Keyboard Shortcuts?

×
  • Next step
  • Previous step
  • Skip this slide
  • Previous slide
  • mShow slide thumbnails
  • nShow notes
  • hShow handout latex source
  • NShow talk notes latex source

Click here and press the right key for the next slide.

(This may not work on mobile or ipad. You can try using chrome or firefox, but even that may fail. Sorry.)

(If the slides don’t work, you can still use any direct links to recordings.)

also ...

Press the left key to go backwards (or swipe right)

Press n to toggle whether notes are shown (or add '?notes' to the url before the #)

Press m or double tap to slide thumbnails (menu)

Press ? at any time to show the keyboard shortcuts

 

Moral Dumbfounding

Hypothesis: Not all unreflective moral judgements depend on reasoning.

We will consider evidence for this hypothesis.

What is moral dumbfounding?

Moral dumbfounding is ‘the stubborn and puzzled maintenance of a judgment without supporting reasons’ \citep[p.~1]{haidt:2000_moral}.
‘Moral dumbfounding occurs when you make an ethical judgement but either cannot provide reasons or provide reasons that are ‘only weakly associated’ with your judgement’ \citep{dwyer:2009_moral}.
To understand this phenomenon, we need to review the experiment that gave rise to the term.
To understand moral dumbfounding, we need to review the experiment that gave rise to the term.

Haidt et al (2000; unpublished!)’s tasks

NB: I’m delibertately not mentioning the Heinz dilemma at this stage, for drama.

Control: ‘Heinz dilemma (should Heinz steal a drug to save his dying wife?)’

morally provocative but ‘harmless’: Incest; Cannibal

nonmorally provocative but ‘harmless’: Roach; Soul

‘(Incest) depicts consensual incest between two adult siblings, and [...] (Cannibal) depicts a woman cooking and eating a piece of flesh from a human cadaver donated for research to the medical school pathology lab at which she works. These stories were chosen because they were expected to cause the participant to quickly and intuitively "see-that" the act described was morally wrong. Yet since the stories were carefully written to be harmless, the participant would be prevented from finding the usual “reasoning-why” about harm that participants in Western cultures commonly use to justify moral condemnation‘ \citep{haidt:2000_moral}.
‘In addition we used two "non-moral intuition" tasks: Roach and Soul. [...] In [Roach] the participant is asked to drink from a glass of juice both before and after a sterilized cockroach has been dipped into it. In the Soul task the participant is offered two dollars to sign a piece of paper and then rip it up; on the paper are the words "I, (participant's name), hereby sell my soul, after my death, to Scott Murphy [the experimenter], for the sum of two dollars." At the bottom of the page a note was printed that said: "this is not a legal or binding contract"’ \citep[p.~6]{haidt:2000_moral}

Method: ask whether wrong; counter argue; questionnaire

Method: (1) ask whether the act was wrong (or whether the participant would perform the action). (2) record the answer, and any argument given. (3) experimenter argue against the participant’s position. (4) questionnaire after each task (‘The questionnaire asked the participant to respond on a Likert scale as to her level of confusion, irritation, and confidence in her judgment, and to what extent her judgment was based on reasoning or on a "gut feeling."’).
Do try this at home!
But is Isabel dumbfounded? Maybe briefly. But is that significant? Hard to measure ...
To understand moral dumbfounding, we need to review the experiment that gave rise to the term.

Haidt et al (2000; unpublished!)’s tasks

NB: I’m delibertately not mentioning the Heinz dilemma at this stage, for drama.

Control: ‘Heinz dilemma (should Heinz steal a drug to save his dying wife?)’

morally provocative but ‘harmless’: Incest; Cannibal

nonmorally provocative but ‘harmless’: Roach; Soul

Method: ask whether wrong; counter argue; questionnaire

This condition is often forgotton but it is important for two reasons.
First, a natural assumption is that we should be able to test hypotheses about relative levels of dumbfounding rather than about absolute levels. Second, ... we’ll see late in covering Dwyer‘s argument
On the importance of the control (Heinz) task: ‘Planned contrasts were performed between the Heinz task and each of the other four tasks, because we predicted that the Heinz task would be unique in encouraging analytical reasoning’ \citep[p.~8]{haidt:2000_moral}

Results

NB: unpublished data

‘it often happened that participants made “unsupported declarations”, e.g., “It’s just wrong to do that!” or “That’s terrible!”

Note the comparison with the control!

They made the fewest such declarations in Heinz, and they made significantly more such declarations in the Incest story.’

Results ctd

NB: unpublished data

Informal observation: ‘participants often directly stated that they were dumbfounded, i.e., they made a statement to the effect that they thought an action was wrong but they could not find the words to explain themselves’ (p. 9)

‘Participants made the fewest such statements in Heinz (only 2 such statements, from 2 participants), while they made significantly more such statements in the Incest (38 statements from 23 different participants), Cannibalism (24 from 11), and Soul stories (22 from 13).’

Importance of the method: this is a control

Study 2 (not reported!):

Cognitive load increased the level of moral dumbfounding without changing subjects’ judgments.

‘In Study 2 [which is not reported in the draft] we repeated the basic design while exposing half of the subjects to a cognitive load—an attention task that took up some of their conscious mental work space—and found that this load increased the level of moral dumbfounding without changing subjects’ judgments or their level of persuadability.’ \citep[p.~198]{haidt:2008_social}.
This will be important later when we are thinking about dual process theories of moral abilities.

replication / extension / review?

Royzman et al, 2015 : more recent study doubts that dumbfounding occurs \citep{royzman:2015_curious}. This study involves three experiments. They partially replicated Haidt et al, 2000, then go on to test whether subjects really believe that the incest is harmless.
Note that, unlike Haidt et al, 2000, these researchers did not use the comparison with Heinz!

‘a definitionally pristine bout of MD is likely to be a extraordinarily rare find, one featuring a person who doggedly and decisively condemns the very same act that she has no prior normative reasons to dislike’

\citep[p.~311]{royzman:2015_curious}

Royzman et al, 2015 p. 311

‘3 of [...] 14 individuals [without supporting reasons] disapproved of the siblings having sex and only 1 of 3 (1.9%) maintained his disapproval in the “stubborn and puzzled” manner.’

\citep[p.~309]{royzman:2015_curious}

Royzman et al, 2015 p. 309

Warning: Note the absent comparison with the Heinz dilemma.

summary: moral dumbfounding

we know the definition;

some evidence --- weak, but probably occurs.

‘Moral dumbfounding occurs when you make an ethical judgement but either cannot provide reasons or provide reasons that are ‘only weakly associated’ with your judgement’ \citep{dwyer:2009_moral}.

why is this relevant?

Hypothesis: Not all unreflective moral judgements depend on reasoning.

Should we make a stronger claim ... no judgements moral judgements depend on reasoning?

Remember Heinz!

There is some further relevant evidence ...
What is the role of reasoning in moral judgement? Some appear to have suggested that moral reasoning merely serves to confirm prior intuitions, special cases aside \citep{greene:2007_secret,haidt:2001_emotional}.\footnote{Although in fact Haidt’s view is more interesting. Compare \citet[p.~181]{haidt:2008_social} ‘Moral discussion is a kind of distributed reasoning, and moral claims and justifications have important effects on individuals and societies’; yet they go on to write that ‘moral reasoning is an effortful process (as opposed to an automatic process), usually engaged in after a moral judgment is made, in which a person searches for arguments that will support an already-made judgment’ \citet[p.~189]{haidt:2008_social}.} Opposing these views, \citet{hindriks:2015_how} argues that in ordinary cases of moral disengagement, moral reasoning provides anticipatory rationalization.

post-hoc rationalization

Moral reasoning merely serves to confirm prior intuitions, in nearly all cases (Haidt; Greene)

Some theorists have proposed that oral reasoning merely serves to confirm prior intuitions, in nearly all cases (Haidt; Greene). If they are right, then Dwyer’s suggestion that moral judgements are not consequences of explicit reasoning involving known principles appears sound. However ...

ante hoc reasoning

In ordinary cases of moral disengagement, moral reasoning provides anticipatory rationalization (Hindriks, 2015)

‘Moral disengagement occurs in situations in which someone is tempted to flout his own moral standards, and thereby to frustrate his desire to maintain self-consistency’ \citep[p.~243]{hindriks:2015_how}.

My view.

Moral dumbfounding shows that some ethical judgements are not consequences of reasoning from known principles

Other phenomena (e.g. moral disengagement) indicate that some ethical judgements are consequences of reasoning from known principles

The important thing for me isn’t whether you find the argument compelling or not. There’s surely much more to say. It’s that the motivating for it gives us a good question, a puzzle even.

puzzle 2

Why are ethical judgements sometimes, but not always, a consequence of reasoning from known principles?

To solve the puzzle we have to think about the role of emotion in unreflective moral judgements ...