Porosity, virality and the digital study of contagion

Abstract

How affects, ideas, dispositions and ‘facts’ spread in a digitally mediated milieu is an increasingly prevalent component of how we interrogate the performance of the ‘social’. This paper uses recent and ongoing work around the idea of ‘contagion’ as a means to open up conceptual and methodological debates in order to chart such digital geographies. In particular, we reflect on work concerning the contagion of ideas in social media through searching, scraping and graphing techniques (Marres and Weltevrede, 2013) to ask how do things spread, or better, how can we access the bio-social atmospheres that are the very conditions for contagion?

Exploring the affects of technological ‘black-boxing’, in proprietary digital networks, highlights both the challenges and the potentials of social media research. Thus, in this paper we critically reflect on the practices and technological engagements that underpin investigations of data and online life. Using examples of events performed with and through social media, this paper will explore the rise of ‘digital contagions’, their architecture and formation of networks. Such activities allow for understandings of how contagious atmospheres arise and how little of our thinking, reasoning, emotions or even our cells are ‘ours’. From micro-biomes to somnambulant subjectivities, ‘we’ are, it seems, porous selves.

This porosity has been understood for over a century (Deleuze and Guattari, 2004; Tarde, 1899, 1902, 1903, 2012) but it is only now, as Latour and Lipenay (2009) and others (for example: Sampson, 2012) have argued, that we can start to navigate the bio-social using the affordances of digital trace data. Nevertheless, this remains challenging work and this paper critically engages with both the opportunities and difficulties present in researching our digitally mediated porous selves.

Full text

Porosity, virality and the digital study
of contagion
Sam Kinsley, Exeter
Steve Hinchliffe, Rebecca Sandover, Exeter
RGS-IBG 2014

this talk
1. Research context
2. Some pragmatic (methodology) issues
3. Case studies
4. Theorising collective & contagion

context
• The project – ‘Contagion’
– Team: Prof. Steve Hinchliffe, Dr Rebecca Sandover & Prof. Clive
Sabel (Bristol)
– Funded: ESRC “Transforming Social Sciences Research” theme
– Rapid turnaround
• Starting points
– Connected, online, mobile
– A dizzying world, competing for attention
– Desires and beliefs are contagious
– ‘Chatter’ a key element in generating social life

twitter
• ‘an algorithmically managed infrastructure company’ —
James Bridle (misappropriated)
• A commercial social networking platform for publishing &
exchanging short messages (<=140 chars)
• A communicative infrastructure
– Retweets: boosting or reiterating a message
– Mentions: referring to another entity by their account name e.g.
‘@contagionexeter’
– Hashtags: folksonomic categories
• A ‘big’ data repository (over 1010 tweets added per day)
• An advertising & user-profiling system

pragmatic issues
• Twitter data is proprietary
• Access is a commodity
• ‘geo–’ is just another data point (tiny minority)

nature of the data
Twitter data is proprietary
– access is tightly controlled
– its use is subject to contractual obligations
So:
– one cannot share it in a repository

access to data
• Twitter is a commercial service
• Access is mediated (at a cost)
by resellers of access to the full
historical data set
• There is a free API but…

access to data
The free API limits your ability to
collect data
• Search a maximum of 6 days in
to the past (often more recent)
• Collecting streaming data is
limited to a maximum of 1% of
total traffic, beyond that: subject
to sampling (black-boxed)

the ‘geo’ in the data
Some tweets are ‘geo-tagged’
– Opt-in appending of GPS co-ords
– Only approx. 2% of all tweets
– The ‘geo’ is therefore ‘just another
bit of metadata’
-> Need to think in terms of
topology not topography

techniques for our project
‘Raw’ Twitter data
Twitter API
CSV / JSON download
Cleaning & faceting
Social networking analysis Data interrogation & visualisation
/

techniques for our project
‘Raw’ Twitter data
Twitter API
CSV / JSON download
Cleaning & faceting
Social networking analysis Data interrogation & visualisation
/

the UK badger cull
Range of activists & policy-makers, for and against,
the pilot cull of badgers in Gloucestershire and
Somerset.
Two particular hashtags: #badgercull and #TBfree

the UK badger cull
Oct-Nov 2013
#badgercull and #TBfree

the UK badger cull
Oct 2013
‘badgers’ and ‘goalposts’

the UK badger cull
#TBfree — modularity of a conversation
farmers'
support'
network'
NFU'
policy'makers'
an67cull'/'
pro7vaccine'
pro7cull'

counterpoint — Peaches Geldof
Over 300k tweets in 24 hours
– Biggest spike following
@BBCBreaking announcement
– High proportion of RTs
‘Recreational grieving’? Or –
Intervention by algorithmic agency
– ‘trending’ breeds ‘gamesmanship’
– ‘bots’ account for ~ 8.5% of ‘monthly
active users’
Example: ‘her best photos’, 6223
retweets across 7+ accounts

what collective?
Twitter is not
– a ‘mirror’ of/for society or the world
– a performance of The Social
…it is a performance of social relations
– in a system with some infrastructural constraints

ad-hoc publics
Infrastructural characteristics aid convening
particular kinds of emergent publics:
– ‘adhoc (hashtag) publics’ (Bruns & Burgess 2011)
– ‘concerned publics’ (Whatmore 2009)
The convening of devices, infrastructures, people,
software (and so on) into publics is an ontogenetic
phenomenon – it is revealed in its tracing

social media revealing ‘transindividuation’
• Assume no a priori entities
• Collectives are process(es)
• Publics are constituted in the event of their relation
– the terms of the relation are constituted by the relations
themselves
– ‘the whole is always smaller than its parts’ (Latour et al.
2012)
– ‘a partial and relative resolution manifested in a system
that contains latent potentials’ (Simondon 1992, 300)

what spreads?
Energies flow through systems in diverse forms
– affective and visceral flows, flows of signification and data
etc.
but — not ‘memes’
Potential to affect/be affected generated in each
instance of flow ~ difference in repetition.
– Expression of ‘transindividual desire’ ~ translation of
energies in the forming / performance of relations as such

what spreads?
(Sampson 2012, 21 after Tarde 1903)

Society ‘is a kind of contamination that moves constantly, from point to
point [… with] subjectivity always [referring] to the contagious nature
of desires and beliefs’ (Latour & Lippenay 2009, 9)
Contagion is not merely ‘contact’ but involves differences in repetition,
such as adaption and mutation.
Contagion is a matter of intensity more than/ as well as extension.
‘it is not the network technology… that distributes the repetition and
contagion of desire-events. The network is “the relationality of that
which it distributes…the passing-on of the event”.’ (Sampson 2012, 123
citing Massumi 2002)
what spreads?

porous ‘dividuals’
‘dividuals’ ~ physically bodied selves rendered endlessly
divisible and reducible to data representations
– a separation of the ‘self’ from its representations in data?
The ‘larval subject’ of the mediated selves is multiply
performed, opening out different performances within
different transductive milieu
Echoes of Tarde’s suggestion that the ‘social [hu]man is a
somnumbulist’ (Sampson 2012, 12-14)

an insubstantial pageant?
Digital media are not an immaterial
performance of social relations
– there are material consequences
• legal
• political/ethical
• Infrastructural
Such research is a ‘performative
accomplishment’
– data is contingent on actions of
researcher
The cloud-capped towers, the
gorgeous palaces,
The solemn temples, the great
globe itself—
Yea, all which it inherit—
shall dissolve,
And like this insubstantial
pageant faded,
Leave not a rack behind.
We are such stuff
As dreams are made on,
and our little life
Is rounded with a sleep.
The Tempest, Act. 4, Scn. 1 (Prospero)

Pass it on…
http://www.contagion.org.uk/