parapluie elektronische zeitschrift für kulturen · künste · literaturen -> übersicht | archiv | suche
no. 17: improvisation -> jazz

Habitual Spontaneity

The Improvisational Paradigm in Jazz and Comedy

by C.B. Davis

How do we develop the habitual spontaneity that is one of the central features of improvisation, and which we constantly take for granted in everyday conversation? It is quite common to enlist linguistic models in the attempt to answer this question, but such models cannot account for the largely unconscious processes that inform improvised conduct. A look at the practice of improvisation in both jazz and comedy from the performer’s point of view, aided by the embodied theory of meaning of Cognitive Linguistics can provide insights into the aspects of improvisation that the linguistic model leaves unexplained.

Improvisation, especially in music, endures both mystification and dismissal from non-practitioners. My experience as a non-sight-reading piano player bears this out. On the one hand, there are those who ask me, usually while I am playing, “ Are you just making that up as you go along?” In a certain tone of voice that choice of words seems to accuse me of cheating, as if playing from a “fake book” was like writing the answers to a test on the bill of your cap. On the other hand, there are the people who wait until I stop playing to ask, “Are you playing by ear?” which is a polite way of asking, “Is that natural?” It seems obvious to me that the way I play piano is a result of the way I learned how to play, but people imply with such questions that improvisation is a either a dubious skill or a freakish talent. I am one of those musicians who are more comfortable improvising than reading music (accurately) or playing a memorized piece (as written). When I’m not self-conscious about playing I can make coherent musical statements on the piano much the way I would tell a story or describe something in conversation. The language I am “speaking” when I improvise is a dialect of jazz. When I am thinking too much, interrupting the self-organizing dynamic of the improvisation, the jazz comes out as if I am translating in my head rather than speaking fluently. It is not as coherent, and it does not sound intentional until I get back into the flow by incorporating something from the “mistake” into the next phrase.

Framing improvisation as art creates a space in which self-consciousness is a normal reaction, as anyone who has performed in front of an audience knows. Most people can get around this self-consciousness in performance by concentrating on the memorized text or score they have rehearsed. Overcoming the self-consciousness of performing ad-lib can be the most difficult part of learning to improvise for many people. When first asked to improvise, a former jazz student of mine “ran into a wall”, even though she possessed excellent technique and could play the scales and arpeggios needed to get through the chord changes. I had the image of her being frozen at the edge of a diving board, or afraid to grab onto a swinging trapeze. She was used to having a net, that is, a score or memorized piece to play. The wall she hit was her self-consciousness, her inner monologue telling her she didn’t know what to play. Finally I told her to hit one note in whatever rhythms she wanted while I played the chords or bass line. After these first few awkward baby steps she began to realize that she did have a net: she could play something simple. Her technique was making her too self-conscious to play; she expected to be able to play jazz at the level she could play Chopin. I think she realized after a while that she wasn’t thinking about what to play when she played Chopin, but was rather relying on her muscle memory, prompted by the score, or an occasional flashing acknowledgement of having finished one idea and moving to the next. Improvising, she understood, was more like talking than reciting. Bebop is too complex a territory to navigate by thinking about it. Whether the map is in the “fake book” or in your brain, you still have to rely on muscle memory to walk, run and jump through the terrain. We humans are most regularly improvisational in everyday actions and conversation, when we are not self-conscious about performance. When I am improvising fluently I am not thinking about the stylistic and historical parameters of jazz music, just as I am not thinking about etymology or grammar when speaking my native language, and yet the history, the rules and idioms are still there in both cases, informing my movements and the sounds that result. The realization that we all improvise when speaking in everyday contexts was useful in helping my piano student overcome her “block,” but it did not solve the problem of how to develop the habit of spontaneity in performance. Even the mastery of the grammar and vocabulary of a second language does not assure access to spontaneity. Knowing the rules is a necessary but not sufficient condition for acquiring the skills that lead to fluency.

When I first began writing this essay, I thought that my task would be to describe those unique elements that distinguish jazz playing from other improvised conduct. However, I immediately found it necessary and more fruitful to first identify the elements that were common to improvised performance in general. In order to throw the paradigmatic features into sharper relief, I will compare jazz improvisation to a “non-musical” improvisational tradition, that of improvised comedy, specifically the Commedia dell’ Arte. Just as all improvised music is not jazz, all improvised comedy is of course not Commedia, but the Commedia provides a more definable set of parameters for improvisation than more current forms of improvised comedy and is thus particularly useful for the present purpose. In addition to a mastery of underlying structures and idioms, the comparison of jazz improvisation and Commedia illuminates more specific traits. From the performer’s point of view, these paradigmatic elements include: frequent reference to a commonly understood vocabulary of signs, the ability to create while paying attention to audience and fellow players, a social space conducive to experimentation or playfulness, a commitment to taking chances, and perhaps most important of all, a cessation of inner monologue which allows for a quick flow of spontaneous responses and ideas. I will try to draw out these features as analogues in both improvisational genres.

Commedia is analogous to jazz in that it was based on traditional materials. Just as the template for jazz was the American popular song, Commedia was based on typical comic plots and stock characters that recur in Italian comedy at least as far back as Plautus (2nd century BCE). Unlike jazz, which has an arguably unbroken tradition, Commedia supposedly died out in the seventeenth or early eighteenth century. It is also true that the Commedia never really existed in the form in which it is practiced today, which at its worst becomes an overly precious “renaissance fair” anachronism. The postwar Italian playwright and comedian Dario Fo admits that what he learned and taught was not the Commedia of old, but methods based on a comic tradition of performance that has been continuous, even when the masks and scenarios went away, if they ever really did, structurally or semiotically speaking. In his book of lectures, Tricks of the Trade, Fo somewhat sidesteps the controversy over how much actual improvisation was involved in the Commedia by defining improvisation as more a quality of execution than a set of procedures. The act of improvisation in both jazz and comedy is unconvincing without this palpable feeling of spontaneity. The habitual spontaneity of the improviser demands confidence, relaxation, and adaptability; all of which amounts to staying in the present moment rather than anticipating your own and everyone else’s next move,or dwelling on the last.

Confidence and adaptability in improvised contexts is built on the mastery of a very large vocabulary of syntactic structures and idiomatic gestures until they become almost an unconscious template, the langue to the parole of improvising, to put it in Saussure’s terms. This langue is what allows jazz musicians to improvise together without rehearsals by “calling” tunes that are familiar to most players, and I imagine that adepts of Commedia could similarly step in and wear the mask of another character or improvise on a familiar scenario with unfamiliar fellow actors. In Commedia, this langue includes the masks (stock characters), roles (a hierarchy and set of relationships), scenarios (the bare plot with entrances and exits), and bits of business (lazzi). A typical lazzi of the clown mask Arlechinno is “the lazzi of the fly” in which the servant clown mimes following the path of an irritating fly and tries to catch and eat it, usually pulling off the wings first. This is the sort of bit that could be inserted almost anywhere in a scenario as a solo or a comic upstaging of the other performers. The clown’s lazzi correspond to the riffs or musical phrases that jazz musicians frequently fall back on, sometimes as a substitute for inspiration or as a way of cueing transitions to the players and audience. These “licks” and “lazzi” also serve as idiosyncratic gestures identified with individual characters or performers, as in the case of Thelonious Monk’s signature whole-tone runs or Harpo Marx’s penchant for cutting off a gentleman’s tie or coat-tails with his always handy scissors.

Masks and roles are harder to transpose into musical equivalents, but they have parallels in the hierarchy of functions and characteristic roles of each instrument. If Pantalone and Brighella (or any other pair of comic types from August and White Clown to Groucho and Chico Marx) engage in a scene together, one of them is usually feeding the other either straight lines or manipulative tricks to which the more immediately comic figure can react. In jazz, an ensemble will sometimes drop down to the bassist and the soloist, with the bassist performing the “straight man” role, feeding rhythmic and harmonic movement to the soloist. Here we come to one of the major differences between musical and theatrical exchanges, the fact that actors usually don’t speak at the same time. There is however, a lot of “turn-taking” in terms of focus and background roles in jazz, and the functional hierarchy still works as a connecting and structuring principle. For example: the ensemble plays the arranged melody, the tenor saxophonist or trumpet takes an improvised solo, accompanied by drums and bass on the first chorus, with piano or guitar joining on the bridge or second chorus. This is very much a scenario of entrances, exits and scenes between the players. Whoever is leading the scene or soloing relies on the support of the other players to keep the momentum going and to reinforce the improvised musical ideas or dialogue. This reinforcing of forward momentum is as important in comedy (and in most dramatic genres) as it is in music and dance. Comedians “get on a roll,” and jazz musicians “swing” or get into a “groove.” On a more micro-level, the development of action and dialogue in improvised comedy is analogous to the use in jazz of sequential musical phrases. Each succeeding phrase makes use of the material that has come before in the improvisation, and the rhythm section’s role is both to push the momentum forward by reacting to the sequences rhythmically and harmonically. A common exercise in learning to improvise dialogue is referred to as “yes-and-ing.” To keep an improvised scene moving it is best not to introduce new subjects, but to practice agreeing and extending the situation or idea indicated by the first person who speaks or does something in a scene. In both jazz and improvised comedy, the push forward requires not only rhythmic but also conceptual momentum.

The linguistic model I have been using in discussing the syntax and semantics of improvisation has dominated explanations and discussions of developmental and pedagogical approaches to improvised conduct. How does a player get from imitation and practicing the fundamental langue to spontaneous creative play? Is improvisation more than just repetition and shuffling of learned strips of behavior sewn together with an attitude or feeling of spontaneity? The linguistic model can describe improvised conduct as well as everyday conversation to a certain extent, but these descriptions are limited because of the fact that in both improvisation and spontaneous speech most of the work happens on an unconscious level. Linguistic metaphors for unconscious processes, including those involved in talking, are limited in their analogical relevance by the categories of semantics, syntax, grammar, etc. The relatively new discipline of Cognitive Linguistics attempts to overcome these limitations by adopting an “embodied” rather than formal theory of meaning. This theory, most canonically stated by Lakoff and Johnson in Philosophy in the Flesh, has yet to be applied to the unconscious processes underlying the paradigmatic features of improvisation. I will attempt to outline possible directions for the application of an embodied theory of cognition, after a brief review of the most salient explanations of improvisation to date.

Until quite recently, the two most insightful and well-researched studies on how people learn to improvise in jazz were Paul Berliner’s socio-cultural study Thinking in Jazz, and David Sudnow’s recently re-written Ways of the Hand, a first-person phenomenological account of learning to play jazz piano. Both Berliner and Sudnow use predominately (but not exclusively) linguistic models and metaphors to describe the process of improvising. Berliner and his informants suggest that a good solo should have a sort of narrative structure, while Sudnow’s emphasis is on the speaking of coherent “jazz sentences.” Both of these books suffer from valiant attempts at explaining the inside of jazz from the outside, that is, without actually using technical musical explanations. Berliner’s book is more readable because he focuses on the social and cultural institutions and practices through which jazz musicians have traditionally passed on their skills and lore. Sudnow’s job is more difficult, and so is his book, a situation that prompted the recent re-write. The book is still hard to read, and even where it is clearer than the original, Sudnow still has to rely on a makeshift vocabulary that focuses more generally on the nature of acquiring a manual skill than the music itself, and the result remains more obtuse than the most complex music theory.

Although not exclusively about either jazz or improvisation, William Benzon’s book Beethoven’s Anvil offers a fresh approach to both subjects, combining a cognitive approach to music with Richard Dawkin’s theory of cultural evolution. Benzon’s theme is basically that music, originally encompassing dance, developed as an effective means of social bonding. In fact, most of the aspects of music in general for which he offers explanations are applicable to improvisation, although I think improvisation more often requires and therefore engenders the kind of ecstatic waking dream states he attributes to high level “musicking” of musicians in all genres. Like a good jazz musician (which he apparently is) Benzon is careful to stop when he runs out of ideas on a theme, rather than trying to push for a theory of everything, as many projects in cognitive science seem to do. He readily admits to the gaping holes in empirical evidence that he must fill with speculation. The boldest of his propositions is that music is an immediate precursor to the development of language and hence, all culture. Biological evolution, in Benzon’s view, gave us the kind of neural circuits that allow for the cultural evolution of music, which in turn molds the nervous systems themselves. For example, music allows for a kind of neural integration among groups of people, and this is turn created new neural connections that allowed for the development of language. Language gives birth to the inner monologue, which for Benzon is roughly equivalent to the human will. I don’t quite follow Benzon’s logic in making an inner monologue necessary for true intentionality, but I do agree that it is this inner monologue that moves into the background of consciousness during an ecstatic musical experience, especially during fluent improvisation, both in jazz and in improvised comedy.

In jazz, the unconscious directing force of muscle memory responds to structural prompts and input from the other players, but I have also heard the term “improv muscle” used by experienced comic improvisers to describe the spontaneous comic reflex they utilize when “on a roll.” This “improv reflex” was first described to me by actress Shelly Stover, a former theatre student of mine who is one of the most gifted comic improvisers I have ever seen. She has described to me the experience of exiting from a successfully improvised scene, hearing the audience’s laughter and knowing that what she had just said and done was very funny, but not quite remembering what had happened. I have had this same experience in listening to playbacks of my piano improvisations and being happily surprised at what I hear. The “improvisation muscle” serves as an apt metaphor for the embodied cognition involved in spontaneous movement, and actual muscle memory is one of its most important inputs.

To supplement his own use of linguistic models, Benzon elaborates on dance as an analogue to high level “musicking.” As he admits, there is nothing new about explaining music through the phenomenon of dance, but Benzon specifically says that the underlying pulse of music is mediated by the same neural structures that mediate locomotion. Even when people are only listening, rather than playing or dancing, the “central motor systems that evolved to drive the skeletal muscles are entrained to the music” (142). In a paper entitled “What is Music?” Danish cognitive semiotician Per Aage Brandt proposes a kind of sensorimotor homunculus that he calls a “private dancer.” He posits that certain sounds, intervals and rhythms affect this inner dancer and cause the tension and release we feel when listening to music. Since emotions have correlates in physical movement and positions (recoiling from danger, flailing in anger, etc.) this suggests that the emotional qualities of music are grounded in locomotive neural schemata.

Benzon makes a rather vague use of Lakoff and Johnson’s theory of embodied cognition to propose that the meaning in music is made through metaphorical projection of an experiential sensorimotor source domain projected onto a more abstract target domain. The most recent work on conceptual blending by Turner and Fauconnier, The Way We Think (2001) convincingly demonstrates that this unconscious semantic projection is actually much more complex than binary metaphors. Turner and Fauconnier show that the source-to-target domain mapping is only one aspect of conceptual blending which involves multiple inputs and partially overlapping spaces. These cutting edge cognitive theories are being applied to studies of meaning in literature and are just beginning to appear in work on meaning generation in musical listening. The current applications of cognitive science to the humanities seem only to explicate the reasonable and rather obvious assumption that we do and think as we do because of the kind of bodies we have. Explaining the complexity of a musician’s improvisational behavior or a knowledgeable listeners’ reception of jazz improvisation benefits more from knowledge of musical history than from either philosophical speculation or scientific measurements. Benzon does say that the neural circuits that connect music to locomotion have to be culturally trained. Like Benzon, I am attracted to Richard Dawkin’s theory of cultural memes, the cultural equivalent of genes, because it emphasizes how the traditions themselves evolve through the uses to which we put them. Work that combines semiotic questions with cognitive explanations seems the most fruitful in liberating theories of meaning in music from purely linguistic models, despite a mostly Anglo-American view that the cognitive turn is incommensurable with semiotics.

Benzon does not fully differentiate improvisation from other kinds of musical performance because he is focusing on the cognitive constituents of musical experience in general. To delineate the paradigmatic features of improvised conduct we need to isolate the process from the generic features of the relevant medium or art form, and that is why I chose a non-musical brand of improvisation to compare and contrast to jazz. As a multi-cultural meme, jazz is more difficult to define than improvisation. During the peak of the New Age/Wyndham Hill fad of the 1980s, and since the more recent “jam band” phenomenon (Phish, Widespread Panic, etc.) I have encountered musicians and non-musicians alike occasionally using the word “jazz” as if it were a synonym for improvised music in general. This is my cue to protest that improvisation is a necessary but not sufficient condition for jazz. Improvisational music was nothing new when jazz developed, and in previous centuries the ability to improvise was expected of most musical entertainers. As a rule, classical musicians before the twentieth century were much more commonly competent as improvisers, and of course many of the major classical composers and performers were celebrated improvising soloists. But, as I always add, “they weren’t playing jazz.” At this point in the conversation, the younger musician usually shrugs and asks “Why not?” I am afraid my almost automatic answer is something that older musicians used to say to me about fusion-jazz, i.e. “It doesn’t swing.” I still cringe a bit when I hear myself say this, remembering how I would have considered it a reactionary position in my youth, but today, all the music I can in good conscience call Jazz either alludes to a straight-ahead groove or to some aspect of the blues vocabulary, usually both.

Many genres of American music share elements of this same idiomatic vocabulary, even if we subsume the blues within jazz or vice versa. Rock and roll and other post-jazz popular genres also exhibit bent notes, minor and major thirds sounding simultaneously, phrasing over the bar lines, triple over duple meters and song form structures such as the 12, 16 or 32 bar chorus. This recognition of common structural and idiomatic elements would seem to put the emphasis back on improvisation as the essential element of jazz, if only in terms of the degree or amount of spontaneous playing involved. The paradoxical pairing of a drive for innovation and an allegiance to tradition also characterizes jazz, so that any definition begs the question of how much innovation a genre or tradition can take before it becomes something else. An improvising pianist I greatly admire, Al Hood, once said to me, “There isn’t much improvisation in jazz.” His curmudgeonly attitude aside, his comment made me cognizant of a question worth asking: “How much improvisation can there be in a traditional or codified style?” My personal idiomatic requirements might be too restrictive to an adventurous player like Al, who probably hears what I consider the “signs of jazzness” as blues and bebop clichés. As a result of the watering down of the art into commercial “smooth jazz” or even neo-bop and neo-swing, many musicians like Al are uncomfortable with the word “jazz.” Even the Art Ensemble of Chicago, the greatest stage show in avant-garde jazz after Sun Ra, refer to their repertoire as “Great Black Music: Ancient to the Future,” emphasizing a continuity of African influence even over their intense commitment to improvisation. For me, a certain level of commitment to improvisation is definitive of jazz, a commitment that can be just as strong regardless of whether the musical situation is competitive or collective, in earnest fury or devotional serenity.

In addition to commitment, improvisation in both comedy and jazz requires a social space that is in varying degrees conducive and accepting of playfulness. “Theatre sports” is in fact a frequently used term for improvised comedy performances in America, and some actors consider improvisation more of a game than a legitimate genre of performance. Although a sport or game is also improvisation on a set of rules, theatre and jazz are not literally contests, at least not simply two-sided ones. Yet competition is a causal force in both domains, as well as a pedagogical method. An apprentice player often rises to a higher level of commitment and desire when bested by someone in improvisational play. This is something that is somewhat lost in an academic environment, especially in these days of consumer-style education. Learning jazz and even improvisational comedy in school greatly limits the appropriate opportunities for “cutting” another player. Gender specificity aside, the violent undertones and macho flavor of jazz competitiveness or slapstick comedy are not paradigmatic of improvisation. However, a social space that favors play is often also conducive to at least symbolic competition. A certain competitiveness seems paradigmatic of improvisation in the sense that it is an ingredient in the playful, risk-taking atmosphere necessary for spontaneous creativity. In both comedy and jazz, negotiating the levels of freedom and constraint in improvisational situations is in great part a matter of paying attention to the other players, whether it is a monologue/solo or an ensemble section. Clowns not only compete as adversaries in the narratives they enact, but especially in improvisation when comic one-upmanship is a driving motivational force for creativity.

A competitive attitude can hinder an improviser if it becomes self-conscious rather than directed toward the other performers. As with audiences for sport, Jazz and comedy audiences often take sides in the enacted conflicts, and audience reaction becomes a spur and a gauge for the competitive edge in performance. Perhaps more attention is automatically directed toward the audience in Commedia than in jazz, due to the expectation of laughter on the part of both actors and spectators. Audience reactions to jazz performances differ greatly, from polite applause at the ends of solos to shouts, standing ovations and awestruck silence, but in comedy the desired response of the audience is specifically laughter. Other musicians and educated jazz listeners tend to laugh out loud at dazzling displays of creativity and bursts of virtuosi technique. The humorous aspect of jazz shares many features with clowning, which is at its best also the domain of the physical virtuoso. It is difficult to imagine that music and dance evolved before laughter, which is also an almost completely social activity, as is the wail and cry of mourners. Extending Benson’s developmental logic, perhaps group laughter and crying preceded music. I think that it is the “laugh” and the “cry” in jazz that I miss when players do not engage the blues vocabulary. But neither these direct vocal correlates of inner states, nor a variety of synchronized rhythmic movements are universally paradigmatic of improvisation. The most paradigmatic aspect of conscious attention that covers both jazz and Commedia is the consideration of one’s fellow players and the audience in guiding your own improvised conduct.

Everyone improvises most of the time, but it is not usually framed as art. Everyday conversation is improvised on vast sets of mostly unconscious social and linguistic guidelines. The less constraint these structures impose, the more “free” the improvisation and the more unsure you are of when to speak and what to say. Trying to hold on to a thought in order to wait your turn to speak inevitably makes you self-conscious and distracts you from the conversation at hand. This breaks the flow as much as interrupting or “trying to get a word in edgewise.” Groups can both listen and speak more sensitively when they have become socially and emotionally bonded. One reason for this is that roles become more firmly established in a bonded group, whether it is a family conversing at the dinner table, musicians playing on the bandstand or a comedy team improvising according to their typed roles.

The bonding of a group of improvising performers not only establishes roles in a more egalitarian way but also leads to more sympathetic listening. When I remember performances by the Art Ensemble of Chicago, I often see the image of the late Lester Bowie standing with his trumpet almost to his lips, listening so intently to the rest of the band that he pulled the audience into a deeper kind of listening with his own silence. A silent character onstage tends to take focus when everyone else is talking, and if she finally speaks, the anticipation and surprise of the audience highlights the moment. Perhaps the most celebrated use of this technique in theatre history is in the Agamemnon of Aeschylus, where Cassandra appears to be a mute character until she is left alone with the chorus, at which point her jagged prophetic ravings are all the more powerful. A more apt example for comedic performance would be the decision of Harpo Marx to remain silent in the wake of his two hyper-verbal brothers, effectively framing his signature use of props and sight gags and giving him a special kind of focus whenever he appeared.

In effective improvisation, silence and listening are as paradigmatic as any of the busier, more aggressive elements. Responsive, sensitive listening may in fact be the most important skill an improviser develops, as it is perhaps the most certain path to cessation of the inner monologue and the habit of spontaneous creativity. Both jazz musicians and improvising comedians attest to an outwardly directed attention when achieving that all too rare ecstatic state of being so deeply in the moment that intention takes a back seat to instinct. Perhaps though, from the improviser’s point of view, the most relevant similarity between jazz musicians and comedians is simply that they are always eager and willing to improvise. You definitely have to pay us to stick to the script or read music all night.

Works Cited and Suggested Reading

  • Aksnes, Halljerd, “Meaning Generation in Music Listening.” E-text at
  • Benzon, William. Beethoven’s Anvil : Music in Mind and Culture. New York : Basic Books, 2001.
  • Berliner, Paul. Thinking in Jazz :The Infinite Art Of Improvisation, Chicago : University of Chicago Press, 1994.
  • Brandt, Per Aage. “What is Music?” For the 13th "Nordiske Musikforskerkongres," Århus 2000. E-text at
  • Dawkins, Richard, The Selfish Gene, Oxford ; New York : Oxford University Press, 1989.
  • Fo, Dario. The Tricks of the Trade, trans. Joe Farrell, edited with notes by Stuart Hood. New York: Routledge, 1991.
  • Lakoff, George and Mark Johnson, Philosophy In The Flesh: The Embodied Mind And Its Challenge To Western Thought, New York : Basic Books,1999.
  • Sudnow, David, Ways Of The Hand, A Rewritten Account, Cambridge, Mass.: MIT Press, 2001.
  • Turner, Mark and Gilles Fauconnier, The Way We Think : Conceptual Blending and the Mind's Hidden Complexities. New York : Basic Books, 2002

copyright © 1997-2012 parapluie & die autorinnen und autoren. alle rechte vorbehalten.
issn 1439-1163, impressum. url: