***

It is not only pornography and cinema that encroach upon our agency. Large scale data collection and analysis, of the type practiced by NSA, google and GCHQ, involve the collection of vast swathes of information on the population with the aim of modelling, predicting and, ultimately, influencing human behaviour.

Whilst much of the discourse around the revelations of the extensive online monitoring activities of the National Security Agency has gravitated around the invasion of personal privacy that might come from some bloated hacker in a light-sapped NSA back-office sudorifically soaking the back of his ill-fitting Tshirt into a Rorscharch blot of sweat as he bangs one out over your emailed fuckphotos in the seminiferous and hypaethral comfort of a government-sanctioned toilet cubicle, the real danger has nothing to do with how accessible the pixellated and over-exposed smudgestain of your disgusting genitals in flagrante delicto is, or that you enjoy visiting websites in which people insert cockroaches into their urethras. Instead of the horrorshow of your private perversions, these institutions instead seek the very information with which you move those around you to the very brink of somniferous apathy– what you buy, what you read, what ill-conceived mewlings you wretch up with bulimic regularity onto your social networking profile. In many ways the NSA are more interested in your life than your friends are.

Techniques for collecting and analyzing data and using this for predicting human behaviour have greatly improved, which means that it has become easier to manipulate our behaviour. In the corporate realm, this is done primarily with the aim of selling us things more effectively as well as changing our behaviour to buy things that we would not otherwise. Surveillance and digital media expert Mark Andrejevic claims that one of the aims is “in a sense to pre-empt consumer desire”, “Consider the oft-repeated mantra of database marketers: 'Advanced analytics helps retailers know what their customers want to buy before the customers do'” (Andrejevic, 2013, p. 28). He gives the example of how the retail giant Target looked for correlations between purchasing particular products in particular quantities and recently becoming pregnant.

...Target searched through its giant consumer database to determine what patterns of purchasing correlate with the eventual appearance of female shoppers on its list of customers with babies ... Then Target's researchers determined what patterns of advertising were the most effective, conducting controlled experiments by creating different combinations of advertising appeals. The potentially intrusive character of this type of research was indicated by an anecdote reported to the New York Times about a man who complained to Target after his teenage daughter started receiving advertisements for baby products: it turned out that the store knew before he did that his daughter was pregnant. (Andrejevic, 2013, p. 25)

This type of analysis springs from what Chris Anderson refers to as “The End Of Theory” (Anderson, 2008) in which the increase in computing power and the large amount of data available¹⁴ means that correlations between disparate pieces of data can be found and used to build ideas about the world, but the correlations that they find are not based on a pre-existing hypothesis: “Petabytes allow us to say: “Correlation is enough.” We can stop looking for models. We can analyze the data without hypotheses about what it might show. We can throw numbers at the biggest computing clusters the world has ever seen and let statistical algorithms find patterns where science cannot” (Anderson, 2008)... a specific type of flat ontology...

In the political realm, these techniques have much more profound repercussions – the manipulation of the voting population. Andrejevic explores a whole set of techniques from Big Data number-crunching, through neuro-marketing, sentiment analysis, and body language analysis that make our actions more predictable, all the while reinforcing an asymmetry of power.

At its most dystopian, the resulting landscape is one in which those with access to the database can derive practical, if probabilistic (“post-comprehension”), knowledge about how best to influence populations while members of these population[s] are left with an outmoded set of critical tools...In somewhat more concrete terms, this dystopia would be one in which political parties, for example, might use giant databases to exert influence in the affective register (by determining which appeals result in triggering desired voting behaviour)...This asymmetry would free up politicians to engage in “infoglut” strategies in the discursive register (promulgating reports that contradict themselves endlessly, pitting “expert” analysis against one another in an indeterminate struggle that does little more than fills air time, or perhaps reinforce preconceptions) while simultaneously developing new strategies for influence in the affective register.
(Andrejevic, 2008, p. 154)

Depending on your view, this may lead to either the the collapse of democracy, as free will asymptotically dissipates under the wide-spread and sophisticated manipulation of the public; or the revealing of democracy for the scam it has always been, propagating the illusion of a choice that is in fact non-existent.

The possibilities that have been opened up by Big Data in terms of large scale data collection, analysis and prediction allow for the same type of predictive models to be applied to Contemporary Music. The same collapse of human agency that might happen under the predictive and manipulative models of Big Data has implications for the type of human agency that is utilized in experimodern musical practice.