Rational Trust


Thursday, December 9

12.30 - 13.00: Coffee
13.00 - 14.20: Paul Faulkner (Sheffield): The Practical Rationality of Trust
14.30 - 15.50: Nikolaj Jang Lee Linding Pedersen (Copenhagen): Rational trust and epistemic consequentialism
16.00 - 17.20: Erik Olsson (Lund): Modelling Epistemic Trust: A Bayesian Perspective

Friday December 10
09.30 - 10.00: Coffee
10.00 - 11.20: Kristoffer Ahlström (Copenhagen): Required Trust
11.20 - 12.40: Gloria Origgi (Jean Nicod): Rational trust and reputation: for a second-order epistemology
12.40 - 13.40: Lunch
13.40 - 15.00: David Owens (Sheffield): Testimony and Truthfulness
15.00 - 16.20: Klemens Kappel (Copenhagen): Epistemic Trust

Workshop attendance is free. However, anyone who wishes to attend the workshop should email Nikolaj J. L. L. Pedersen (nikolaj@ucla.edu) to register by Monday December 6.

The workshop is funded by the Velux Foundation, as part of the research project The Epistemology of Liberal Democracy. The project is hosted by the Social Epistemology Research Group (SERG) at the University of Copenhagen.

Title: The Practical Rationality of Trust
Speaker: Paul Faulkner
Abstract: Most action can be explained in Humean or teleological 
terms; that is, in most cases, one can explain why someone acted by 
reference to that person's beliefs and desires. However, trusting and 
responding to trust are actions that do not permit such explanation. 
The acting of trusting someone to do something is a matter of 
expecting someone to act for certain reasons, and the act of 
responding to trust is one of acting for these reasons. It is better 
to say that people act out of trust, rather than for some end. And one 
can act out of trust, one could trust S to A, say, without believing 
that S will A. Moreover, S could then A in response to one's trust, 
without any rationalising desire, but merely for the reason of one's 
trust. Thus, teleological considerations do not suffice to understand 
certain acts of trust. This is the negative claim of the paper. Its 
positive proposal is an account of the practical rationality of trust. 
The key idea is that in trusting one takes on commitments, not merely 
to act in certain way, but also to premise one's practical reasoning 
on a trust-based view of the interaction situation.

Title: Scepticism, rational trust and epistemic consequentialism
Speaker: Nikolaj Jang Lee Linding Pedersen
Abstract: According to some views of the structure of warrant, it is 
not possible to acquire an evidential warrant--a justification--for 
thinking that there is an external world. Due to epistemic 
circularity, any attempt to acquire such a warrant will result in a 
principled warrant-transmission failure. Similarly for other 
anti-sceptical hypotheses. Owing to the principled absence of evidence 
supporting our positive attitude towards anti-sceptical hypotheses, it 
has been suggested that the proper attitude is not one of belief, but 
one of trust. I discuss whether trusting anti-sceptical hypotheses can 
be rational, despite the principled absence of positive evidence 
supporting them. I adopt a consequentialist framework and investigate 
whether trusting anti-sceptical hypotheses can be regarded as rational 
in the sense of maximizing expected epistemic value.

Title: Modeling Epistemic Trust: A Bayesian Perspective
Speaker: Erik J. Olsson
Abstract: Bayesians are committed to the view that an epistemic 
agent's belief state at any given time can be represented as a 
probability distribution over propositions in some language. Bayesians 
also believe that a rational agent should react to incoming evidence 
by means of conditionalization. This is a powerful model of belief 
chance which has had, and still has, many advocates. It could be 
argued, however, that this picture is not entirely complete without an 
account of trust: when we receive information from other sources we 
also tend to adjust our trust in those sources. For instance, if the 
information was expected, this should count, if ever so slightly, in 
favor of the source's credibility. If, by contrast, the information 
was surprising, that might lead us to question trustworthiness of the 
source. How this can be modeled within the framework of Bayesians is 
very much an open question. In the talk, I will present, and try to 
justify, a particular way of modeling and updating trust as 
implemented in the Bayesian simulator Laputa.

Title: Required Trust
Speaker: Kristoffer Ahlstrom
Abstract: There has been a lot of discussion in recent epistemology 
about the conditions under which we are permitted to trust others. 
However, very little attention has been paid to the conditions under 
which we are obligated to trust others. The present investigation 
seeks do something about this oversight by putting forward two 
necessary and jointly sufficient conditions for when trust is 
required. First, the alternatives to trusting the relevant sources 
need to be sufficiently unpalatable from an epistemic point of view, 
for example by being tantamount to stifled epistemic progress or a 
dwindling reduction in the things we can possibly come to know. 
Second, the subjects involved need to have evidence that the sources 
in question are reliable on the relevant issues. In order to assuage 
the worry that the obligations associated with such required trust are 
irrelevant to who or what we are obligated to trust all things 
considered, two recent controversies involving lack of trust are 
considered, where we not only have pro tanto but all-things-considered 
obligations to trust the sources involved.

Title: Rational trust and reputation: for a second-order epistemology
Speaker: Gloria Origgi
Abstract: We do not choose to trust. In most situations we just do not 
have the
choice. That is why trust is an active stance, a fundamental epistemic
resource that allows us to navigate in a reasonable way in the world of
knowledge. Given this view of trust, I will argue in this paper that the
distinction between reductionist and anti-redictionist approaches to
epistemic trust in the epistemology of testimony is, to a certain extent,
artificial. We monitor the informational environment and catch reputational
cues, gather signals from our informants and develop our trustful attitude
in context. I will present the project of an epistemology of reputation as a
way of sustaining trust as a rational tool to acquire information.

Title: Testimony and Truthfulness
Speaker: David Owens
Abstract: By making assertions, we change the normative situation in 
at least two respects. First assertions have *epistemic* significance: 
they change what it is reasonable for the audience to believe. Second, 
they have *deontic *significance: they change what the speaker owes to 
his audience. I assume these two aspects of the normative significance 
of assertion must be congruent and I investigate various ways of 
fitting them together.

Title: Epistemic trust
Speaker: Klemens Kappel
Abstract: Apparently, we often believe what someone says because we 
trust that person, where the trusting is in some sense the reason, or 
at least part of the reason, for the belief. Generally, one cannot 
assume that trust is a distinct type of relation or psychological 
attitude; it is likely that there is a family of related phenomena and 
that 'trust' does not denote a unified psychological attitude or 
relation. Different accounts of trust may simply conceptualise 
different phenomena going under the same name, and in part believing 
someone on trust overlaps with believing on testimony. I shall use the 
term 'epistemic trust' for one kind of trust that I will try to 
characterise. The main focus will be the kind of trust that figures in 
Hardwig's papers, though the aim is not exegetical. First, I will 
discuss the way in which epistemic trust may seem necessary for 
acquiring knowledge in certain knowledge producing social systems. 
Second, I will make some more specific suggestions about what would 
have to be true about epistemic trust for it to fulfil this role 
(suggest that it is a non-inferential disposition to believe, which 
might be emotionally based). Third, I will discuss whether true 
beliefs acquired on the basis of epistemic trust might qualify as 
known and justified. I shall suggest that this is indeed so, though 
the extent to which our actual known beliefs are based on this type of 
trust is an open empirical one. Finally, I will discuss if epistemic 
trust thus understood is different in kind from conceptions of trust 
proposed in the literature.