From the Prescriptive Model to the Informative Model: Formula and Forms of Creativity in Jazz Improvisation ROBERTO ZANETTI Università degli Studi di Torino Dipartimento di Filosofia e Scienze dell’Educazione Torino Italia Email: [email protected] Web: https://unito.academia.edu/RobertoZanetti Keywords: formula, improvisation, jazz music, aesthetics, creativity. I. INTRODUCTION not be true in the absence of a notational system. Indeed, it may not be true for most of the world’s mu- This article deals with jazz improvisation, attempting to sic. But for a great deal of the most valued music of the 6 finding a way to define a model through which it could be West, since the development of a sophisticated musi- COSMOS + TAXIS COSMOS conceived, developed and practiced. To begin with, we will cal notation, it seems to be true that there are musical investigate the traditional schema through which musical works, and that there are performances of them (Kivy ontology has always considered musical entities, i.e. a pre- 1987, p. 245). scriptive one. Later on, we will focus on a dynamic concep- tion of formula—i.e., patterns that are able to express the Kania (2014) offers us an important springboard to reflect musician’s individuality in a given performative context— on the role of the so called “high-order music ontology” and sketch the difference between figural and generative over (approximately) the last five decades. The core ques- models. Particular emphasis will be given to the distinc- tions of this philosophical area are commonly sketched as tion between formula and mere licks, that can be defined follows: as conventional, fixed and ossified formulas. In conclusion, we will see how jazz music can be based on a different, not • what kind of things are musical works? pre-determined kind of structure, which we may call in- • what criteria do we need to identify and recognize them? formative model. This allows us to abandon an inveterate • what is their mode of existence? prejudice, i.e. that jazz is an anarchic music, and, in addi- tion, to better understand what criteria should we refer to As Kivy claims, it seems that the presence of an object in evaluating the creativity which is at stake into a jazz per- called “work of art” and instances of it called “performanc- formance. I do not want to establish a qualitative hierarchy es” is taken for granted, at least in Western music. In other between the prescriptive and the informative model, but I words, if we do not suppose that these kind of things may would only like to highlight the fact that they give rise to have some sort of existence (concrete, abstract or fictional), different productive paths in making music. investigating issues of music ontology could be problem- atic, if not useless. Leaving aside the undeniable accura- II. MUSICAL ENTITIES WITH PRESCRIPTIVE cy and depth of Kivy’s inquiries in this field, I think that MODELS: THICKNESS AND THINNESS these lines reveal something that few ontologists of music (in which Kivy is certainly included) are inclined to admit: It may not be true for improvisations, and it may not music ontology, as it has been conceived since its birth, can be true for certain kinds of electronic music. It may tell us very little (i.e., “it may not be true” for too much) of VOLUME 5 | ISSUE 1 2017 COSMOS + TAXIS the musical experiences that technological, social and in- with the notion of “norm-kind”, i.e. an entity that admits tercultural practices enable (Born 2005). Even so called both correct and incorrect performances. This allows “comparative ontology” (Kania 2014) which tries to apply Wolterstorff to maintain a certain degree of flexibility in the categories that traditional research has developed in the sketching the relation between a musical artwork (a kind) analysis of Western classical music to other genres (such as and its performances (the tokens) (Wolterstorff 1975, p. rock, jazz, folk and pop), faces the same difficulties.1 As a 131). Genuine performances do not simply instantiate the consequence, many scholars have raised doubts about the kind through a proper sound sequence, but are produced usefulness of the notion of a work of art (Kania 2011; Brown with an idea of what should be done in order to have a cor- 1996, 2000a)2 and the philosophical tool usually linked to rect rendition of it. Wolterstorff here is claiming that de- that, the type/token model.3 Others have been even more spite the fact that one makes every effort to obtain a correct radical and have called into question the whole concept performance for a particular piece, it can well happen that of music ontology (Thomasson 2006; Ridley 2003, 2004). one does not succeed. Nevertheless, this does not prevent Maybe it would be excessive and to some extent harmful to us from identifying our rendition as one of that musical throw away decades of philosophical enquiry on music, but artwork, i.e., of that norm-kind. Even if Wolterstorff does it is certainly true that a traditional point of view on today’s not consider correctness as a sufficient condition to produce musical practice does not help us enough, and that the role genuine performance of a musical artwork,6 he maintains of concepts such as work/performance, composition/rendi- that it is at least necessary. This is particularly clear in all tion and composer/interpreter should be sharply reduced. those scholars (in which Wolterstorff could be undoubtedly This is particularly relevant in musical contexts involving included) who see the relation between a musical piece and the complex and the much debated notion of improvisation its renditions in terms of shared properties (Dodd 2007, p. in which most of aforementioned high-order music ontolo- 201). Julian Dodd extends this reasoning to its most extreme 7 gy assumptions seem to be lacking.4 However, before focus- consequence: according to him, the creative process carried ing on improvisation, it could be useful to make explicit the out by an artist simply consists in selecting the properties model that underlies nearly all the ontological perspectives that a performance must possess in order to count as a cor- that consider the ordinary musical event as consisting in rect rendition of the composed piece. Furthermore, the only a more or less faithful rendition of a pre-existent entity or properties that can fulfil this task are thestructural-sonic COSMOS + TAXIS + TAXIS COSMOS structure (Hagberg 2002, p. 189), whether or not we call it a ones played with the timbre that the composer has previ- work of art or not. ously establish: that is why Dodd’s view can be understood The model is based on a prescriptive conception of the as timbral sonicism.7 It should be clear that sonicist theories, musical phenomenon, according to which a performance is both pure and timbral, end up by reducing the creative pro- commonly intended as a rendition of a given pattern (Dodd cess to a matter of selection, and this “may not be true” for 2007; Kivy 1983, 2002). The way in which this pattern is in- the great majority of the musical practices we come across stantiated allows us to decide if the performance is correct in our artistic experience. or not. In other words, performance is a medium through Not only in jazz—which is commonly viewed as the “nat- which we can have access to a certain musical piece that is, ural realm” for musical improvisation—but also in some in principle, detached from the specific rendition we hear contemporary classical composers such as Stockhausen and in a particular occasion. From this perspective, the ontol- Cage, improvisation plays a central role in creating a mu- ogy of music has developed the type/token model in order sical piece.8 In all these contexts, selecting properties does to preserve the identity of a musical structure despite its re- not seem to be the most important feature, because im- peatability (Bertinetto 2012b) and, secondly, to justify the provisation is a creative activity that does not give rise to expressive freedom left to interpreters. Roman Ingarden, for a re-performable entity. Stephen Davies, in his well-known example, intends expressive features as the result of a “fill-in book Musical Works and Performances, focuses on the dif- the gaps” activity: for him, musical scores are constitutively ference between composition and improvisation: partial, because it is impossible to note down every sound inflection on a pentagram (Ingarden 1989). Consequently, Improvisation are not musical works, I say. Does the main task of the interpreter is to complete the spaces this mean that improvisers are not composers? Here of indetermination that musical scores open up.5 Nicholas I think one can say what one likes, so long as the is- Wolterstorff (1975), in turn, identifies musical artworks sues are clear. It is plain that composition and impro- FROM THE Prescriptive MODEL to THE Informative MODEL: Formula AND FORMS OF Creativity IN JAZZ Improvisation visation involve similar processes—inventing tunes, performance framework, “binding” seems more appropri- organizing material, trying to unify the manifold. If ate if related to the first, rather than to the second. To make “composer” is like “swimmer” or “driver”—if it names a very partial proposal, it could be said that a performance the person who creates a certain kind of activity—im- exhibits, displays or shows some properties, and that these provisers are composers. On the other hand, if “com- properties are liable to be indentified and evaluated accord- poser” is more like “fletcher” or “wheelwright”—if it ing to different criteria, depending on the musical context names the person who creates a certain kind of prod- in which this performance takes place (Davies 2009).