Sommers, Fred - Dissonant Beliefs

11/20/2009

Analysis, Vol 69 No 2 April 2009

This article attempts to change the conception of a belief as a propositional attitude. 'Propositionalists', as author calls them, believe that believing is primarily a relation between a believer and a proposition. Author also considers Fodor a propositionalist even though he gives a three-part relation, a believer, a sentential expression, and a proposition. Author's proposal is different: believing is a relation between a believer and the world. This view author dubs the 'mondialist' view, and believes it can account for so-called 'dissonant beliefs', which are beliefs where a person seems to believe a contradiction (or an inconsistency, since 'contradiction' may imply propositions)

Author lays out the standard two kinds of belief: de re and de dicto. He wants to add a third: de mundo, which takes the world to be 'xish'. Here is how the progression might work:
-Formation of a de re belief about the ringing of a smoke alarm
-Formation of a de mundo belief: the world is smoke-alarm-ish
-Possible, but not necessary or automatic: de dicto belief: 'There is a smoke alarm ringing'.
The separation between the first two and the third gives space for non-human animals to have beliefs, de re and de mundo, but not de dicto. (pg270)

In order to make this work, 'the world' needs a bit of a specialized understanding. Author gives it: 'the world is characterized by what is and is not in it.' (pg268-9) So, our world is 'elk-ish' and not 'elf-ish' since there are elks in the world but not elves. The second understanding is that there is a context that constrains mondial beliefs-- a 'Domain under Consideration' (DC). We use this when we say things like "it's raining". It isn't raining in the whole world-- just in our DC-- in our DC the world is rain-ish.

Author believes that mundial beliefs are more primitive than propositional belief-- he loosely translates mundial beliefs as 'being aware of something'. (pg270) The two advantages of having them is that they are good fits for non-human animal beliefs and also that they can account for 'dissonant beliefs'.

Author takes dissonant beliefs to be commonplace in our world. For the well-educated, dissonant beliefs come when we believe a proposition (de dicto) like E=mc2 but have no idea what the world is like because of it, or what it would be like if it weren't true. Author thinks it is rational to keep your mundial and your de dicto beliefs in line, but it may be very difficult given the amount of arcane scientific knowledge we are exposed to. (pg272) But a more mundane example is where someone who doesn't believe in the afterlife still feels as though her dead relative is checking in on her. Here a mundial belief of 'the world is father-ish' is inconsistent with the propositional belief 'my father is dead and gone'. (pg271)

Lastly, author advocates that treating beliefs as primarily non-propositional will allow for uncomplex analyses of difficult examples. The one he uses is of a man who sees a reflection through a window of a man with his pants on fire. He believes de dicto 'there is a man with his pants on fire', but does he believe this man is himself? Propositionalists, author claim, need to do cart in the concept of a 'presenting sentence' ('my pants' or 'somebody's pants'). Author simply says that the man can have two different mundial beliefs: 'the world is my-pants-on-fire-ish' or 'the world is somebody's-pants-on-fire-ish'.