ARTS1092: Working With Data

Week 2: ‘Raw Data is an Oxymoron’: Toward a Situated

Analysis of Data

Zoe Horn
[email protected]
What is data?
Data sets
 A 'data set' is an aggregation of
data that has been organised
and structured.

The Iris data set published by

R.A. Fisher in 1936. This data
set is one of the first scientific
data sets and is often used as a
test case for many statistical
classification techniques and in
machine learning.
Anna Ridler, Myriad (Tulips), 2018

Myriad (Tulips) is an installation of thousands of hand-labeled photographs of

tulips that were taken by the artist. The collection of images is a type of data set
that has laboriously categorized by the artist. The work shows us the skill, labour
and time that goes into generating a data set. It exposes the human labour and
decision making involved in the production of data and its classification.
‘Raw data is an oxymoron’
• Data is more than numbers, statistics, or bits of code. Data represents the world!

• Data is originally product of observations derived from sense perceptions (such

as sight, touch, hearing, smell, etc.) Data is now extracted from computational
processes – yet this data remains an abstraction of phenomena (interactions,
subjects, desires, purchases, etc.)

• Data are inherently partial, selective and representational. Data are abstractions
of phenomena that cannot be fully captured (e.g. height and weight
measurements do not allow us to understand the complexity of human beings!)

• Data is a value or set of values that represents a phenomena or concept. Data

only become ‘information’ when analysed.

• Data are generated, the criteria used in their capture has consequences.

• Data is not synonymous with fact, data is rhetorical. The meaning of data can
vary depending on its context.

• Data is not neutral and objective.

‘Raw’ Data
• ‘Raw’ data commonly refers to data that has
been collected from or by a ‘source’.

• ‘Raw’ data is data that has not been processed.

• Data is often described as ‘raw’ when it is

unstructured and unformatted. In this form, it
contains inconsistencies
‘Raw’ Data is an Oxymoron
‘Raw data is both an oxymoron and a bad idea; to the contrary, data should be cooked with
care’ (Bowker, 2005, pp. 184).

‘Data too need to be understood as framed and framing, understood, that is, according to the
uses to which they are and can be put. Indeed, the seemingly indispensable misperception
that data are ever raw seems to be one way in which data are forever contextualized—that is,
framed—according to a mythology of their own decontextualization’ (Gitleman and Jackson,
2013, pp. 5-6).
The importance of understanding framing.

Image 1: a famous image that was used as evidence that refugees threw their children overboard in 2001
The importance of understanding framing.

Image 2: an image of the same event that shows the larger context in which both adults and children
were being rescued from a sinking vessel.
Louis-Napoléon, the Imperial Prince of France
in 1849.
Meyer and Pierson, The Prince Imperial on his Pony,
Posing for a Photographer, 1849.
Claude Lévi-Strauss, the Raw
and the Cooked, 1964.
 ‘Cooked’ Data

• ‘Raw’ data cannot be separated from the ‘cooked’

(social, political, historical, environmental factors).

• Data comes from a 'source' and the source is always

situated.

• The harvesting of data always requires criteria and

these criteria must also be analysed in relation to
social, historical, political, environmental factors.

• How we think about data – i.e. how it's stored,

formatted, integrated with data structures, queried,
and so on – has a larger relation to the ways in
which we understand our forms of knowledge and
social organisation.
A situated analysis of data
Situated Knowledge: first critique (‘scientific’
objectivity)
Situated knowledge is an acknowledgment of contingencies that
operate in the production of that knowledge. The term comes to us
from the feminist science and technology studies scholar Donna
Haraway. She uses the metaphor of vision to develop this idea.

• For Haraway, the presumption of objectivity and impartiality

reproduces a god-like vision of the world, a ‘view from above,
from nowhere’ (589).

• This view from above transforms a very specific position (male,

white, heterosexual, human) into a universal. It position
knowledge produced from this position as neutral and objective.

• Haraway describes this as ‘the god trick’.

‘the god trick’
‘the god trick’

Axon police body-worn cameras are a contemporary example of Haraway’s view from above. The
body work camera is positioned as a truth-capturing device that will result in less police violence.
But the deployment of body-worn cameras has not prevented the killing of Black people by police.
Situated Knowledge: second critique (relativism)
• Haraway also critiques the idea of knowledge as purely relative and socially
constructed.

• She argues that this also constitutes a type of ‘seeing from everywhere and
nowhere’ (581).

‘Relativism is a way of being nowhere while claiming to be everywhere equally. The

“equality” of positioning is a denial of responsibility and critical inquiry. Relativism
is the perfect mirror twin of totalization in the ideologies of objectivity; both deny
the stakes in location, embodiment, and partial perspective; both make it
impossible to see well. Relativism and totalization are both “god tricks” promising
vision from everywhere and nowhere equally and fully, common myths in rhetorics
surrounding Science’ (584).
Situated Knowledge / Situated Data
Haraway argues that there is no such thing as a ‘view from nowhere’. Rather,
knowledge production must always proceed from the assumption of a
locatable and partial perspective.

Situated knowledge ‘privileges contestation, deconstruction, passionate

construction, webbed connections, and hope for transformation of systems of
knowledge and ways of seeing’ (585).

A situated approach to the study of data considers both the data (numbers,
measurement and metrics that have been collected) and the things we bring
to the data (the protocols, organisational processes, measurement scales,
technologies, categories, standards, ideologies, biases and so on). Such an
approach is concerned with studying what data does, that is, how it shapes
the way we understand the world.
Data is like matryoshka
dolls
 Or networks of
entangled relations
and interests.

Mark Lombardi, Oliver North, Lake Resources of Panama, and the

Iran-Contra Operation, ca. 1984-86 (fourth version), 1999.
Looking under data
We will both look into data and under data rather than simply looking at data sets,
numbers, and figures. This approach will lead us toward a consideration of the
ways that data intersects with social, political, economic, and environmental issues.

An excerpt from Kate Crawford and Vladan Joler’s Anatomy of an AI System schematic, 2018.
Seeing data like an Atlas

‘An atlas is an unusual type of book. It is

a collection of disparate parts, with
maps that vary in resolution from a
satellite view of the planet to a
zoomed-in detail of an archipelago.
When you open an atlas, you may be
seeking specific information about a
particular place—or perhaps you are
wandering, following your curiosity, and
finding unexpected pathways and new
perspectives.’ (Crawford 2021, 9-10)
Seeing data like an Atlas
‘An atlas presents you with a particular
viewpoint of the world, with the imprimatur of
science—scales and ratios, latitudes and
longitudes—and a sense of form and
consistency. Yet an atlas is as much an act of
creativity—a subjective, political, and aesthetic
intervention—as it is a scientific collection.’

‘Maps, at their best, offer us a compendium of

open pathways—shared ways of knowing—that
can be mixed and combined to make new
interconnections. But there are also maps of
domination, those national maps where territory
is carved along the fault lines of power: from the
direct interventions of drawing borders across
contested spaces to revealing the colonial paths
of empires.’ (Crawford 2021, 10)
Two maps of the same space that tell two different stories about
place, belonging, and sovereignty
Case study: Anatomy of an AI System
The Amazon Echo (3rd Gen) Smart Speaker with Alexa
The Amazon Alexa Super Bowl Ad 2018
The Amazon Echo (3rd Gen) Smart Speaker with Alexa
L: Cecilia Vicuña’s portrait of Karl Marx; R: Marx’s critique of political economy: Capital
Case study: Deaths Inside Database
Warning:
The following slides contain references to, and images of, deceased Aboriginal and
Torres Strait Islander people. The example we are about to look at contains
information about Indigenous deaths in custody. If you think this material may
distress or upset you, please feel free to stop watching now.
Source: Statista
‘The statistics should elicit shock and
outrage, but they also can flatten the
complexity of this issue and overshadow
the stories of the people who lost their
lives.

Numbers conceal the full extent of the

violence, just as official explanations for
deaths – “natural causes” – have a way of
rendering invisible the multiple forms of
state-sanctioned violence that force
Aboriginal people into jail in the first
place.’

– Amy McQuire, The Saturday Paper (June

6-12, 2020)
Situated Rendering: Deaths Inside Database
Ruth Wilson Gilmore tells us that abolition is not merely an
endpoint but is a practice. That is, abolition is about building
different systems of support and infrastructures that seek to
resolve existing inequalities.

‘Abolition abolition isn’t just absence… abolition is a fleshly and

material presence of social life lived differently.’

‘So, abolition is a theory of change, it's a theory of social life. It's

about making things.’

– ‘Prisons and Class Warfare: An Interview with Ruth Wilson

Gilmore’