Critical Data Studies

Definition

Critical data studies is a contemporary, primarily social science reflection on the social, political, and scientific challenges posed by Big Data. The prefix "critical" means that CDS primarily discusses the epistemological and methodological issues of big data in the context of power and hegemonic relations, and its questions are also formulated within this context.

Big Data

Not only quantity but a system
Kitchin: data assemblage: he technological, political, social and economic apparatuses and elements that constitutes and frames the generation, circulation and deployment of data

characteristics and nature:
- large quantity
- fast
- exponentially growing
- relational
- high resolution/granulation
- can only be created, managed, and analyzed using computers
 

1. not new

Critique of modernity/Enlightenment

Max Horkheimer

Theodor W. Adorno

Dialectic of Enlightenment

1947

Enlightenment and modernity inherently led towards totalitarianism and authoritarianism

measurements and data = uniformisation and control

expectation

reality

Promises of Enlightenment (- positivism, scientism)

empiricism and measurements will result in neutral data and information, which will lead towards a more equal, more just society, more rational state administration and organization, more fair distribution of wealth, more fair political representation.

NATURE

HUMANS

Adorno and Horkheimer, 1947 (2002): 164

"The blindness and muteness of the data to which positivism reduces the world passes over into language itself, which is limited to registering those data. Thus relationships themselves become impenetrable, taking on an impact, a power of adhesion and repulsion which makes them resemble their extreme antithesis, spells. They act once more like the practices of a kind of sorcery, whether the name of a diva is concocted in the studio on the basis of statistical data, or welfare government is averted by the use of taboo-laden words such as "bureaucracy" and "intellectuals," or vileness exonerates itself by invoking the name of a homeland."

Data is not neutral

Geoffrey Bowker: data is not raw, is always cooked
Daniel Rosenberg: data is rhetoric (data = what is given)

origins/roots?

definitions and concepts?

main problems addressed?

main questions to ask?

what are the frontiers?

what to do? (applications? activisms)

Big Data as a myth

Chris Anderson: "end of theory", correlation vs. causality, end of hypothesis-test-result trinity, end of ontological questions, data is objective and neutral

boyd and Crawford, 2012

- three levels of Big Data: technology, analysis, mythology
- challenges and debates the claims related to big data on epistemological, methodological and moral grounds
 

Claims:
1. changes the definition of knowledge (ep.)
2. objective and accurate (m)
3. the bigger the better (m)
4. meaningful without context (m)
5. ethical (et.)
6. equal (et.)

Dalton and Thatcher, 2014

Questions

• What historical conditions lead to the realization of ‘big data’ such as it is?
   
• Who controls ‘big data,’ its production and its analysis? What motives and imperatives drive their work?
   
• Who are the subjects of ‘big data’ and what knowledges are they producing?
•        
• How is ‘big data’ actually applied in the production of spaces, places and landscapes?
•        
• What is to be done with ‘big data’ and what other kinds of knowledges could it help produce?

Iliadis and Russo, 2016

"Data are form of power"