Science, data and decision making in an era of public controversies

A sociological perspective

Francis Chateauraynaud (GSPR, EHESS, Paris)

Eurostat Conference, Luxembourg, 28 nov. 2016

1. Ways of knowing, forms of evidence and abductive reasoning 2. Dealing with non-linear processes. Some epistemic Lessons from environmental Issues 3. The rise of public Expertise : four Models between Science and Politics 4. Data, Governmentality and the Chaos of Controversies 5. A socioinformatics Experimentation : following complex Dynamics by modelling evolutive Configurations of Actors and Arguments

Source : Prospero/Tiresias, extracted elements from the database on nuclear controversies 1/6

Source : Prospero/Tiresias, extracted elements from the database on nuclear controversies 2/6

Source : Prospero/Tiresias, extracted elements from the database on nuclear controversies 3/6

Source : Prospero/Tiresias, extracted elements from the database on nuclear controversies 4/6

Source : Prospero/Tiresias, extracted elements from the database on nuclear controversies 5/6

Source : Prospero/Tiresias, extracted elements from the database on nuclear controversies 6/6

1.

Ways of knowing, forms of evidence

and abductive reasoning

“Contemporary forms of government are marked by the rise of indicators, measures and new metrics to compare, certify, codify and evaluate. In many countries, performance measurement has become one of the symbols of the transformation of governance. […] Focusing on instrumentation makes it possible to avoid the functionalist reification of instruments (the resolution of problems) and the limitations of constructivism (actors do not invent everything, all the time, and their trajectory often explains nothing of significance). Empirically, instrumentation involves associating reflection on the development and choice of instruments with their implementation in order to identify their uses and understand their outcomes. There are products and instrumentation outputs in terms of the choice and selection of specific procedures of policy implementation through instruments, budgets, rules, norms and standards” (Patrick Le Galès, Performance Measurement as a Policy Instrument, Sciences Po, June 2016)

Studying controversies leads to consider three different regimes used by players in order to produce some evidence :

• Axiomatic Regime : computational spaces based on autonomous axiomatic (formal or syntactic dimension of reasoning and proof)

• Conventional Regime : production of data organized under isomorphic constraints or principles of equivalence, involving categories and collections of validated facts (semantic dimension of proof)

• Phenomenological Regime : generating cross-checking tangible elements by grasping a variety of empirical experiences in specific contexts (pragmatic dimension of proof)

This distinction produces a different understanding of evidence in the making than the three regimes of objectivity presented by Daston & Galison (2010), but they can be interlinked

“Hormones and Endocrine-Disrupting

Chemicals: Low-Dose Effects and

Nonmonotonic Dose Responses”

Laura N. Vandenberg, Theo Colborn,

Tyrone B. Hayes, Jerrold J. Heindel,

David R. Jacobs, Jr., Duk-Hee Lee, Toshi

Shioda, Ana M. Soto, Frederick S. vom

Saal,

Wade V. Welshons, R. Thomas Zoeller, and

John Peterson Myers

Endocrine Reviews, June 2012

Dealing with a plurality of forms of knowledge

without any linear pathway

Scientific controversies make visible permanent oscillations between: The attempt of a paradigmatic purification of scientific production, under

the hegemony of a discipline seeking to impose a formalism in front of the resistance of other epistemic communities;

Conflict between different paradigms with the introduction of discontinuity

and incommensurability; Tipping points due to a saillant discovering or a marking event, with more or

less deep cognitive shift; The extension of a postnormal science dealing with complex systems, non-

linearity, uncertainty and variability depending on scales of observation;

A rise of participatory research and of open social investigation or expertise dealing with the heterogeneity of sources of (sometimes contentious) knowledge.

Evolving environmental challenges: from specific to systemic

Characterisation of

key challenges

Key features In the spotlight in Policy approaches

(examples)

Specific linear cause-effect,

large(point) sources,

often local level

1970s / 1980s

(continuing today)

targeted policies

and single-issue

instruments

Diffuse cumulative causes,

multiple sources,

often regional level

1980s / 1990s

(continuing today)

policy integration

and raising public

awareness

Systemic complex causes,

interlinked sources,

often global level

1990s / 2000s

(continuing today)

policy coherence

and systemic

approaches (long-

term transitions)

Source: EEA, SOER2010

From monotonic to non-monotonic reasoning 1 Classical mode of reasoning: Modus Ponens / Modus

Tollens

Monotonic Logics

2 P implies Q; P is asserted to be true, so therefore Q

must be true.

Foundation of classical rationality

3 Pragmatics of Justification (Pragmatics according to

Boltanski & Thévenot, 1991)

Putting Rationality in context and connecting

social situations to universal principles of

foundation

4 S (situation) congruent to R (Regime of Justification),

then S implies R – if not true, then S implies D (Dispute)

The plurality of Principles or Regimes of action

and/or justification explains the numerous

disputes in social life

5 Pragmatist sociology of transformation: Events,

actions and argumentations are transformed through

interpretive activities, producing a new series of

actions, argumentations and decisions, with a huge

uncertainty on practical consequences

The plurality of interpretations creates both

constant reevaluations of norms and rules, and a

creative/collective activity with emerging actors,

new ways of arguing and innovative devices or

tools. 6 In a situation S, E (event) or A (action) seized by I

(interpretative activity) produces A (argumentation)

from which proceeds the evaluation or valuation (V) of

available rules (R0… Rn) but, at the same time, of

known and unknown consequences (C) which

evaluation or valuation (V) retroacts not only on the

situation S but on a series of future situations (F)

Non-monotonic or complex logics with emerging

and retroactive consequences (non-linear process

and feedback)

2.

Dealing with non-linear Processes

Some epistemic Lessons from environmental Issues

Figure SPM.9 from IPCC report Climate Change 2014:

Impacts, Adaptation, and Vulnerability: Opportunity space

and climate-resilient pathways

Environmental issues are entering a new period shaped by a number of on-going processes from international to local. We (in social science) observe an increasing tension between the expansion of environmental standards and norms on the one hand, and the emergence of multiple sources of conflict on the other.

F. Chateauraynaud et alii, « Towards collaborative environmental studies in an era of conflicts over global environmental issues », Position Paper written by the ‘Barcelona Group’ (spring 2015)

General Model of Transformation for Warnings and Controversies (2011)

Accident

Conflict

Legal action

New alert

Public consultation

Unveiling conflicts of

interests

New scientific findings

…

Andrew Abbott, “On the concept of turning point”

in Time matters. On theory and Methods

The University of Chicago Press, 2001.

“The concept of turning point is a « narrative concept ». That is, the concept has reference to two points in time, not one. What makes a turning point a turning point rather than a minor ripple is the passage of sufficient time ‘on the new course’ such that it becomes clear that direction has indeed changed.” (Abbott, p. 245)

“What makes the trajectories trajectories is their inertial quality, their quality of enduring large amounts of minor variation without any appreciable change in overall direction or regime.” (Abbott, p. 248)

How do actors discuss the reversibility or irreversibility of a turning point, try to avoid it upstream – as they do in risk assessment – or to interpret the causality and consequences afterwards or downstream, after a marking event has occurred.

Le Monde, 10 11 2016

3.

Public Expertise and collaborative foresight : four Models between Science and Politics

Expertise is at stake

in all the processes observed

• The classical model of expertise has been under huge critique: asymetry, lack of transparency, conflict of interest, late knowledge, old fashioned verificationism and epistemology, monodisciplinary reasoning etc

• New forms of expertise dealing with more complex issues: contradictory expertise, collective, distributed, participatory …

Expertise / Counter-Expertise

Dissensus and conflict on facts and matter of facts

Legal suits on courts Political debates

Resolving the conflict with a decision

Greenpeace CRII-RAD / ACRO CRII-GEN Robins des Toits NRDC

Collective Expertise

Plurality of viewpoints and multidisciplinary approach oriented to a consensus-based assessment

The procedure is conducted by a national or international agency, generally after new scientific publication or marking events creating a shift in assessment

Making a strong consensus

Comité Dormont ESB IPCC INSERM amiante 1997 WHO 2012 INSERM pesticides 2013 IPBES

Distributed Expertise

A diversity of actors with different skills and tools, interests and knowledge, agencies, private labs, NGOs, contribute to a trend in balancing certainties and uncertainties, knowledge and ignorance, standard evaluations and specific experimentations or experiences

Flux of studies, reports, research and reviewing, which define, without a central organizer, the common knowledge and the collective norms - e.g. good practices

Convergence and divergence which produce a space of relevant scenarios

Nanos Endocrine disruptors Energy scenarios (IEA, Ademe, NegaWatt… )

Dialogic or participatory Expertise

Direct interactions between scientists, enginers, stakeholders and the public

Open consequentialism by which lay experiences and the plurality of life forms are taken into account for collective assessment

Reframing trust in public expertise and decision

Danish Board of Technology Citizen panels Digital democracy Participatory research

Michel Godet and Fabrice Roubelat

Creating the future : The use and misuse of scenarios

(1996)

A sociological

Ballistics of

complex critical

processes (2011)

“The future is not yet written but remains to be constructed” Gaston Berger

We must distinguish contigent events, weak signals, emerging issues, long run processes and mega trends, by bringing together ruptures and gradual transformations…

“I have alerted very early about the problem of lack of technical control on off-shore platforms and now we are in front of the biggest oil slick in American history! How would we avoid this kind of catastrophe in the future? How to be sure that it will never occur again?” (intervention by an inspector, in may 2010, in the course of the big controversy surrounding the management of the disaster caused by the explosion of Deepwater Horizon Platform - fragment extracted from a corpus built from American news sites)

Peter Gärdenfors and David Makinson, « Revisions of

knowledge systems using epistemic entrenchment » (1988) • A major problem for knowledge representation is how to revise a

knowledge system in the light of new information that is inconsistent with what is already in the system. Another related problem is that of contractions, where some of the information in the knowledge system is taken away.

• Gärdenfors attacks the problems of modelling revisions and contractions in

two ways. First, two sets of rationality postulates or integrity constraints are presented, one for revisions and one for contractions. On the basis of these postulates he shows that there is a natural correspondence between revisions and contractions.

• Second, a more constructive approach is adopted based on the "epistemic

entrenchment" of the facts in a knowledge system which determines their priority in revisions and contractions. We introduce a set of computationally tractable constraints for an ordering of epistemic entrenchments.

4.

Data, Governmentality and the Chaos of Controversies

“There are three main challenges. Firstly, economic theory has de-coupled from statistics, weakening the triangle between statistics, economic theory and political practice. Secondly, current management ideologies demand measurements of progress in the public sector, in a context where many dimensions cannot currently be quantified. Finally, while big data is seen as a solution to improving our understanding of how our social, economic and environmental systems intersect, this overlooks the need to refine data and ensure quality before drawing conclusions. We need to expand our ability to measure complexity in systems.” W. Radermacher, contribution in the Report of the EEA Scientific Committee Seminar on emerging Systemic Risks, Copenhagen, 24 February 2016

4.

Big data, Governmentality and the Chaos of Controversies

Pragmatics of transformations: 6 spheres/6 sociologies Chateauraynaud & Debaz, forthcoming (2017)

5.

A socioinformatics Experimentation :

following complex dynamics by modelling evolutive configurations of actors and

arguments

Prospéro

Semantical and

statistical text

analysis

Tirésias

web crawling,

data mining

and data

crunching

developed by a

community of

users

Chéloné

Database interfacing multiple

corpuses for sharing and archiving

Marlowe

natural language

interface e-

sociologist (AI)

producing

reports for

researchers

Text databases, algorithms, tools, procedures and … digital sociologists

This graph is based on statistical distribution of the years cited in discourses and texts that compose

the French corpus concerning asbestos analysed with Prospéro.

Asbestos public trajectory in France

0

200

400

600

800

1000

1200

1898

1902

1906

1910

1914

1918

1922

1926

1930

1934

1938

1942

1946

1950

1954

1958

1962

1966

1970

1974

1978

1982

1986

1990

1994

1998

2002

2006

2010

2014

2018

2022

2026

2030

workplaceaccidents law

First signs and alarms

Mesotheliom : official recognition as asbestos illness

Huge mobilization, anti-asbestos group inUniversity of Jussieu

Contamined Air Scandal (1995)

Ban of Asbestos in France (1997)

Senate report recognizing state responsability

Thousands of deaths are coming …

First Social Crisis

Silent periodno mobilization

Political CrisisJudicialCrisis International

struggle goes on

Mapping the main issues connected to alerts and risks (Prospéro trans-corpus)

Tweetoscope

Capture of a window from Prospero on Climate change

Source Nbtxt First date Last date ling

Yahoo News 12086 29/08/2013 00:00 01/04/2016 00:00 en

AP 6516 24/01/2014 00:00 31/03/2016 00:00 en

TheConversation-uk 5157 13/08/2013 00:00 24/11/2016 00:00 en

The World Bank 4780 04/04/2013 00:00 23/11/2016 00:00 en

nuclear-news 3793 30/12/2015 00:00 24/11/2016 00:00 en

BBC 3751 26/07/2010 00:00 25/11/2016 00:00 en

Reuters 2949 02/04/2014 00:00 01/04/2016 00:00 en

World Nuclear News 1142 10/06/2015 00:00 24/11/2016 00:00 en

TheEcologist 537 01/03/2014 00:00 17/08/2016 00:00 en

UNEP 383 04/04/2013 00:00 15/11/2016 00:00 en

WHO 269 12/11/2013 00:00 21/11/2016 00:00 en

European Commission 236 04/12/2014 00:00 24/11/2016 00:00 en

Corporate Europe Observatory 192 08/07/2014 00:00 17/11/2016 00:00 en

European Environment Agency 129 20/11/2013 00:00 23/11/2016 00:00 en

Nuclear Industry Association 126 31/07/2015 00:00 03/11/2016 00:00 en

EFSA 88 03/03/2016 00:00 08/07/2016 00:00 en

Jonathon Porritt 66 24/07/2015 00:00 07/11/2016 00:00 en

The Food & Environment Reporting Network 63 22/05/2013 00:00 18/06/2015 00:00 en

OCDE 39 21/01/2016 00:00 24/11/2016 00:00 en

CBS News 6 02/07/2015 00:00 25/02/2016 00:00 en

Towards a collaborative observatory of controversies in Europe (inspired by the experience with ANSES (2006-2014) 1/2

Preserving the memory, in the long run, of many risk issues that would be

difficult for individual human agents to describe and to analyze ; making the history available for further evaluations and discussions.

Evaluating news flows, by monitoring the full history of each issue

through iterative procedures and within previous corpuses.

Conducting fieldwork on various sites (with scientists, institutions, corporations, NGOs, local populations…) allows to complement data from the public sphere.

Identifying emerging or re-emerging issues and anticipating their potential development

Dealing with uncertainties and indeterminacies through an approach assuming that we do not know in advance the trajectory of a public problem

Strengths and Weaknesses of the socio-informatics observatory developed with ANSES (2006-2014) 2/2

Closely examining the diverse interactions between public controversies:

risk issues are not independent from each other ; Identifying the properties of turning points leading to disagreements, on

the same footing with consensual processes. Finding a way toward a form of “objectivity” by taking seriously all the

positions, even when issues are clearly characterized by struggles between powerful vested interests.

Helping academics, agencies, NGPs etc deal with conflicting temporalities of politics, science and media

Developing and/or applying collaborative digital methods in order to

generate a continuous process of collective learning – a sort of reflexive socialization of digital data and AI

Thank you for your attention