Stone Interview, part 1, August 15, 1974, https://www.rollingstone.com/music/music-news/glenn-gould-the-rolling-stone-interview-part-one-180448/