Personality & Individual Differences · Simplified
Five Explanations, All of Them Fit
Explaining somebody after the fact is easy. That is the problem with it.
The case
What Priya did
In a team meeting, a difficult and unglamorous project was described and volunteers invited. Three people gave reasons why they could not. After about forty seconds of silence, Priya volunteered.
Priya is fictional, and her behaviour was written to be compatible with every explanation below. That is not a trick. It is the ordinary condition of explaining somebody from the outside.
Step 1 of 2
First, a judgement
You are about to be given five explanations, and every one of them accounts for what Priya did. What would make one of them better than another?
Step 2 of 2
The jury room
Three pieces of evidence can be requested. Ask for them in any order and watch what each one does to the five explanations.
Key terms
- Fit and prediction
- Whether an account can accommodate what has already happened, and whether it commits itself to something not yet observed. Only the second can be wrong.
- Underdetermination
- More than one account remaining compatible with all the evidence there is. It is the normal situation, not a failure.
- Risk
- How much an explanation stakes on evidence that could have gone the other way. An account that predicts nothing cannot be wrong, and gains nothing when it turns out to survive.
| Explanation | Where it stands | Evidence it committed itself on | What it predicted that came true |
|---|
What this shows
Surviving is not the same as being supported
Key idea: Two of the five explanations were ruled out, which is worth noticing on its own: evidence does work, and an account that commits itself can be shown to be wrong. Three survived, and they did not survive in the same way. The achievement account staked itself on all three pieces of evidence and all three went its way. The self-efficacy account did nearly as well. The trait account survived having committed itself to one piece of evidence out of three, because a broad disposition is compatible with almost anything a person does on a single occasion. It is not refuted, and it is not thereby supported, and the difference between those two matters. The last row of the table is the one that will not go away. The observation that would separate the survivors is what she does next, in a different setting, and that evidence does not exist yet. Ending with more than one live account and a clear statement of what would decide between them is not a failure to reach a verdict. It is the verdict.
Priya is fictional and her behaviour was written to fit every account offered, so nothing here is evidence about anybody. Each explanation's predictions were also written down in advance by whoever built the page, which is the part real life does not supply: in practice, working out what an account actually commits itself to is most of the difficulty, and accounts are often stated so loosely that they commit themselves to nothing. Two further cautions. Some of these explanations are compatible rather than competing, so ruling one in does not rule another out, and forcing a single cause is the commonest error in this kind of reasoning. And an explanation surviving the evidence is not the same as its being true, since the next observation is always outstanding.
The longer version offers eight explanations and six pieces of evidence, asks you to rate each account before investigating, scores your confidence against what the evidence can support, and marks which pairs of accounts are compatible rather than competing. It is at Explain This Person: The Courtroom in the main collection.