Improvisation and Classical Guitar – Part 2

My article on improvisation and classical guitar (Classical Guitar Summer 2016) considers the history and context of classical guitar improvisation (and classical music in general); discusses why and when to improvise; and provides some strategies on how to get started.

I had the distinct pleasure of interviewing three imminent classical guitarists—Roland Dyens, Dušan Bogdanović, and Andrew York—asking them many of the same questions. Herein is an edited transcript of these conversations that expands on many of the techniques and topics discussed in part 1.

Photo by Caroline Strunk

Question - John W. Warren: You are among the few prominent classical guitarists who improvise regularly in concert; you are also among the few concert classical guitarists who also compose music.  Is there a connection—are improvising and composing related?

Roland Dyens: To me yes, they are related, but music is strange enough that you could be a good, or even a great, composer without being able to improvise three notes together. It’s strange but true; I know some friends who are outstanding composers who really don't know how to improvise. And they’re like, “How? How can you do that?” To me, being a composer implies being an improviser, but no, it’s not the truth. But I couldn’t imagine it, for myself, any other way.

Dušan Bogdanović: I’ve written elsewhere (in my book Ex Ovo) that at best, improvisation would be as much structured as composition and that at best, composition would be as fresh and inspired as improvisation. It is also not an accident that many great classical composers were improvisers (such as Bach, Mozart or Chopin) and vice versa. In Renaissance and the Baroque, it was expected of musicians to know how to improvise. We need not even mention jazz or world music.

Andrew York: There is a connection, or maybe multiple connections, of sorts, because both are involved in the creation of something new. It’s a very different use of the mind than practicing, or playing a piece that you’ve been working on and memorizing. It’s quite a different mental process. We use one part of the brain with technique, in executing a piece from a manuscript; what happens in the mind is quite different with both improvisation and composition as compared to performance.

One way to think about it is that improvisation is composition on the spot. That’s not completely true, but that’s one way to look at it.

When we improvise, we’re in the moment, and ideas are coming to us and we’re playing them, hopefully, without too much thought. Someone who improvises well, however, can have a larger temporal sense—a larger time sense—and get an idea of a form developing. It’s a pretty complex thing, when you think about it—as you’re improvising you have to become aware of formal possibilities. At any given moment you’re just in that moment, and playing something on the spot. So to have this larger perspective at the same time you’re in this Zen-like moment in the present.

Q: I would imagine that sometimes your improvisations may lead to compositions, where you start improvising and you later refine it and turn it into a piece?

AY: It certainly happens that way for me. I lot of my pieces were based on or were born out of an improvisation. I don’t know if that happens for you as well. I might be improvising, just playing around on the guitar, I stumble on something that seems really meaningful, kind of poignant, just has a bit of magic to it. I try to record that very quickly on my iPhone, to capture the intent, that emotional moment. Later, I can hear it again and use that as a springboard for developing a theme. But the problem is that as soon as you start messing around with technology, you’re completely out of that state of creation.

I’ve got Pro Tools here at home, but if I have an idea I’m playing around with in the moment, the last thing I want to do is start launching Pro Tools, pushing buttons—the moment is gone. I used to have the cheapest cassette tape recorder that money could buy, and it had one touch recording. The quality was awful, but it captured the intent, and all I really needed to hear was the intent behind the idea. All I had to do was reach out with an index finger, push that button, and keep playing.

Now I use my iPhone, but even that irritates me, because there are a bunch of thumb strokes to get to the recording app, to click it, to make sure it’s working—it’s just enough work that it’s irritating. I think it’s critical to stay in that frame of mind where creation is happening, and not start using parts of the brain that you use to manipulate technology.

Q: You compose and perform music over a wide range of genres, beyond strictly “classical.” Does this genre crossing or genre–bending influence your improvisation, and vice-versa? Is it a requirement?

DB: That is a good question. I actually do not see big borders between genres and for me it is all a spectrum of defined or not-so-defined stylistic characteristics. Surely, improvising a ricercar is not the same as improvising a blues, but at this point I think that the styles are largely a matter of history and marketing and not of creative reality.

AY: Yes, that is a good question. I don’t think it’s a requirement, precisely, because anybody that improvises probably has some experience with different styles. Maybe not many, and maybe not that deep, but they at least have some idea of what different styles sound like. But I don’t think it’s absolutely necessary.

I’ll give you an example. On one of my earlier albums, Centerpeace, about half the album is free improvisation. I was playing with a pianist named Allaudin Mathieu, and he’s one of the best improvisers in the world, I think. What I love about him is that when improvising together there’s no pressure to go with any stylistic direction, which means that he can go in any stylistic direction. If you improvise with a jazz player, you’re going to play jazz, which is great if you’re comfortable with that style. Most people have their bag, their thing that they’re comfortable with, and that’s the way they’re going to improvise. If you’re familiar with a number of styles, maybe you have a number of different stylistic approaches.

But what I like about Mathieu, is he can go anywhere from serialism-sounding academic music, very well formed, to humorous circus music, to beautiful, luscious jazz harmonies. He can go anywhere, like Indian raga, wide open… This is true improvisation. He’s not bound by stylistic considerations. Because ultimately, styles are limitations, aren’t they?

Q: Yes, that’s probably true, but sometimes having a limit, or constraint, might be helpful, in a way.

AY: I would put it a different way: sometimes having preset parameters is helpful. That can actually aid the creative process. Maybe it’s just a semantic thing. Styles are open ended, there’s no limitation as to what can be done within a particular style, but when you bind yourself to a certain set of rules, in a sense you limit yourself from anything outside that set. If you’re playing bebop, you’re using the standard harmonic approach to moving through changes in bebop, and what essentially constitutes the rules of bebop improvisation, you’ve set yourself on that path—you’ve chosen a set along an infinity of sets. So in that sense it’s a limitation. It’s not a limitation in the sense of the quality of music.

This, to me, is a very interesting philosophical part of music, what is music and why at any certain time are certain rules accepted, as the way music is or should be. Music history is the exploration of the evolution of musical rules.

Q: One of the concepts of improvisation is surrendering—letting the music takes you where it will. But on the other hand, we’re still conscious of certain things, like what key signature we are in, what notes will work, keeping a steady beat, etc. How do you balance these?

RD: Yes, I know what you mean. Improvisation presupposes a lot of musical knowledge. You have to have a background in the history of the guitar, geography of the guitar, harmony, counterpoint, etc.

In other words, you can’t just improvise improvisation—you need a deep knowledge base.

I’m not talking about atonality, only noises, playing with timbres, etc. Anyone can do that, but that’s different. We’re not talking about that; we’re talking about an organized improvisation, right?

To have an “organized improvisation,” something that sounds good, it’s like the “little sister of a composition.”

Roland Dyens - Improvisation - 20th International Festival of the Guitar, Cittá di Mottola, 2012

AY: Even though it’s creation, like composition, with improvisation there’s a lot less intellection, at least on the surface. It’s a subconscious process. If you think too much when you improvise, I’m pretty sure you’re going to be a bad improviser—I know because I struggle with this; I have a hard time turning my mind off. If you try to think and control everything, it’s really not a good recipe for creating good improvisation. You might be able to pull it off if you think quickly, but that doesn’t really open the gates to true creativity.

Composition is different, where you need to think about what you’re doing—you want to blend inspiration with all the skills that you have, the knowledge that you have of music, forms, and styles, and you can craft things, and there’s not necessarily the pressure to do it in the moment. You can erase a note when you’re composing; you can try different things. So there are some similarities but there are also some distinct differences.

Q: How did you begin improvising? Was it early in your guitar career? Did you have teachers who taught improvisation as part of their instruction?

DB: I pretty much started improvising and composing at the same time that I started to play guitar. Well, perhaps not immediately, but a few years later. Nobody taught me how to improvise, but I had a violin-guitar duo with my father and I think that that made the practice very lively and spontaneous. When I grew up, I played quite a lot with friends in groups including electric guitar, acoustic jazz and contemporary avant-garde music later on.

RD: I think you are a born improviser; it comes naturally, if at all. When I first got a guitar in my hands, when I was about nine years old, I think the first thing I did was improvise. You can learn, but to do it deeply, it’s part of your nature. It’s like having a good sense of direction, which is not my case. I have a good sense of direction on the neck of my guitar, I think. But even in Paris, going left or right, it’s a tragedy for me! I was my own teacher. I had teachers in other topics, like harmony and counterpoint, but not in improvisation.

AY: For me, it was a personality thing; even as a kid I was fascinated with the idea of creating new things. My earliest memories were always trying to create something fantastic or amazing. So I think it was inevitable. As soon as I could play anything, even a few chords on the guitar, I started writing, cataloging harmonies, even though I didn’t know what they were called yet, I started writing down these shapes and how they seemed to me. I would describe them, give them adjectives, so I could use them. This was before I new anything about [music] theory.

Improvisation is not really something you can decide to do and then learn; it has to be already part of you. Of course, you can enhance your skills by studying with someone who can point you in the right direction, and facilitate your learning process, but I think you already have to have that desire.

Q:  How is improvising in concert different than improvising in practice? You have the audience, but are there other ways it’s different?

AY: On stage, it’s not a comfortable process. When you’re by yourself, you can do anything. On stage, however, there are consequences. When you improvise in front of a lot of people, you hope that it’s going to go well. It doesn’t always go well. I’m not the most consistent improviser in the world. Some people I respect deeply, like Dušan—he’s a real genius at improvisation. His knowledge base is just absolutely vast. He’s got it all at his command. I think I improvise well, but I don’t completely trust myself to do a grade A improvisation every single time. It’s not always going to happen. It’s not like learning a piece to the point where you’re 95 percent sure you’re going to play it really well. With improvisation, I guess I’ll play something well enough that someone won’t complain, hopefully, but sometimes I just know it’s not happening. It’s difficult to find that Zen-like state that allows it to flow. But when it works, that’s the magic, when there’s something happening right at the moment, I get a feeling that I can’t get playing prepared pieces.

Well, that’s not entirely true. At this stage of my life, when I perform anything, be it Bach, or my own pieces, there is sort of a coming together of states of mind, where it’s almost like I’m improvising, even though I know exactly what I’m going to play, because it’s a practiced and prepared piece, but it’s a state of mind that becomes more immediate, more in the present, more Zen-like. That gives a feeling of freshness, like you’re creating it on the spot. This is a direction I’ve been going for decades, where you feel like you’re creating it anew, even though you already know what it is.

DB: Like anything else, playing in front of people is a challenge. Of course, it is not the same responsibility as playing just for yourself, but on the other hand, it is also very inspirational and it can produce good results—after all, you are sharing your experience with a larger entity.

RD: To be honest with you—and I have a lot of defects but I usually try to be honest—most of my improvisations are simply not very good. It’s rare—some times, but very rarely, do I think, wow, this one would have deserved to be a piece. Some times, but rare.

So, to use an example, “Tarantos” by Brouwer, has seven fragments in squares [ed. “Enunciados” (Statements)], represented by roman numerals, I to VII. Another set of passages are represented by letters, A to F [ed. “Falsetas” (in flamenco music, falsetas are short melodies played by the guitarist between verses)]. The rules of the game, as it were, consists in associating the letters with the numbers, but you can do this however you want, in concert or on a recording—you can pick different letters with different numbers: i.e. VI • A • III • C • I • B, like that.

When I improvise, it’s not exactly the same, but it’s similar; what I’m playing are some things I know that I think might work together, some of the thousands of things I know might sound good. So it’s not completely random. This will fit this, or this will match that. Rarely is it pure creation, sometimes, yes, because I do get inspired. I believe in inspiration, it’s rare, but it exists.

But the problem is, in concert, you have to be a genius now.

I love to improvise, and through the years, it became almost a superstition, to start with an improvisation. It’s the best way for me to check the sound, the reaction of the public, myself, my guitar. To warm up a bit, feel the guitar.

I’m not inventing anything here; it was a deep tradition in the past, as you know, with lute players. The “prelude” was an improvisation, it meant “before the beginning” in Latin. I’m making the same connection.

Q: It’s a big risk to improvise in concert.  Do you usually have a short segment—such as a few bars or a phrase—to start with and go from there as the muse takes you, or do you start completely fresh with nothing in mind?

RD: It’s a risk, but I’m glad to do that. It’s a luxury really. I can’t think of anything better to warm up and feel the ambience.

DB: For some people it is a big risk to improvise, for some people it is a big risk to play written music (think of Keith Jarret, for example, who always hangs on to the written score when he plays classical music). I tend to oscillate between the two, but actually very often I feel the freest when I start from nowhere, but that really depends on the moment.

For a while I have developed a tactic where I just “noodle” on something before the concert and then I simply continue with the same idea or atmosphere as I enter the stage. That way, there is no separation between the spaces, as ideally there shouldn’t be anyway—at least that is what I think.

AY: I love it that Roland starts his concerts with an improvisation. I sometimes do that too. There’s something comforting about coming out and just beginning to play. There’s actually less pressure, in a way, because once you begin, you can do it as long as you want, you can segue into something, you can do anything. It’s difficult to improvise in the middle of a piece, because if you’ve already got this structure around it, and you have to fit your improvisation into that. I do that as well, but in some ways, it’s more difficult.

I rarely have anything prepared. It has happened, like sometimes backstage a phrase may occur to me, and I think, “That’s cool;” that’s happened to me more than once. But often, it’s just like a tabula rasa, a blank state, I just hit a note and I try not to know where I’m going.

I’ve noticed that it’s dangerous to have something in mind to improvise over—often it becomes derivative. If I improvise something back stage, then I try to recreate that, it changes the way you play. It’s sort of a double-edged sword, because once you’re in the process of re-creation, you’re no longer in the process of creation.

Q:  One of my ideas on this topic is that guitarists can find certain pieces that work well because they may have “launching off” places for improvisation. For me, Brazilian music works well—the music of Baden Powell comes especially to mind.

RD: Yes, absolutely. Brazilian music to me is just like jazz, to me, the difference is rhythm; you have lot of 6/8 vs. 2/4 or 4/4 time, the kind of rhythm.

You know, the most difficult thing is the instrument itself, the guitar. Especially in the case of classical guitar—this is a big issue, because it’s not about improvising while someone else plays the chords for you; it’s doing it by yourself. This is the crux of the issue.

On classical guitar, we have to do two parts at the same time, the melody and the harmony. On piano, things are so much easier and clearer. On the piano, the right hand can play the melody and the left plays the harmony, the accompaniment. We don’t have that on the guitar. It’s much more difficult; you cannot compare. Maybe, somehow, I’m a frustrated pianist, because in some ways, in my improvisations, I try to recreate the piano on the guitar.

In my teaching, for example, I notice that most players play the accompaniment way too loud; I tell them to do it like the piano. For example, in a “Sonata” by Manuel Ponce, I tell them to play the harmonic background softer, until it’s almost nothing, instead of playing the accompaniment and the melody at the same level.

On piano, clarity is easier; on the guitar, wherever you travel on the neck, even up high on the neck, you have to keep the accompaniment going—that’s very complex and demanding. Also, on the piano you have the pedal, so it’s easier to create sustain. On the guitar, to improvise or even just to play well, you have to be a sort of wizard, a magician, by creating the illusion that the notes are long lasting—which is not the case. You create the illusion that the notes are longer than they really are, that is one of the secrets to improvisation.

The image of the piano is in my head when I’m improvising.

Roland Dyens

Q: Improvisation is familiar in jazz guitar, rock, flamenco, etc., so why do you think there is a lack of standard practice and instruction in classical guitar (and classical music in general)?

RD: Two little dots… I damn don’t know. I just don’t know. Yes, you can extend this to classical musicians in general. Maybe it’s more surprising that this is the case for players of a polyphonic instrument. I’m not sure, but it seems classical pianists are more predisposed to improvisation than classical guitarists.

DB: Mostly it has to do with the historical development of classical music, but to simplify, I think that the specialization of performer via composer is the main “culprit”.  The structure in classical and contemporary music is so complex and demands so much attention and technique that it is very tough to match that with symmetric work on improvising. My opinion is that one has to sacrifice one in order to gain the other. It’s a matter of personal choice and there is no recipe for the right balance; every person finds what works for him/her.

I do think that classical musicians and guitarists in particular should have consistent instruction in improvising and I see some signs of this, especially in the US. Europe is more dependent on tradition and I think that it will take time to change that, but most new things require effort. Early music is a good example.

AY: I can only speculate, but one reason is that as the pedagogy increased, and classical guitarists were playing more and more complex pieces, it took a lot of effort to play at this high of a technical level. If you spend all your time learning technique, you don’t have a lot of time to study the guitar in a well-rounded way, which means learning harmony and other concepts necessary for improvisation, which are very different. It takes an equal amount of energy and time; it’s not a casual thing to really do it well.

The other thing is it’s a lot easier to be an extremely proficient technician than a very good improviser or composer. Just look at the numbers. Every year, we’ve got a deluge of hotshot young musicians winning competitions, right? Yet every year, you don’t have a bunch of great composers emerging on the scene, you just don’t. It’s another thing entirely. It’s much, much more difficult to create good music than it is to play the guitar well. I don’t know if people realize this. It takes a lot to be a great player, but it takes, I think, exponentially more skills to be a great composer.

If it sounds derogatory, it’s not; it’s just a fact.

Improvisation, as we all know, used to be part of being a classical musician, but only organists seem to have kept that tradition alive; they still learn improvisation as part of their craft.

Andrew York, photo by Terry De Wolf

Q: In classical music, like in the baroque period, there was improvisation, but then things got more codified, set in stone.

RD: Yes, the best example of this is the cadenza in a concerto—it was originally a part in between 2nd and 3rd movement where the interpreter would show off his skills as an improviser, so the soloist would show how he could improvise. But through the years the cadenza became written down by the composer, and instead of being free it was stratified. So I guess you can blame those first composers who wrote those parts down and made it fixed instead of improvised.

I remember when I was about twenty years old, I played my first big Parisian concert, and I was going to play a Concerto by Villa-Lobos—it was also my final exam for my studies—and I both improvised the cadenza and played the cadenza that was written down by Villa-Lobos. I couldn’t avoid the official one because of the requirements of my exam, so I played two cadenzas.

Q: Andrew, you have a personal interest and depth of knowledge in math, including topics such as equations and the natural word, software and encryption. Does this have an influence over your improvisation—for example, do you utilize forms of equations or mathematical progressions in your improvisations, as they have influenced your compositions, such as your recent six-movement suite "The Equations of Beauty"?

AY: I think it does, simply because I have an interest in them. Even if I didn’t know anything about math, I would probably intuitively use the ideas of forms and sets, because music is usually ordered in a way that can be considered mathematical. Musicians are typically good at math; one way I like to think of it is they’re good at subconscious math… Humans are musical creatures, every culture is musical, and music essential does have at its roots a mathematical base—it’s based on ratios, vibrations, forms, rhythm, these are all mathematically accessible. When people are naturally good at music, that’s kind of a natural mathematics. It doesn’t mean they know anything about math, per se, it’s still an expression of what mathematics represents. When you learn math, you go from a subconscious appreciation of it, or a subconscious utilization of mathematical processes, to a conscious awareness of how they work. It’s actually a huge leap. It fascinates me to know that musicians are good at math, because they’ve used these areas of the brain, without any kind of conscious awareness of what they’re doing.

Andrew York performs "The Equations of Beauty" suite at Guitar Salon International (Guitar: 1957 Hermann Hauser II, ex. Bream)

Q: What advice do you have for classical guitarists who want to get started improvising on guitar? Are there particular keys that are easier to start with?

AY: I think that one of the best bases on which to start improvising is having a really good ear. So developing your ear in a way that you can hear the different qualities of each pitch in relation to the whole piece. In other words, if you’re in the key of G, and you play a G major scale over it, each note of the scale should have a noticeably different quality to you, or a different color. The more aware you are of the way notes relate to a home key, the better you can improvise. If you can’t play what you hear in your mind, you can’t really improvise.

Of course, you have to know the guitar very well, but first, you need to hear the notes in your mind and recognize how each pitch relates to the others.

When I do workshops on this, I talk about these ratios, like a G, and a D, the fifth, a very simple ratio—a ratio of 3 to 2—a beautiful sound, very open and powerful. Works great for rock and roll, too, for example, there’s nothing like a rock and roll guitar playing a fifth, with all its harmonics it can be very pure. And you continue on from there, what does the fourth sound like, a C, the major third. Every one should have a unique, identifiable color. When you get good at that, you can begin to improvise without so much fear, because you recognize what you hear in your mind and translate that onto the instrument.

So the best advice I can give is to develop your ear.

RD: It’s very complex, and somehow disappointing, because there are no tricks, or tips, as you say, to improvise. I tried, long ago, to do workshops or master classes on improvisation, but I gave up. I had very good players in front of me, very good, who were unable to improvise after two or three workshops, and, on the other hand, very humble players who had a strong intuition and could improvise very freely. So despite what I brought them—my teaching, ideas, and suggestions—it didn’t work. It’s nondemocratic.

But, if anything, the first step to learn improvisation, and perhaps most important, is to learn how to accompany songs. And then get free of that and try to do things with the harmony, to discover the inner voices of the song, contrapuntal motion in polyphonic music. This is good for popular songs. We have a strong tradition of French chanson, like Georges Brassens, Édith Piaf, Jacques Brel. Or a similar thing is with jazz standards, because as you know, most jazz standards used to be popular songs, like “Over the Rainbow.” The songs in general in Brazil are also popular songs, so it’s the same that way too.

DB: As in anything, just start with inspiration and sincerity. As far as particular keys, from Blues and Carcassi studies to Britten’s Nocturnal all roads lead to E, both major and minor.