Psychoanalytic roots:
Psychoanalyst Erik Erikson is generally credited with coining the term generativity in his 1950 text Childhood and Society, in which he defines generativity as “the concern in establishing and guiding the next generation” (1950: 267). Erikson’s context is a psychosocial one, but the notion of a productive creativity between metaphorical generations has developed a broader resonance.
Linguistics and grammar:
In 1957, Noam Chomsky published Syntactic Structures, in which he defined a concept of transformational generative grammar which used “rewrite rules” that could be used not only to describe a language but also to generate sentences in a language. Differentiating between deep and surface structure of a sentence, Chomsky wrote that deep structure is “that aspect of the [syntactic description] that determines its semantic interpretation, and the ‘surface structure of a sentence’ [is] that aspect […] that determines its phonetic form” (Chomsky, 1966: 16). Transformations, then, are applied to the deep structure to generate its surface structure, or individual enunciation. The earliest recorded generative grammar, incidentally, is reported to be Pāņini’s Așțādhyāyī, a Sanskrit grammar from the 4th century BC that outlines more than four thousand grammatical rules to generate words and word combinations.
From semiotics, Ferdinand de Saussure’s differentiation between langue (a system) and parole (an enunciation) may also offer a helpful framework for understanding generativity at a textual level (Cours de Linguistigue Generale, 1916).
Generative music:
At the 1996 San Francisco Imagination Conference (documented in inmotion magazine), Brian Eno outlined a concept of “generative music” in opposition to a broad category of recorded music he termed “classical music”, defining a rule-based, open, responsive system (as opposed to a closed, replicable artefact):
“Classical music […] specifies an entity in advance and then builds it. Generative music […] specifies a set of rules and then lets them make the thing […]. Generative music is unpredictable, classical music is predicted. Generative unrepeatable, classical repeatable. Generative music is unfinished, that’s to say, when you use generative you implicitly don’t know what the end of this is […]. Generative music is sensitive to circumstances, that is to say it will react differently depending on its initial condition, on where it’s happening and so on. Where classical music seeks to subdue them.”
Scape screenshot from iTunes store
More recently, Eno has defined his iPad app Scape (2012), developed with musician and programmer Peter Chilvers, as “a new form of album which offers users deep access to its musical elements. These can be endlessly recombined to behave intelligently: reacting to each other, changing mood together, making new sonic spaces […]. It employs some of the sounds, processes and compositional rules that we have been using for many years and applies them in fresh combinations, to create new music.” (http://www.generativemusic.com/). Generativity is about rules and protocols, then, rather than following notation; that is to say, the system, rather than the particular enunciation.
[Incidentally, a very interesting early [1983] text on the relationship between theories of generativity in linguistic and music is Lerdahl and Jackendoff’s A Generative Theory of Tonal Music (MIT, 1996)].
Tomotaro Kaneko’s Generative Music Workshops, dating from 2010, have followed a similar principle to Eno’s work with Chilvers, setting about reproducing “masterpieces of generative music […] by following the original instructions and images of past realizations instead of notation” (2011: 179). Kaneko’s conclusion is that “replayed sounds never progress in the same way because of the specific relationship between components and environment. Such generativity of the relationship would be stressed by mobile computing […]; future developers [of generative music apps] need to consider such expanded environments”.
This seems to raise important questions about the material world and our mediated experience of it through art, in the wake of the era of mechanical reproduction. If mobile computing is facilitating a new way of thinking about music that is more like our relationship with sound in the real world, then might this change our expectations of how we will engage with art? If we expect each enunciation to be different (generative) every time, created anew at the moment of our engagement with it, then will our engagement be alert, active and engaged? Has our 20th century habituation to identicality of experience, via the MP3 track, CD or vinyl record, promoted a passive automatism in perception? Eno writes: “Putting on a record and knowing I’m going to hear the same thing I did last time has actually become a little bit irksome. It feels quite Victorian.” I agree!
Of course, fan participation in terms of mash-ups of existing released tracks is not new: here’s one of my all-time favourites, by The MonstaMunch Music: https://www.youtube.com/watch?v=HeRLLwKxIJM. More recently, though, a significant further step towards generativity in music production is the emerging participatory fan culture of stem remixing, in which components (stems) of a multi-track recording are made available to fans by (usually) record companies – that is, the owners of the music – allowing fans to remix individual components of the track. 
Examples include Deadmau5’s Sofi needs a ladder (2010), REM It Happened Today (2011), and Bon Iver’s album Bon Iver, Bon Iver (2012). For more on this, see Sam Bennett of Australia National University, who spoke on this at Cambridge CRASSH‘s March 2014 conference on Creativity, Circulation and Copyright: Sonic and Visual Media in a Digital Age.
Eno seems to set out a similar process in the authoring of the system behind Scape, saying, in Meet the Developers: “we began by deconstructing some pieces of music that I already had in multi-track form […] all their separate sounds separated, and seeing what would happen if we took those apart and built a system where they could reconfigure in semi-random or indeterminate ways, and that’s how Scape started.”
As media theorist Henry Jenkins points out in Convergence Culture (updated edition, 2008), such a fan culture is not new, but what has shifted is [its] visibility” (135). Further, I would propose that a second important shift is the emerging normalisation of this more (inter)active, collaborative, creative mode of engagement, and its apparent – although undoubtedly partial and strategic – sanctioning by the original authors/owners (in this case, typically the record companies).
This raises the important question of generativity in opposition to sterility: in digital terms, a generative technology is one that can be built upon by anyone (an open system) rather than one that is controlled and can only be changed by the original author or owner (a closed system).
Finally, a more general word on Digital: In 2008, in The Future of the Internet: And how to stop it, Jonathan Zittrain defined generativity as “a system’s capacity to produce unanticipated change through unfiltered contributions from broad and varied audiences” (70). Like Eno’s definition, then, and whether through human or environmental intervention (as Kaneko), this suggests a system that responds to external stimuli to constantly evolve its expression or enunciation. Can a text like the Commedia function, or simulate functioning, in this way too? This is one of the questions I’m asking in my research.
N.B. Additional section coming soon on gaming and procedural generation.
generating dante 