Why do audiophiles prefer jazz recordings? Why is Miles Davis Kind of Blue the classical recording that warrants high quality reproduction? Reflecting on these questions can inform our understanding of Golden Era tango recordings, why they warrant spending money and effort to get the highest quality reproduction possible, and why nothing else even remotely compares.
Getting good quality transfers of tango from shellac records to high resolution 24bit files is not easy. Luckily I recently discovered The Great 78 Project by George Blood LP:
The Great 78 Project is a community project for the preservation of 78rpm records. From 1898 up through the 1950s…
Fortunately for the fans of Argentine tango Epoca de Oro music this includes many recordings of tango from the late 20s to the mid-50s. And best of all, they’re all uploaded to the Internet Archive and available for free download.The downside is that while these are amazing transfers they retain the clicks of the shellac records. So I invested in some special software (iZotope RX8) to remove most of the distracting clicks and noise but without losing the amazing sound quality.
I was able to get some amazing transfers of D’Arienzo, Canaro, Troilo, Biagi, Pugliese, Fresedo, De Angelis and more. The process of removing the clicks is fairly simple once you learn how to use the software, and I’ve been able to restore some of the most amazing sounding tango recordings I’ve ever heard. You can hear the musicians and singers almost as if they’re in front of you. I’ve never heard tango like this, so clear and immediate!
TangoTunes transfers come pretty close and I hear that they’re coming up with new releases from Troilo 1943-1945 and also the Biagi and D’Arienzo recordings will all be re-transferred again as they have improved transfer techniques and also better quality shellac records. I’m looking forward to these new releases. In the meantime, it’s exciting that I can restore these recordings myself from the George Blood 78 Project as they come out with more amazing transfers of tangos.
We have many “wonderful” tango DJs on the scene these days, or at least that’s what all the marketing on Fakebook Argentine Tango pages tells us. In reality more often than not what I see is something like this: the “wonderful” DJ plugs the audio device into the headphone jack of the laptop (ie., no external DAC); and then plays the music from MP3 files on iTunes. The music that results is downright depressing: thin, muffled, low energy, lacking in space or transparency, basically a dying corpse. If that wasn’t enough the DJ then attempts to resuscitate the victim by amping up the volume to ear-assaulting levels.
It took me a while to figure this out and now I can spot the mediocre audio setup almost the moment I enter the space. In the early 2000s this was understandable as little was understood about tango music and its requirements, at least outside of Buenos Airs. Today, however, we have access to online DJ lists and tandas, quality transfers to high resolution audio files on websites like TangoTunes which also provides ample information about tango music, and any aspiring DJ can find information on how to get acceptable sound of these in terms of the computer software and digital-to-analog converters or DACs.
Despite all of this we continue to be subjected to really awful sound at most milongas (really “practilongas”) organised by dancing teachers who are clueless about DJing themselves or about hiring competent DJs. Apparently the costs are too high and one place where these people are comfortable skimping, apart from the audio equipment, is the music files. Apparently, MP3 music files are perfectly adequate because “most people can’t tell the difference anyway”.
I’m old enough to have spent a decade or two listening to music on vinyl and then CDs using fairly good quality Hifi equipment: NAD amplifier, Harman Kardon CD player, Mordaunt Short speakers, Dual turntable. So I was a bit of a hi-fi buff. When the internet came around and you could download mp3s I was excited about the possibilities. But I noticed the clear difference in the quality of the sound compared to the CD. Clearly a 320kbps sounded better than 128kbps, but it did not sound anything like a CD.
More recently I decided to look into music reproduction for tango and in the process learned about the most recent developments in computer hi-fi or “computer audiophile”. The rules of the game have changed quite radically in particular because the traditional source of vinyl or CD was replaced by the sound or audio file (MP3, ACC, AIFF, FLAC, ALAC). Consequently where in the past one had to consider a good quality turntable and/or CD player today you must consider the audio player software (iTunes, Traktor, Audirvana 3, JRiver Media Center, etc.), additional sound processing software or “filters”, and the digital-to-analog converter or DAC (AudioQuest DragonFly, Chord Mojo, Chord Hugo 2, etc.)
Generally speaking the discussion of a computer audiophile setup as it relates to classical music will also apply to tango, although tango might have some special considerations. Not all classical music is recorded in the same manner. I find that classical orchestras that use “period instruments” have a similar considerations as tango recordings. These orchestras use instruments that would have been in use when the music was originally composed rather than their more modern versions. So they tend to be recorded in a way that tries to capture the timbre and ambience of the individual instruments rather than just the content of the music as it emerges out of the whole orchestra as a sort of general image.
So we want an acoustic image where there is a clear separation of the instruments that captures the characteristics of the various instruments whether they occupy the space in the low, midrange, or high range of the acoustic spectrum. If find that the ÉdO recordings also preserve the timbre of the instruments as it would have been immediate to the players and audiences at the time. With a good recording you can capture the separation of the instruments and their individual timbre. All these acoustic instruments emit complex sounds that are due to harmonics, so that a given note played on an instrument is actually a collection of several notes. So in the process of recording and reproduction you want to capture the richness of the acoustic signal as much as possible.
Now, when the audio industry was transitioning from analogue formats such as the vinyl record and the tape to the digital medium of the compact disc it was decided that to capture the maximum amount of perceptible information the ideal format is 24bit/96kHz. I have a vague understanding of what this means, but basically this format samples most of the information that is perceptible to the human ear, although advocates of high resolution or HiRes audio argue for higher, and the CD is able to contain more information, much more than eg., the vinyl record. But 24/96 was considered adequate even for most audiophile purposes.
Now, when the first computer audio files were brought out, their original purpose was to be able to transfer sound over what was the fledgling internet. An uncompressed sound file of a 3 minute song can take around 30Mb of hard drive space. In the 80s through 90s hard drives were small and the internet was slow, especially compared to today. I remember that in the early 90s a computer costing two thousand dollars or more would have no more than 20Mb of hard drive memory. Such files would take hours to upload and then download over the internet. The MPEG standard was developed for compressing music files to a 10th of the size or less. As Wikipedia tells us:
In regard to audio compression (the aspect of the standard most apparent to end-users, and for which is it best known), MP3 uses lossy data-compression to encode data using inexact approximations and the partial discarding of data. This allows a large reduction in file sizes when compared to uncompressed audio. The combination of small size and acceptable fidelityled to a boom in the distribution of music over the Internet in the mid- to late-1990s, with MP3 serving as an enabling technology at a time when bandwidth and storage were still at a premium. … With the advent of portable media players, a product category also including smartphones, MP3 support remains near-universal.
MP3 compression works by reducing (or approximating) the accuracy of certain components of sound that are considered (by psychoacoustic analysis) to be beyond the hearing capabilities of most humans. … Compared to CD-quality digital audio, MP3 compression can commonly achieve a 75 to 95% reduction in size. For example, an MP3 encoded at a constant bitrate of 128 kbit/s would result in a file approximately 9% of the size of the original CD audio. (emphasis added)
So the main purpose of the MP3 format was eliminating a lot of auditory information in order to reduce the file size while retaining an “acceptable” quality of sound. Now, most discussions today concern the issue of what is “acceptable fidelity” and what is “beyond the hearing capabilities of most humans”, with many claiming that “most people can’t tell the difference”. This begs the question why Sony and Philips decided in 1980 on the Red Book or CD standard of 16bit/44kHz if most of that extra information makes no difference because it’s not perceptible by humans?
While the original intent of the MP3 was the compression of the music file so that it can be more easily stored and transferred, this has subsequently become an informal standard for newly recorded music. Studios aimed to record music so that it could sound “good” even when reproduced from an MP3 file on a cheap audio device. The result of this development is that since the 90s most people are mostly exposed to music reproduced from MP3 and rarely hear music reproduced from CDs on quality audio equipment.
Now when you compress a recording of classical or tango music to the MP3 format you effectively strip the music to its bare musical content losing most of acoustic information that contributes to its atmosphere that is critical to the aesthetic experience of the music. People who are trained in listening to music or have listened to high quality reproduction of the music are able to immediately perceive these aspects. Most people are still able to respond to these aspects without being consciously aware or being able to articulate them. The problem is that most people have become so used to MP3 that they accept them uncritically.
So the idea that MP3s are acceptable and that Red Book 16bit/44kHz or HiRez 24bit/96kHz files are redundant strikes me as utterly absurd, so-called comparison tests notwithstanding. It is true that people are nowadays constantly exposed to poor quality sound reproduction to the point of being unable to discriminate between acceptable and impoverished sound reproduction. It is the sad reality that we are exposed to more “music” then ever before, but most of that is really musical noise that only succeeds in blunting our sense of hearing. I reproduce below an excerpt from an article written in 2001 by the musician/composer Mark Polscher that addresses some of the issues relating to the proliferation of MP3s. I used Google Translate and did some minor editing to improve clarity.
8-bit music is produced in 24-bit studios, sold in MP3 quality and consumed through 6 channels
By Mark Polscher
The data-compressing MP3 algorithm has the logical consequence of a general reduced perception of music
What was initially intended as a compromise and concession due to low transmission speed on the internet is establishing itself as a musical norm and not only on the sales and retail side. For many musicians and music producers MP3 has become a reference for artistic production. With growing awareness that 8 bits are completely sufficient for most contemporary music anyway the MP3 format with its tonal inadequacies represents an artistic loss for only a few. AB comparisons to prove that the difference in sound between MP3 and CD is hardly audible are proudly announced.
This is not surprising. Almost all of the music samples used were “composed” from the outset in a multi-evaluation manner and industrialized in terms of sound technology, so that a new compression by MP3 could no longer harm the production of such quality. The fact that the MP3 format was able to spread so successfully in a short time has nothing to do with the original idea of compressing music files so they can be quickly sent over the internet with low bandwidth, but with the fact that the data-compressing MP3 algorithm is the logical consequence of a general lowering of the perception of music.
In times when the decisive cultural trends emanate from the standard pop machine and an incessantly sounding data stream terrorises our world, it is not surprising that music is only used as a permanent karaoke for a universal attitude to life. Admittedly, it doesn’t matter at which tonal and aesthetic level the global soundtrack occurs. MP3 has long since mutated from the Internet file format to the sound standard.
In this light, the whole question of copyright appears like a historical apercu, where entertainment device manufacturers and consumers remain unimpressed. I am afraid that more and more musicians and music producers are constructing their music in terms of the MP3 format right from the start and are completely uncritical as to both industry standards and renewed inferior standardisation. And all without need: the Internet business with music is idle and, more importantly, the Internet is further away than ever from becoming a mass medium. There are various options for non-data-reduced archiving (storage has never been so cheap), and the commercially available CD will undoubtedly remain the medium for music for a long time to come.
But of course the whole point is not with MP3. The fact that music has already been conceived and created with reduced data in its creative phase is the result of a reduced-imagined and position-compressed stance towards the work itself and only clears the way for the standardised sound porridge of our day. As long as composers and producers continue to turn away from the essence of music and give up the creative moment in favour of a consumer-oriented fabrication the door to regression and flattening in music remains open.
All of this is also in grotesque contradiction to the efforts of the music and entertainment electronics manufacturers to blast the high-definition DVD technology in 24 bit and 96 kHz for both the musician and the listener. In the meantime, every Aldi PC has been equipped with a 24-bit converter and the countless samplers in children’s rooms in the western world process their audio snippets in DVD quality. The manufacturers promise the music producers a sound engineering paradise and the music consumer consciousness-enhancing sound experiences in front of the television. On the one hand it is rather uncomfortable to see how you have been trying to define various DVD standards for years, and once again solely and exclusively favor economic and political rituals with this or that technique, on the other hand it is a sad pleasure to experience, how the advertisement suddenly discovered sound and depth as the highest quality feature in music. But what kind of works are they supposed to be? The hit parade in 24 bit and 6 channels for breakfast television?
I doubt that the benefits of DVD and other new sound carriers can be conveyed to music consumers. The system blocked the access itself – and the sad thing is that the musicians, composers and interpreters, who actually concern it the most, contributed to it themselves with their humility towards the phono companies.
MP3 may have been a political move
The fact that the manufacturers of music reproduction equipment and home electronics still get their money’s worth can be attributed to McLuhan’s phenomenon “the medium is the message”: 8-bit music is produced in 24-bit studios, sold in MP3 quality and consumed in 6 channels. When the movement started a few years ago, after some hesitation, I decided to put two of my works on my homepage as an MP3 download. Still intoxicated by the utopian idea of digital democracy and excited about the technical possibilities of having alternative publication and distribution channels, I thought that this was an artistically adequate way of presenting my music to a wider public.
Regardless of the unsatisfactory sound results in the MP3 encoding of my pieces and always hoping that this algorithm was only a transition format, which will soon be replaced by an acceptable further development, it was also important for me to be part of this movement to be and take a position against the phono empire. But that was of little interest and the low audio quality quickly met the demands of both producers and consumers.
In doing so, immensely important musical information is lost during coding: the criteria used in information technology processes for data reduction, redundancy and irrelevance are both used in MP3 technology. The lossy coding of audio material uses the effects of frequency and temporal concealment and forms the basis for reducing the depth effect, dynamics and spatiality to a minimum or eliminating them completely. But these are essential parameters that serve to clarify music and coordinate multi-layered perception for musical deepening.
The fact that there are only a few musicians involved and only a few listeners can and want to perceive this is sad enough, but to say that this is not relevant musical information is simply stupid. The fact that one wants to do without these artistic expressions fits all too well into the one-dimensional thinking and feeling of our day. If music is produced from the outset according to the new industry standard, it certainly says something about the quality of the creative process. If the composer can’t or refuses to perceive losses in coded audio this says something about his sensitivity. If you find this sort of music, you should avoid it.
I have now decided not to make any further MP3 encoding of my music and for the two works “are there two? Are there more than two” and “vme” expressly on my homepage to point out that the recordings, similar to the Real streaming format that I also offer, are only a reference to the studio production, for which an appropriate sound carrier is required. Incidentally, when I download one of my MP3-encoded pieces, I no longer have any copyright concerns, because it is not a digital copy of the work, but only an incomplete reference to it. MP3 could have been a political move. Composers and interpreters could have assumed a high degree of responsibility and self-determination. Instead, however, they again prefer to continue to allow the industry regulate our aesthetic and economic future.
“Tango is a feeling”
Traditional tango dancers in Argentina are often heard saying that tango is a feeling and one has to express that feeling in dancing tango. There is the technique of dancing and there is the culture of the milongas, but in the end all of that is subsidiary to the ultimate goal of tango which is the feeling. Learning dancing technique and tango culture may distract us from this fact because we may suppose that learning these other things will ultimately guide us to attain that feeling, to experience it, and to express it in our dancing. Yet it seems that this outcome is not inevitable and that it is possible for a person to go through the whole process without succeeding in feeling what tango requires. It is not entirely clear whether feeling is something that one should automatically or naturally experience, or whether it is something that is also learned.
One may suppose that the feeling in question is just whatever one feels when one is doing tango. The problem seems to be somewhat like the problem of art: is art some specific type of work that we put in art galleries because it satisfies some artistic standards set by the critics, or as some postmodernist thinkers suggest, is it art simply by virtue of being located in the art gallery. In that case, could a fire hydrant be an artwork simply by virtue of being located in an art gallery?
If we want to claim that some people don’t feel the music we must provide some sort of a standard for telling the correct way of feeling the music. Here we must assume that feeling is not merely an individual and subjective experience, but is something that can be shared and agreed upon. Otherwise, how can two people dance if they have no way of telling whether they agree on the feeling. That means that the the feeling must be expressed in the way the dancer expresses, interprets and hence understands the music, which may be more or less correct or agreed upon.
The idea that the feeling of the music is a purely subjective matter might be in part due to the fact that a lot of music is listened to in silence without any overt expression. The paradigm of listening is people sitting and listening to a symphony. How are we to tell whether they feel the music correctly or not? However, there are fairly clear cases where music either does or fails to make sense to a person, and where different people can agree on whether the music makes sense to them. One such case is the case of tonal vs. atonal music. Most people who are used to tonal music find that they do not understand atonal music.
The latter is music that most people do not understand. Wikipedia defines tonality as follows:
Tonality is the arrangement of pitches and/or chords of a musical work in a hierarchy of perceived relations, stabilities, attractions and directionality. In this hierarchy, the single pitch or triadic chord with the greatest stability is called the tonic. The root of the tonic chord forms the name given to the key; so in the key of C major, the note C is both the tonic of the scale and the root of the tonic chord (which is C–E–G). Simple folk music songs often start and end with the tonic note. The most common use of the term “is to designate the arrangement of musical phenomena around a referential tonic in European music from about 1600 to about 1910”. … Harmony in jazz includes many but not all tonal characteristics of the European common practice period, known as classical music.
In other words, tonal music respects certain relationships between notes whereby a ‘key’ or ‘root’ note establishes the tonality and the functional status of the other notes of the scale. The interval or distance from the tonic note determines the character of the note and what can be done with it, eg., whether it is possible to begin and end a phrase on that tone. The relationships between the tones have a push-pull relationship such as that they create tension, gravity, etc.
Tonality is an organized system of tones (e.g., the tones of a major or minor scale) in which one tone (the tonic) becomes the central point for the remaining tones. The other tones in a tonal piece are all defined in terms of their relationship to the tonic. In tonality, the tonic (tonal center) is the tone of complete relaxation and stability, the target toward which other tones lead. The cadence (coming to rest point) in which the dominant chord or dominant seventh chord resolves to the tonic chord plays an important role in establishing the tonality of a piece.
Because of these features of tonal music, we understand or feel the music when we perceive the relations between the sounds in terms of movement. Movement in music and its various ‘spatial’ characteristics are generated by the relations between the different tones. Musical movement is not inherent in the physical sound. Rather, it is ‘intentional’: we hear movement in the line of notes as connected and moving even though there is nothing that is literally moving in the physical sound and each pitch is separate from another. There may be silence between the notes, but we still hear a continuous line of tones connected in a melody (Scruton 2009).
The intentional character of music is expressed in language through spatial metaphors such as distance and movement. Music moves and it moves faster or slower, moves up or down, etc. Music has rhythm and beat, both of which are to be distinguished from the musical metre. Scruton argues that music can have beat without much rhythm. Rhythm arises out of the music and its accents, and generates the forward motion. Music can have beat without much rhythm, and vice versa, music can have rhythm without a strong beat. In the Western classical tradition up to the 20th century rhythm has been generated by the musical line. With the emergence of the drum kit the beat is introduced from the outside to music that has relatively little of either rhythm or beat.
So what room is there for a person to fail to understand tango music and how would that be manifested in their dancing? If we consider for a moment how one learns to play a piece of music on an instrument, the quality of the playing depends on things like rhythm and expression. Rhythm refers to the grouping of notes, accent, emphasis or suspension. Suspension, playing a note on an offbeat and holding it till the first beat of the next phrase, is present music that is syncopated which includes tango. Expression has to do with the phrasing of notes, the correct gesture, dynamics (playing loud or soft), crescendo (gradually increasing loudness), and so on. A teacher will make gestures to communicate to the student how to play a given passage. Otherwise, the performance will be mechanical and without feeling, like what one might get from software that plays from a written score.
We can see here that learning musical interpretation in dancing is going to be very much similar to learning musical interpretation in playing. Scruton suggests that it is possible to divide music that is influenced by speech and song, and music that is influenced by dancing. If tango emerged as dancing music then its beat and feeling will be strongly connected to the movement of dancing, eg., one should be able to tap one’s foot to it. The music will have a pronounced beat or pulse that’s danceable. There will be a perceptible ‘on’ and ‘off’ beat, and if the rhythm is syncopated, the ‘off’ beat will be emphasised and suspended, driving or dropping into the ‘on’ beat of the next phrase.
Now, different tango orchestras will interpret tango music with different emphasis, with more or less beat, rhythm or expression. This makes the task of interpreting the different recordings in dancing somewhat complicated. However, dance education devotes the majority of the time to movement and little or not time to learning how to listen to tango music. This is largely due to the prevailing false idea that listening is not a skill. One should already know what one likes and that’s the end of the discussion.
But the idea that listening and experiencing music is spontaneous and effortless is a new one, and it has really only arrived at a time of pop. Pop music (including Rock, RnB, Rap, etc.) is really a new kind of thing that departs from prior musical traditions, and moves away from the Western classical tradition. Western classical music has evolved to a high level and requires effort and cultivation in order to understand and appreciate it. Pop is based on a very narrow set of structures, in particular, an ostinato backing which is external to the music and which appears to require no effort to enjoy but is actually addictive. Pop music requires the harmony and melody to be fitted to the ostinato rhythm which absolves the music itself from generating a rhythm through melody and harmony.
Because there is no ostinato backing in classical music or tango the forward motion and rhythm is created through melody and harmony: grouping, accent, and emphasis. Rhythm is here to be distinguished from the musical metre. It is commonplace to hear that while milonga music is 2/4, tango is 4/4, and vals is 3/4. These, however, designate the meter in which the music is written, that is, the way that the musical bar is divided into sections. Rhythm, however, is not created through these and it is possible to have music in these meters that have little or no rhythm. For example, not all music written in 3/4 can be danced to as a waltz. Rhythm is also different from the pulse or beat. The pulse or beat is the point where you would clap or tap with your foot. But rhythm consists of all the up and down beats and stresses that happen in between, so that you have different rhythms with the same sort of beat.
So the first step to learning to dance with feeling, with the correct expression and interpretation, is to learn to listen. Dancing tango is difficult because one has to move with the partner to music. If we dance complicated patterns we will have little attention left to give to the music. Then we will be dancing mechanically, robotically, without expression. At first it is better to develop the practice of listening to a given orchestra without dancing. If you feel like moving you can stand and make gestures with your hands like an orchestra conductor. This gesture will flow into your body and ultimately into your dancing. Start by listening to music you can easily tap your foot to. When you’re finally on the dancefloor, dance to music that you are already somewhat familiar with. Before venturing out, give the music a few moments of attention and have caught onto the gesture that it elicits. This is the gesture that you communicate to your partner as you move into the first step.
Roger Scruton (2009) Understanding Music: Philosophy and Interpretation. Bloomsbury Academic.
Here is a series of videos that are helpful in understanding music:
 Examples of atonal music, which Roger Scruton thinks we don’t really understand because we naturally seek tonality.
 A lecture explaining tonality and atonality with examples:
 Listening to tango music is more like listening to classical than to pop, so the following videos apply to tango as well:
 Interpretation classes with Benjamin Zander are very instructive on how to think about emotion and interpretation in music:
 A lot of useful insights into the history of music (classical music was originally for dancing), structure and layers in music.