Evidence Dialogues Inaugural Workshop Thursday 12 May 13:00-17:15 CEST

[ISHPSS]

From: Mackor, A.r. <a.r.m...@RUG.NL>

Evidence Dialogues Inaugural Workshop 

Programme and Schedule 

13.00-13.15: 

Welcome by Professor Anne Ruth Mackor, Professor of Professional Ethics, University of Groningen 

Introduction to Evidence Dialogues by Professor William Twining, Emeritus Quain Professor of Jurisprudence, University College London. 

13.15-14.15: Trudy Dehue - What’s in our facts? The example of facts produced with the randomized controlled design 

The ingredients of scientific facts, just like the ingredients of food, shape people’s lives. This is because of the range of normative criteria and definitions that precede scientific fact production. For instance, poverty research is based on a normative definition of ‘poverty’ that subsequently shapes poverty policies, whilst testing the efficacy of an antidepressant is based on normative definitions of both ‘efficacy’ and ‘depression’ that subsequently shape treatments for depression, as well as wider views on unhappiness. That is not to say, though, that facts are little more than mere opinions. Rather, that some facts can be better than others - not so much because they represent a pre-given reality, but because they are more thoroughly warranted in a philosophical, ethical and social sense. Put differently, apart from statistical precision, ‘deliberative precision’ is the difference between ‘facts’ and ‘opinions’. It follows that researchers need to devote much more time to thorough reflection and open democratic knowledge-production than is currently the case.

 

14.30-15.30: Dave Lagnado - Causal models and Bayes: natural or uneasy bedfellows? 

Can we formalize principles of good evidential reasoning? If so, how well do people conform to these principles? And how can we improve their reasoning when they deviate? In this talk I will argue that taking a causal model perspective on evidence helps us address all three questions. First, a causal framework moves us beyond a purely probabilistic conception of evidence, helping us to analyse complex interrelations between hypotheses and evidence, and to address issues of relevance, reliability and strength of evidence. Second, in the psychology of evidential reasoning, causal thinking is typically cast as a heuristic that conflicts with normative principles. But once we adopt standards of inference based on causal (probabilistic) principles we can see how and why people deviate from normative standards. Third, once we accept that people are inherently causal thinkers, we can devise better methods for improving reasoning and avoiding biases. 

15.45-16.45: Joe Roussos - What kind of evidence do climate simulation models provide? 

The core problem of climate science in the 21st century is to predict the course of climate change, so that we can mitigate the worst changes and adapt to those which remain. To accomplish this task, climate scientists make use of complex simulation models. These attempt to capture the core dynamics of the climate, in order to predict its changes as various inputs—such as anthropogenic greenhouse gas emissions—change over time. But what kind of evidence about the future do these simulations provide? Some philosophers and scientists take them to describe possible future states of the climate, but no more. This seems insufficient to support their policy-guiding ambitions, however. A more mainstream approach uses them to construct conditional probabilistic predictions for what will happen. But while this is in principle more useful to decision-makers, it is not clear that it is warranted. In some particularly difficult cases, climate scientists are beginning to move away from the seeming objectivity of mathematical simulation models towards ‘softer’ but more flexible expert judgements. I close with some reflections on the social and political pressures which influence such choices. 

16.45-17.15: Evidence Dialogues future plans and closing. 

This workshop will be delivered online via Zoom. Please email Lynne Chave (l.c...@ucl.ac.uk) to obtain the link and passcode for the event, or if you have any questions. 

Please note, the lectures (but not the subsequent discussions) will be recorded and made available on the Evidence Dialogues website. 

https://evidencedialogues.wordpress.com/

To receive details of future news, activities and events, please send an email to l.c...@ucl.ac.uk or follow us on Twitter@EvidenceDialog.