When watching videos, think about what’s outside the frame! In a musical context a piece can come across differently depending on how elements are mixed or what the performers / conductor choose to bring out.

This is something I’ve been noticing in my own life - over the past couple of years I’ve been saving some of my family’s home videos from my childhood on my computer to preserve them from the forms they were in, and going back and watching some of them again, and it’s been an interesting experience. I have memories of some of the things we have video of, but many of the videos were of things I did and the videos were recorded by my parents, such that the angle and position that I see them from in the video do not correspond to my first person memories, which can be a little jarring. I also only have videos of certain things. I grew up mostly in the era before smartphones (I got my first smartphone when I was 20, and my first iPhone the year after I graduated college), so we had a couple of tape-based video cameras but they were clunky and we didn’t have them all the time. We only really had videos of major events in our lives - our recitals, Christmas and birthdays, etc. These days I have lots of little videos of rather more mundane, day-to-day parts of my life that we couldn’t realistically film back then, so the parts of my life that the videos help me “remember” are skewed. I’ve also come to the conclusion that some things might have been better left as memories, but that’s a story for another time 🙂.

Connecting this back to music, an audio recording captures a specific version of a performance of a piece of music, and the way in which it is captured and mixed can greatly change the way in which it comes across compared to how it might in a live performance context. In some cases, the recording is the only version of the piece of music in sonic form, as it was never intended (and may not be possible) to be performed live. This applies particularly to many forms of electronic music, and to a lesser extent to commercial music like film scores. Many of those could be more or less performed live, they just often won’t be, but many kinds of electronic music are made with computer-based synthesizers and stitched together in midi in a sequencer and bounced out to an audio file to be shared on a platform like SoundCloud or the commercial streaming platforms (or maybe even a CD or vinyl record), and there never was a live performance of the piece. You can argue that there actually is a form of live performance of a lot of that, which is a DJ playing it at a dance club, where the atmosphere of the club and the people dancing brings something of live energy to the experience that you wouldn’t have just playing it on your computer or phone, but that’s a different kind of live performance, as the music track itself is usually pre-recorded and would sound the same each time.

In musical genres where live performances are more common, such as most acoustic music settings, the genres themselves often to greater or lesser degrees expect music to be more like “real-life” as described above, in the sense that no two performances of a piece will be the same. Certain genres, like classical music, will be on the “lesser” side of that spectrum, where the performer is often expected to practice a particular piece until they can execute it roughly the same way every time. This can actually get to be so intense as to become crippling, leading many classical performers to have bad cases of stage fright, particularly if they had a recent experience where they forgot a passage during a performance, or in some other way screwed up what they were trying to do. I think some of this is also actually “aided” by listening to a lot of recordings, which are often heavily edited to ensure that every passage is perfect even if no particular playthrough of the piece was perfect, and it sometimes can be useful to attend more live performances with professional ensembles to be reminded that even the pros don’t always have perfect performances. In this sort of music, one should try to be as accurate as possible, but it might be unhealthy to take it to the extreme of “making any mistake at all ruined the night”. In most contexts the audience may not even notice, and if they do they are often more forgiving than some of us are led to believe in our training. Auditions and competitions are not the only performance experiences, and not even representative of most gigs in real-life anyway!

Other genres will use various forms of improvisation to greater or lesser degrees (see my forthcoming essay “Spectrum of Improvisation” for more on that once it’s posted). In these styles, it would actually be seen as bad form to have any two performances of a piece be identical, as that defeats the spirit of improvisation in these forms.

In classical music, there is another version of “living” music, which is minor fluctuations in tempo relative to strict “metronome” tempo. This applies to most forms of non-electronic music, where groove is more flexible than a fully-quantized tempo grid in a sequencer, and that can sound robotic and unnatural. Even in music where the tempo doesn’t officially change (like a lot of non-electronic dance music in various genres), some level of fluctuation is expected from measure to measure and phrase to phrase. In those contexts it would be considered the music “breathing”, and as such shouldn’t really be noticeable unless you tried to count strict metronome time or actually put a metronome on it. In other pieces, the tempo will change explicitly, either between sections or even over time within a section, and of course it should be noticeable in those contexts if people are paying attention to the music.

Another related version of living music applies even to multiple performances of the same classical piece, as interpreted by different players or even by the same person over the course of their lives. To witness this in action, pick any conductor who recorded more than one cycle of the same set of pieces (all the Beethoven symphonies would be a good choice, or all the Brahms symphonies, etc). You will find that some of them have radically different tempos for the same pieces at different points in their careers, and other elements will change too - how much to bring out various counterlines, how smooth (or not) to make certain passages, etc. This also applies to solo performers who record, say, solo piano works, or the Bach Violin Sonatas or Bach Cello Suites over time. See below for a mini-study I prepared about this.

I’ve even found this in my own piano playing as I learn a piece. I’ve found that as I get more comfortable with a piece, and particularly as I memorize a piece, my “ideal” performance of a piece changes. Not just what I can play, but how I think a good performance should be whether or not I can achieve it yet. As a general rule, my ideal tempo increases as I learn a piece, compared to my impression of it before I learned it myself, or as I was working on it. Again, this is separate from doing “slow practice” where I’m intentionally playing it slower than I’d want to ultimately. This is my target speed itself that’s changing.

Celtic music is a good example of one of the genres where the music shouldn’t be performed the same way each time. One of the problems with using tunebooks to learn some of the tunes is that this encourages the mindset that there is one correct way to play a given tune, especially for people coming from a classical background. But actually, each tune can take on many forms, and one thing you need to do when learning this music on any instrument is to learn how to vary the music successfully using your instrument. The precise techniques will vary from instrument to instrument, but they will all basically come in the form of ornamentation, melodic variation, and phrasing (slurs, connecting notes to each other or leaving space around them, etc). Once you know some of these techniques, then when you learn a tune you need to learn where in a given tune you can introduce variations. Listening to several players playing a tune, or even to one player playing a tune several times through, will begin to show how this works. Grey Larsen has a fantastic book called The Essential Guide to Irish Flute and Tin Whistle, which goes into the most detail of any book I’ve ever found for any instrument on these subjects, and is recommended even for people who don’t play those instruments. He’s focusing on Irish Traditional Music in that book, but similar concepts apply broadly to Scottish music and its offshoots in Cape Breton (Canada) and Shetland Island music, and even to Old-Time American music, though the details will differ. This leads to this music feeling freer in many ways than classical music from a performer’s perspective, and learning something about that genre (or any number of other genres) if you are a classical musician can really expand your musical mindset (one point of Multi-Genreism in the Capital-M Music approach).

All of this comes together to make a living, evolving interpretation and performance of a piece of music, which cannot be captured by a single recording of a piece. So while I still say that a recording of a piece is “real” existing music compared to something like sheet music, or like a player piano roll that is not currently actively being used, a recording is not a “living” piece of music. Only a live performance can be that, and in some cases, only a set of live performances over time can really show that. They can each be recorded, of course, and then each one is a snapshot or “video” in time of a specific interpretation, but the piece of music itself lives beyond any one of the performances, and certainly beyond the fixed ink on the page of a piece of sheet music.

P.S. to add - there is another way in which a recording could be said to have a kind of life of its own, and that is in our relationship to it over the course of our lives. There are particular recordings that I have strong associations with because of the context in which I first heard them, either the time of life I was in or other events surrounding my initial experience of the recording. Anecdotally, it seems like most people have a stronger relationship with music they first experienced around jr. high - college years of their lives (basically as a teenager or a little after that), compared to music we first hear in our late 20s or 30s or later. The extent to which this is true may vary from one individual to the next - I suspect that musicians have a different relationship to music they hear later than non-musicians do, though even in my case I can tell that I know music better that I learned several years ago than tends to be the case with music I first learn today in my early 30s. But I will also note that some songs, both classical pieces and non-classical songs, speak to me in a different way today than they did earlier, and there are certain songs that I didn’t really understand 5 or 10 years ago that I relate to better today with more life experience, which will no doubt continue in the future as well. This is one reason why a person’s interpretation of a given piece of music changes over their lifetimes in a live performance context, but it can also influence our experience of the same recording, even if the recording itself hasn’t changed, because we have.

-- --

Here’s an excerpt from a forthcoming guide to rhythm that I’m in the process of putting together. This mini-study examines different conductors and orchestras interpreting the same piece of music to show just how different it can be even in styles that pride themselves on sticking to the “true intentions” of the composer.

-- --

In classical music, the tempo is not usually set by the composer, at least not in common practice music (roughly music from the 18th and 19th centuries in Europe). Or rather, there is more than one tempo that could be considered “right”. Even when composers (or at least the scores) provide metronome marks, you will hear a wide range of tempos from different professional conductors and orchestras, not to mention student and amateur groups! To demonstrate this, I’m going to pick three recordings of the same piece by three different highly respected ensembles and conductors. I happen to like one of these tempos and not quite as much the other two, but we can talk about it a bit after I show them.

This is Franz von Suppe’s Light Cavalry Overture, one of my favorite pieces of all time for very personal reasons - this was the first piece I ever specifically asked my dad to get me a recording of after hearing one of my friends play it in the Chicago Youth Symphony Orchestra when I was somewhere around 8 or 9 years old. I loved it that day, I still love it today, and I’ve now gotten to play it three times myself!

For this demonstration, I’ve picked a recording from the Berlin Philharmonic with Herbert von Karajan as the conductor, and then another recording from the Academy of St. Martin in the Fields with Sir Neville Marriner as the conductor. I was hoping to find one with Sir Georg Solti and the Chicago Symphony Orchestra (my hometown symphony!), which is my other favorite orchestra along with these other two, but it doesn’t look like Solti recorded this piece with them, so I’ve pulled one with Solti and the Vienna Philharmonic, another great orchestra, which will certainly demonstrate the point I’m trying to make here. Take a listen to at least the opening of each of them, and then I’ll talk about it below:

That may have been an eye-opening experience for you if you haven’t done this kind of back-to-back comparison of multiple versions of the same piece before, especially if you come primarily from the school that says “we must do exactly what’s on the page and nothing else”. For reference, the score says q=80 for this opening, and Marriner is close to that - about 76 to 78. Herbert von Karajan is more like 65, and Solti is somewhere around 90, though they are all using a bit of rubato so it’s hard to clock it exactly. As we get further into the piece, von Karajan raises the tempo to roughly match Marriner in the faster parts (starting around the wind steady 8th notes at about 2:15 in von Karajan’s recording). He sticks with the faster tempo until the big string unison (4:38 in his recording), where he slows down substantially again compared to Marriner, before once again roughly matching Marriner in the last fast part. Solti takes the tempo faster than Marriner throughout the entire piece.

As you can hear and see, von Karajan takes the piece significantly slower than Marriner does in the slower sections of the piece, so much so that it almost sounds like two different pieces to me. I’ve seen this phenomenon with some other pieces too, and sometimes it almost feels to me like von Karajan’s tempos are so slow that the melody gets lost a little bit. Solti’s tempos, on the other hand, are about as fast as I think one could reasonably get away with in this piece, and are maybe a little too fast for my taste.

I tend in general to like Marriner’s tempos on most pieces, as well as other aspects of the Academy’s performances, to the point that nowadays if I’m looking for a recording of a new piece I’m working on or just want to listen to, I default to the Academy of St. Martin in the Fields if I can find a recording from them. I like a lot of von Karajan and the Berlin Philharmonic’s output too - the recordings are always technically brilliant, the playing is always great, but I just don’t always like his tempos. Out of all of the “recent” conductors of the Chicago Symphony (back to Fritz Reiner, at least), I tend to like Solti’s recordings the best, between the raw sound and playing quality and the tempos he picks. I even met a few of the players over the years I lived in the area - several of them lived in my high school district and I knew some of their children growing up, which is one reason I like the group so much (that and they are one of the top orchestras in the world!), and why I was hoping to find a recording of them for this study. But as a general rule, if Solti takes a different tempo than I like, he tends to go faster than I like, and this recording with the Vienna Philharmonic certainly demonstrates that aspect of his conducting.

I’ve been working on rhythm exercises with some of my tutoring students at CSUN, and given the complexity of some of the rhythms, we’ve been trying them slowly. But I’ve found that some of my students want to go so slowly that the rhythms devolve into individual notes and lose any sense of groove. There is a place for super-slow practice if you recognize that you are basically getting rid of rhythm and you know why you are doing it, but in this case since these are unpitched rhythm exercises, I don’t think there’s much point in taking things quite that slowly. I want to be sure that we can still sense the overall groove of the piece as a groove, even if we slow it down from performance tempo for practice. von Karajan isn’t that slow in this recording, but I think he might be pushing the bounds of acceptable tempo for this particular piece. Marriner’s tempos feel right to me in this case (I’ve phrased it before as “he’s taking the tempo I would want to take if I was conducting it”).