Ep18. Can improvisation be recognized by listeners?

Discussion

---

Introduction to the controversy

There is some controversy over whether it is possible to tell while listening in real-time whether a musician is or is not improvising. Jazz and Rock music college textbook's author (The History and Tradition of Jazz 6th ed., Modern Sounds: The Artistry of Contemporary Jazz 2nd ed., and The History of Rock and Roll 6th ed.), Tom Larson (Assistant Professor of Composition (Emerging Media and Digital Arts) at the Glenn Korff School of Music, University of Nebraska-Lincoln) maintains that “most of the time, it is easy to tell” (quoted in full context below).

Composer of instrumental and electronic music, teacher of courses in composition, theory, computer music, and director of the Gassmann Electronic Music Studios and the Gassmann Electronic Music Series, Christopher Dobrian, argues that audience recognition of improvisations can be “central to the performance.” (quoted in context below)

Opposed to Larson and Dobrian, some theorists, such as Harvard professor and well-known jazz musician Vijay Iyer denies that one can tell just by listening whether someone is improvising or not. Let's get into the details of their respective positions in the order just presented.

---

---

In his textbook, Thomas E. Larson History and Tradition in Jazz (2002) explicitly states improvisation consists of "simultaneously composing and performing.” If one is composing, then one is concurrently producing a composition. Larson believes when listening to a small jazz group that it is “usually easy to tell which member of the group is improvising a solo.” He puts it this way:

“Improvisation is defined as the act of simultaneously composing and performing. It is an essential element in the performance of most, but not all, jazz. For instance, most of what you hear when you listen to a jazz big band is written down and not improvised. But jazz is an art form of individual expression, and most jazz contains a great deal of improvisation. When you listen to a small jazz group, it is usually easy to tell which member of the group is improvising a solo, but it is important to remember that the other members of the group are also improvising within the framework of their responsibilities to the group sound.” (bold and bold italic not in original)

Christoper Dobrian, Professor of Integrated Composition, Improvisation, and Technology in the Music Department of the Claire Trevor School of the Arts at the University of California, Irvine, argues that some improvisations are recognized by audience members during a performance.

“The aesthetic power of a sense of spontaneity is important to most performance. In a performance of composed music or of a written play, the audience suspends, at least to some degree, its knowledge that the performance has been largely predetermined and rehearsed at great length. If the performer does a successful job of "acting"—of discovering the subtext implied by the text and convincing the audience that all actions are entirely motivated by that subtext—the audience is helped to forget that the performance is the result of laborious preparation. In improvised music the sense of spontaneity is mostly genuine and often quite evidently so. An audience can determine, by musical and visual cues, certain interactions between players or certain musical phenomena that could not possibly have been planned or foreseen. The appreciation of the spontaneity is therefore not forced and becomes central to the performance.” (bold and bold italic not in original)

There are several significant things to notice about Dobrian's positions. First, Dobrian believes that improvisations can often be recognized by an audience based upon musical cues as well as other contextual factors combining with the extreme unlikelihood that the musical interactions one sees and hears could have been previously rehearsed.

To address this question it is a good idea to learn more about a composing duo who write music so that the opposite of how it seems is what is presented. These mischievous performers ave improvisational elements that seem pre-rehearsed and rehearsed components that appear improvised. This duo of Edwin Hawkins and Phillip Parson are known as [THE].

“The performance duo [THE] (Edwin Harkins and Philip Larson) plays very successfully and cleverly with an audience's conceptions and misconceptions about what is spontaneous and what is not. Their performances include composed mistakes (simulated unintentional spontaneity), pretense that actual mistakes were in fact composed and intentional, unison activities that appear coincidental but were in fact accomplished by secret cues to each other, improvisations which--because of their intuitive understanding of each other--seem precomposed, etc. The aesthetic experience for the audience is one of appreciating being fooled, or appreciating seeing through pretense, sharing "inside" jokes, or simply being uncertain as to whether the performers are sincere or mocking. Furthermore, the aesthetic experience for each audience member may be vastly different, depending on that person's prior experiences and knowledge--depending on whether one "gets" the point, or gets the hidden (possibly conflicting) subtext, or both. Elements such as those discussed in this section--spontaneity, difficulty, predetermination, intention--which are implicit in more traditional performances, are explicity the subject of a [THE] performance.” (bold not in original)

Second, is his hyperbolic use of possibility/impossibility that most non-philosophers consider an UNacceptable way to use the terms possible or impossible. The vast majority of American English language users, certainly in the 90% and higher, have a looser sense to standards of what us and us not possible. when they say that's impossible only mean to ve asserting in virtually all actual cases that the event to which they refer us highly unlikely. They seldo, if ever, use the term in the philosopher' style sense of being not pissibke because self-contradictory.

A second misuse of impossibility is when Americans say and they think that they mean it that, "Everything is possible." Fortunately, everything is not possible. For example, anything that is impossible is not possible. An example of the impossible is for one object to exist and not exist at the same time. Now defenders of the view that everything is possible may wish to argue that the alleged counter-example is not actually a thing. It is instead a compound statement with the same subject but with opposing predications attributed simultaneously where both conjuncts of the compound statement cannot simultaneously be true (or false) since they attribute incompatible predications. Since

Vijay Iyer commits himself to the impossibility of recognizing improvisations in his article "Improvisation: Terms and Conditions" (2009).

“Improvisation. If we define it provisionally as real-time decisions and actions, then what isn't improvisation? We're improvising from the moment we acquire sensation and motion—you could say it's prenatal. The process by which we acquire cognition—embodied action, situated in an information-rich environment—is improvisation itself. There is a fundamental identity between improvisation and what we more generally call "experience." They are one and the same. Life begins at improvisation. Life is a sustained improvisational interaction with the structures of the world, of the body, of culture. Improvisation is a condition of being alive. (p. 171)

Composition and Performance. Exceptions to the above immediately come to mind: for example, the composition of works to be performed to the composer's exact specifications, and the performance of such compositions. This composer-performer bifurcation is among the more peculiar hierarchical differentiations we have in culture. . . . However, this claim does reposition composition as highly unordinary in the scheme of things. This phenomenon--the micromanipulation of the actions of others at the behest of a single composer-ego—is accepted unproblematically as a structural necessity of music, dance, and theater in the West. We swear by this order of things, so much so that we tend to view improvisation as a rare and dangerous behavior. We tend to forget that the performance of fully composed works is an "extreme occasion," to use Edward Said's phrase, whereas improvisation is not only everyday and ubiquitous, but moreover an utterly central phenomenon in our lives, fundamental to who and what we are. (pp. 171—172 with the following six paragraphs omitted)

A recurring conceit among classical musicians, critics, and listeners is that the best performances of composed works are those that "sound improvised." (What, I always wonder, are the hallmarks of that sound? What exactly does improvisation sound like and how do we know that it's occurring? The short answer, it seems to me, is that we don't; more on this below.) (p. 172)

The "Whether" Question. It is common to ask "whether" a musical act was "truly" improvised. . . . Another complicating factor is that there's nothing inherent in musical improvisation that "sounds" improvised. It's easier to identify things that "sound composed": ensemble synchronies and unisons, very difficult to achieve spontaneously, typically signal to a listener that something was planned. In talking about improvisation, witnesses tend to invoke performers' body language, eye contact, visual cues, or of so-called "mistakes" to prove that these acts were (or were not) spontaneous. But short of such things (which are not essential attributes of improvisation) it seems (end of p. 173)

awfully difficult to pinpoint improvisation as such. There are reliable attributes of a specific improvised idiom—culturally specific principles of variation, parameters of expression, and choices to be made—but nothing in the sound of a musical action that announces, "Yes, I was 'authentically' improvised! You can tell just by hearing me . . . I came into existence in the moment that you heard me. It was only just decided that I should exist. My creator chose me over all the options within reach at that time." And so on ad absurdum. Listeners often want to decode the level of real-time agency in what they just heard, so they tend to ask questions like, "What percentage of that was improvised?" This would seem to suggest that, at least to the uninitiated, it's impossible to know just from listening. And if this is indeed the case, then how does anyone ever really know? There's certainly no point in asking what was "intended" in an improvisation, or what an improviser was "thinking" at that moment. And we still have that "impression" of improvisation discussed earlier, that improvised "quality" that is supposedly conjured by the best classical performers. (It is also common, conversely, to praise improvisations for the apparent presence of compositional elements: motific developments, a "natural" arc, and so on.) If a listener can knowingly allow himself to believe that a fully composed work has just been improvised into existence, or that an improvisation was instead composed, then of what use is the distinction at all? But it is a valuable and meaningful distinction. It brings us to a central paradox: the drama of improvised music involves the understanding that those sounds were chosen and deployed at that moment by those people. And yet, you cannot tell this to be true just by listening; you have to already know that this is happening. It follows that you only really know by referring to something beyond the sound. (p. 174)

Improvisation as a Condition. What I submit (after George Lewis and others) is that we agree to understand improvisation not as a quality but as a condition. . . . Is "improvisedness" audible? It doesn't matter. To paraphrase Peter Falk's character in "Wings of Desire": I can't hear it, but I know it's there. (p. 175) (bold and bold italic not in original)

---

---

Critique of Iyer

Iyer has a point that there are no intrinsic sonic marks that reliably indicate when an improvisation occurs. “There's nothing inherent in musical improvisation that "sounds" improvised.” On the other hand, Iyer agrees that when one hears “ensemble synchronies and unisons,” that these may sound composed because of the higher probability that such coordinated effort among multiple independent musicians is more likely to have been planned out and pre-composed prior to their performance.

There are two sides to this same coin 🪙 then. If probabilities are involved of the likelihood of pre-planning and pre-composing when there are extreme coordination and intricate integration of musical passages, then, on the other side of the coin, is where the probabilities also increase because of the likelihood that when what is played is so erratic, so wild, so in-the-moment that it is more likely it is improvised than that it had been planned and pre-composed beforehand. Iyer cannot have it only going one way. If probabilities increase for composition from highly coordinated musical ensembles, then the opposite must be true as well. Extreme sounds are more likely not pre-composed but improvised.

Iyer strives to persuade that improvisations cannot be heard by sound listening only. But this is too restrictive. One also cannot tell from only listening to the sound the meaning of the sentence "The man was fishing off of the bank" as to whether the man was fishing 🎣 on the side of a river 💦 or off of the top of a financial institution 🏦. Iyer restricts us too much. There is no reason not to permit other features from aiding in improvisation recognition besides the sound. Indeed, Iyer perhaps inadvertently concedes quite a bit when he writes that “"What percentage of that was improvised?" This would seem to suggest that, at least to the uninitiated, it's impossible to know just from listening. And if this is indeed the case, then how does anyone ever really know?” Didn't he now concede that the initiated can recognize improvisations when they occur from only listening to the sounds? Who are the initiated? This will be a matter of degree from the highly initiated, to medium, to low, to not initiated. The more initiated, the easier it will be to recognize when improvisations happen.

The author is also too skeptical in his pronouncements. Merely because some one person, the uninitiated, cannot tell or 'know' when improvisation happens, it does not follow that no one can know, which Iyer appears to concede when referencing the initiated. Additionally, since the issue is over whether people can recognize an improvisation when it occurs, there are no restrictions on how we do it, and Iyer certainly seems to think it can be done (“It's easier to identify things that "sound composed": ensemble synchronies and unisons, very difficult to achieve spontaneously, typically signal to a listener that something was planned”), although it may have to be through some non-sonic means that he lists: “performers' body language, eye contact, visual cues, or of so-called "mistakes" to prove that these acts were (or were not) spontaneous. But short of such things (which are not essential attributes of improvisation) it seems awfully difficult to pinpoint improvisation as such.” So, please don't cut them short; use them, and they provide evidence and increase the likelihood that when musicians extemporize others can recognize it.

CONCLUSION: Therefore, Iyer's claim is mistaken that it is impossible or 'awfully' difficult in the extreme to hear improvisations as improvisations.

Iyer most likely believes that it is not possible to know with certainty from just listening whether or not a musical passage has been pre-composed. It is possible to have ensemble synchronies and unisons that were improvised. Such improvisational unison ensemble playing may well occur in the early work of Ornette Coleman and Don Cherry. Cherry's tremendous ear and timing, and his extreme familiarity with Coleman's approaches to music-making, may well have sometimes permitted unison improvisations that were not pre-composed.

“There are reliable attributes of a specific improvised idiom—culturally specific principles of variation, parameters of expression, and choices to be made—but nothing in the sound of a musical action that announces, "Yes, I was 'authentically' improvised! You can tell just by hearing me . . . I came into existence in the moment that you heard me. It was only just decided that I should exist. My creator chose me over all the options within reach at that time." And so on ad absurdum. Listeners often want to decode the level of real-time agency in what they just heard, so they tend to ask questions like, "What percentage of that was improvised?" This would seem to suggest that, at least to the uninitiated, it's impossible to know just from listening. And if this is indeed the case, then how does anyone ever really know?”

Extreme blowing performances on a saxophone as done by John Coltrane, Albert Ayler, or Peter Brötzmann would more than likely be improvisational and not pre-composed, there can be sonic evidence for improvisations, even if no outright marker exists.

Furthermore, in stressing his point that improvisational sounds need not be distinct from pre-composed sounds, Iyer tends to throw some of the baby's bathwater out that could help distinguish between the babies. Iyer accepts that there may be some non-sonic evidence indicative of improvisational occurrences, including “there's nothing inherent in musical improvisation that "sounds" improvised. It's easier to identify things that "sound composed"': ensemble synchronies and unisons, very difficult to achieve spontaneously, typically signal to a listener that something was planned. In talking about improvisation, witnesses tend to invoke performers' body language, eye contact, visual cues, or of so-called "mistakes" to prove that these acts were (or were not) spontaneous. But short of such things (which are not essential attributes of improvisation) it seems 173

awfully difficult to pinpoint improvisation as such. There are reliable attributes of a specific improvised idiom—culturally specific principles of variation, parameters of expression, and choices to be made—but nothing in the sound of a musical action that announces, "Yes, I was 'authentically' improvised! You can tell just by hearing me . . . I came into existence in the moment that you heard me. It was only just decided that I should exist. My creator chose me over all the options within reach at that time." And so on ad absurdum. Listeners often want to decode the level of real-time agency in what they just heard, so they tend to ask questions like, "What percentage of that was improvised?" This would seem to suggest that, at least to the uninitiated, it's impossible to know just from listening. And if this is indeed the case, then how does anyone ever really know? There's certainly no point in asking what was "intended" in an improvisation, or what an improviser was "thinking" at that moment. And we still have that "impression" of improvisation discussed earlier, that improvised "quality" that is supposedly conjured by the best classical performers. (It is also common, conversely, to praise improvisations for the apparent presence of compositional elements: motific developments, a "natural" arc, and so on.) If a listener can knowingly allow himself to believe that a fully composed work has just been improvised into existence, or that an improvisation was instead composed, then of what use is the distinction at all? But it is a valuable and meaningful distinction. It brings us to a central paradox: the drama of improvised music involves the understanding that those sounds were chosen and deployed at that moment by those people. And yet, you cannot tell this to be true just by listening; you have to already know that this is happening. It follows that you only really know by referring to something beyond the sound.

Improvisation as a Condition. What I submit (after George Lewis and others) is that we agree to understand improvisation not as a quality but as a condition. . . . Is "improvisedness" audible? It doesn't matter. To paraphrase Peter Falk's character in "Wings of Desire": I can't hear it, but I know it's there.”

---

---

It is not impossible that someone could recognize improvisations while listening

Of course, most of the time, the musician listening to herself can know whether or not she is improvising, so the claim that no one can tell by listening is false.

➢ What about other band members? Can they recognize when their bandmate is improvising?

Most of the time, other band members should know whether their soloist is improvising since they have played the same song together. They recognize that what their soloist is playing is different from what that soloist ever played before with them. Furthermore, in the context, they also know that this is the part of the tune where their soloist is supposed to be improvising. They anticipate that an improvisation should be occurring now because they have already previously agreed that this would be the location in the tune where the soloist would improvise.

To test this, suppose that a soloist in private, unbeknownst to any of her bandmates, rehearses and then memorizes a new specific solo. When she performs this rehearsed and memorized solo, she will not be improvising. Could her bandmates tell whether her solo is or is not improvised?

One can argue that they might be able to tell that the soloist is not improvising. Here's how. The soloist they are used to hearing when she improvises sometimes is less than smooth. She makes some mistakes or plays some odd note choices that don't quite work. (See Ontimpr4. On mistakes and improvisation). When her bandmates hear her rehearsed solo, it is coherent and smooth with appropriate architecture that fits the form of the song they are playing. Their soloist plays her rehearsed piece flawlessly. Band members might then infer that because this flawlessly executed solo contained no hesitation that it probably had not been improvised by her.

---

---

It is not logically or metaphysically impossible to recognize when a passage of music has been improvised

➢ What do we require for something to be logically impossible?

For a proposition to be logically impossible, it must either be self-contradictory or lead to a contradiction. The contradictory nature makes any proposition logically or metaphysically impossible because it is not possible for the same situation in all respects simultaneously to be true and false.

There is a non-contradictory possible world where audience member Fred listens and recognizes that the soloist played an improvisation. Because there is no contradiction involved, this is a logically possible universe and proves anyone who denies the mere possibility of improvisation recognition is wrong.

Perhaps all the deniers mean is that it is unlikely that anyone can know with certainty that a soloist is improvising.

The degree of certainty required for recognition of an improvisation is relevant to assessing the truth of such a recognition claim. Under many circumstances, we say that you recognized a stop sign 🛑 when you braked your vehicle. Could you still be wrong it is a stop sign? Yes, you could be mistaken if the sign is fake, or an illusion, or a hologram, or a mirage, or a confused perceptual hallucination, or trickery by an evil genius. The absolute impossibility of being wrong does not foreclose a community from crediting Fred with recognizing a stop sign. The community standard is lower or weaker than the impossibility of being mistaken. What is the norm? It is high probability and context that communities typically require for successful recognition of a Stop sign and not that someone needs infallibility in their recognition achievements.

Similarly, then, also for jazz, this should be the standard. Communities hold that highly probably true is good enough for recognizing stop signs or jazz improvisations.

---

---

It is false that it is always implausible that one recognizes when a soloist is improvising from listening.

➢ What considerations show these claims of unlikelihood or extreme difficulty are false?

To refute that it is always unlikely to determine when a soloist improvises, all we need is ONE counter-example that denies the always claim.

There are numerous scenarios where it is believable that one could tell whether a soloist was improvising.

CX1: Unlimited multiple takes of the same solo, all different each time and each responsive to the rest of what other band members play. Since the soloist reacts differently in real-time to what others play, she must be improvising to account for these appropriate responses. If the solo were previously structured, bandmates would not find the soloist responsive to their differing contributions. When a band member does three different things on separate occasions at the same point in her solo while maintaining the same level of responsiveness to what other musicians play, she has to be improvising. She would be non-responsive to alternative input by bandmates if her solo on three different occasions were always unresponsive to bandmates altering musical inputs. She is not improvising but playing a pre-composed piece.

CX2: No likelihood that what one is hearing could even be written down or previously scored. Peter Brötzmann solos, with their squeaking and extemporaneous honking and bleating, have a low probability of pre-composition because they are so wild sounding. Why would anyone even bother to pre-compose music with so few structural internal relations?

CX3: Band members know when a soloist in their band is improvising because of many things. The soloist is expected to improvise, has always improvised in the past, says that they (it's a gender-neutral soloist) will be improvising during their solo.

CX4: Soloist makes appropriate musical quote commenting upon what is currently happening in the room. A made-up example based on an actual performance would be if, upon seeing a patron walk into the club wearing a bright red vest Charlie Parker during his improvised solo quotes the song "When the Red Red Robin (Comes Bob Bob Bobbing Along)." (Hear him play it here during his improvised solo on "52nd Street Theme," performing in Waukegan, WI in 1947.)

CX5: fMRI Scientific brain monitoring and different brain regions light up more when soloist improvises than when playing a prepared pre-composed solo.

CX6: Supported by testimony because a reliable soloist informs you she is going to improvise her solo. You know her to be an honest witness, and afterward, she confirms that she did improvise her solo. Is it probable that you can recognize that what she did during her solo was an improvisation? Well, doesn't it at least raise the probability that you can realize her solo was improvisatory?

Suppose it does raise the likelihood of recognition of improvisation. In that case, audiences in general at a jazz concert have expectations that any soloist will be improvising, thereby increasing the probability of recognizing an improvisation when heard.

How likely or unlikely one will be in recognizing when someone improvises is a relative matter with different contexts and scenarios making significant differences in the likelihood of identifying improvisations.

---

NOTES