Read
Build the mental model
Move through the guided explanation first so the central distinction and purpose are clear before you evaluate your own work.
Abductive Logic
Capstone: Comparing Explanations in Real Cases
An integrative lesson that asks students to run the full best-explanation cycle on mixed cases: list candidate explanations, apply explanatory virtues, pick the best, and check whether the winning explanation is actually good enough or merely the best of a bad lot.
Read the explanation sections first, then use the activities to test whether you can apply the idea under pressure.
Start Here
What this lesson is helping you do
An integrative lesson that asks students to run the full best-explanation cycle on mixed cases: list candidate explanations, apply explanatory virtues, pick the best, and check whether the winning explanation is actually good enough or merely the best of a bad lot. The practice in this lesson depends on understanding Explanatory Fit, Simplicity, and Best of a Bad Lot and applying tools such as Live Rivals Required and Fit the Evidence correctly.
How to approach it
Read the explanation sections first, then use the activities to test whether you can apply the idea under pressure.
What the practice is building
You will put the explanation to work through comparison exercise and quiz activities, so the goal is not just to recognize the idea but to use it under your own control.
What success should let you do
Run the full best-explanation pipeline on at least 3 mixed cases, producing candidate list, virtue scoring, winner, and best-of-a-bad-lot check.
Reading Path
Move through the lesson in this order
The page is designed to teach before it tests. Use this sequence to keep the reading, examples, and practice in the right relationship.
Study
Watch the move in context
Use the worked example to see how the reasoning behaves when someone else performs it carefully.
Do
Practice with a standard
Only then move into the activities, using the pause-and-check prompts as a final checkpoint before you submit.
Guided Explanation
Read this before you try the activity
These sections give the learner a usable mental model first, so the practice feels like application rather than guesswork.
Framing
Running the unit pipeline end-to-end
Earlier lessons taught the parts in isolation: what an explanation is, how to formalize an abductive inference, how to apply explanatory virtues, and how to critique best-explanation arguments. The capstone asks you to combine them on mixed cases.
Real best-explanation reasoning is never just 'pick the explanation that fits'. It is a disciplined comparison: list rivals, assess each against the virtues, pick the best, and — crucially — check whether the best is actually good enough or merely the best of a weak set.
What to look for
- Generate at least two rivals before scoring.
- Score by virtues, not by feel.
- Always perform the best-of-a-bad-lot check.
Best-explanation reasoning is a disciplined comparison, not a vote for the first idea.
Strategy
Choose the move that matches the case
Use a fixed pattern: (1) list at least two rival explanations, (2) score each against explanatory virtues (scope, fit, simplicity, coherence with background knowledge), (3) pick the best, and (4) ask whether the winner is actually well-supported or whether the candidate set was simply thin.
The best-of-a-bad-lot check is the move most often skipped. Even when one explanation clearly beats the others, you should pause to ask whether any of them actually meets a reasonable threshold. If none does, the right answer is to widen the candidate set, not to accept the winner.
What to look for
- List scope, fit, simplicity, and coherence for each rival.
- Pick the winner and state the margin.
- Ask whether the winner crosses a reasonable threshold at all.
Explicit virtue scoring makes the comparison defensible to someone who was not in the room.
Error patterns
How integration failures actually look
The commonest failure is treating the first explanation that comes to mind as the only one. Generating rivals takes conscious effort; without that effort, your 'best explanation' is whatever explanation you thought of first.
The second commonest failure is scoring by intuition rather than by virtues. Saying 'explanation A feels right' is not a diagnosis. Saying 'explanation A has wider scope and is simpler than explanation B, though B has slightly better fit to data point X' is.
What to look for
- Do not settle for the first explanation that comes to mind.
- Do not score by intuition.
- Do not skip the best-of-a-bad-lot check.
The fastest way to a bad abduction is to stop at the first candidate; the fastest way to a good one is to generate rivals on purpose.
Before practice
What this lesson is testing
The cases below are mixed: some have a clear best explanation, some have two nearly tied explanations, and some are genuinely best-of-a-bad-lot situations. Part of the exercise is telling these apart.
A case is only complete when you have produced the candidate list, the virtue-based comparison, the winner, and the best-of-a-bad-lot check.
What to look for
- List rivals before scoring.
- Score each by virtues.
- Include a best-of-a-bad-lot check on every case.
The capstone measures whether you can run a disciplined abductive comparison under mixed conditions.
Core Ideas
The main concepts to keep in view
Use these as anchors while you read the example and draft your response. If the concepts blur together, the practice usually blurs too.
Explanatory Fit
How closely a hypothesis matches the specific features of the observations, as opposed to merely being consistent with them.
Why it matters: Two hypotheses may share scope but differ in how precisely they fit the details.
Simplicity
A virtue of a hypothesis that explains the observations without introducing unnecessary assumptions or entities.
Why it matters: All else equal, simpler hypotheses are preferred — but not at the cost of scope or fit.
Best of a Bad Lot
A concern that the 'best explanation' might still be poor if the real explanation was never among the considered candidates.
Why it matters: This warning keeps abductive reasoning honest — 'best in the comparison' is not the same as 'correct'.
Reference
Open these only when you need the extra structure
How the lesson is meant to unfold
Review
This step supports the lesson by moving from explanation toward application.
Guided Synthesis
This step supports the lesson by moving from explanation toward application.
Independent Synthesis
This step supports the lesson by moving from explanation toward application.
Reflection
This step supports the lesson by moving from explanation toward application.
Mastery Check
The final target tells you what successful understanding should enable you to do.
Reasoning tools and formal patterns
Rules and standards
These are the criteria the unit uses to judge whether your reasoning is actually sound.
Live Rivals Required
An argument to the best explanation should compare more than one plausible hypothesis.
Common failures
- Only one hypothesis is presented.
- Alternative explanations are ignored or mentioned only to be dismissed without analysis.
Fit the Evidence
The preferred explanation should account for the relevant evidence better than its rivals, covering more of the observations and fitting their specific features.
Common failures
- The preferred hypothesis leaves central observations unexplained.
- A rival hypothesis explains the data equally well or better but is not acknowledged.
No Certainty Jump
The conclusion should be framed as the best current explanation, not as deductive certainty.
Common failures
- Writing that the hypothesis is definitely true.
- Treating explanatory superiority as proof.
Widen the Candidate Set
When every candidate hypothesis seems weak, the responsible move is to widen the candidate set rather than pick the best of a bad lot.
Common failures
- Accepting a weak hypothesis merely because it is the best of those considered.
- Failing to look for additional candidate explanations.
Patterns
Use these when you need to turn a messy passage into a cleaner logical structure before evaluating it.
Argument to the Best Explanation
Input form
natural_language_argument
Output form
structured_explanatory_comparison
Steps
- List the observations that need explanation.
- List at least two plausible candidate hypotheses.
- Compare the hypotheses using explanatory virtues.
- Rank the explanations.
- State which explanation is currently best supported, and at what level of confidence.
Watch for
- Listing only one hypothesis.
- Skipping the comparison stage.
- Writing a conclusion stronger than the evidence supports.
Explanatory Virtue Matrix
Input form
list_of_hypotheses
Output form
virtue_comparison_table
Steps
- List the candidate hypotheses as rows.
- List the explanatory virtues (scope, fit, simplicity, coherence) as columns.
- Score each hypothesis on each virtue with a short justification.
- Identify which hypothesis leads overall and which virtues decide the contest.
- State the conclusion in proportion to the size of the lead.
Watch for
- Weighting only one virtue (usually simplicity or familiarity).
- Filling in scores without actual justification.
Worked Through
Examples that model the standard before you try it
Do not skim these. A worked example earns its place when you can point to the exact move it is modeling and the mistake it is trying to prevent.
Worked Example
Virtue-Scoring Walkthrough
A virtue-based comparison makes the abductive choice defensible even when the first-mentioned candidate is not the winner.
Winner
Missing local file is the strongest single-explanation winner because the failure is consistent, not intermittent.
Passage
A program's test suite passes on the developer's laptop but fails on the CI server.
Virtue Scores
Race Condition
High scope; moderate fit (usually intermittent rather than consistent); lower simplicity.
Missing Local File
Narrower scope; very high simplicity; fits if the failure is reproducible.
Environment Difference
High scope, good fit, moderate simplicity; matches the 'works on my machine' pattern.
Candidate Explanations
- A difference in environment (library versions, OS, shell settings) between the two machines.
- A race condition that appears only under CI's parallel execution.
- A broken test that depends on a local file the developer has and CI does not.
Best Of A Bad Lot Check
The winner crosses a reasonable threshold: it is testable by listing the test's dependencies. If that test fails, widen candidates.
Pause and Check
Questions to use before you move into practice
Self-check questions
- Did I generate at least two rivals before picking a winner?
- Did I score each rival against specific virtues?
- Did I perform the best-of-a-bad-lot check?
Practice
Now apply the idea yourself
Move into practice only after you can name the standard you are using and the structure you are trying to preserve or evaluate.
Comparison Exercise
Best ExplanationFull-Cycle Best-Explanation Comparison
For each case, produce: (1) at least two candidate explanations, (2) a virtue-based comparison (scope, fit, simplicity, coherence), (3) a winner with the margin explained, and (4) a best-of-a-bad-lot check.
Integrative cases
Work one case at a time. These cases are deliberately mixed; part of the exercise is deciding which moves from the unit each case requires.
Case A
A company's sales dropped 30 percent last quarter. Three things changed that quarter: a competitor launched a similar product, the company's website was down for several days, and a major customer renegotiated terms.
Which single change best explains the drop, or does the best explanation involve multiple factors?
Case B
A patient has three symptoms: fatigue, joint pain, and low-grade fever, lasting several weeks. The lab results are borderline for two possible conditions.
When is the best explanation still not strong enough to act on without more evidence?
Case C
A research team finds that their new catalyst produces a 12 percent yield improvement. A rival lab cannot reproduce the result.
Candidate explanations include experimenter error, uncontrolled variables, and genuine effect. Compare them.
Case D
Traffic on a highway is suddenly lighter than usual at rush hour. No accidents have been reported and the weather is clear.
Generate three candidate explanations and score them before picking a winner.
Use one of the cases above, compare the competing explanations, and defend the one that best fits the evidence.
Quiz
Best ExplanationCapstone Check Questions
Answer each short check question in one or two sentences. These questions test whether you can articulate the reasoning you just performed in your own words.
Check questions
Answer each question from memory in your own words. No answer should need more than two sentences.
Question 1
Why is generating rivals a prerequisite for good abductive reasoning?
A best explanation is only meaningful relative to a set of alternatives.
Question 2
What is the best-of-a-bad-lot check, and when does it change the verdict?
It rejects even the winning explanation if none of the rivals meets a reasonable threshold.
Question 3
Why is scoring by explanatory virtues more reliable than scoring by intuition?
Virtues are explicit and checkable.
Question 4
When should an abductive verdict be expressed as tentative rather than confident?
When the margin between winner and rivals is small or the winner is best of a weak lot.
Use one of the cases above, compare the competing explanations, and defend the one that best fits the evidence.
Argument Mapper
Build an argument diagram by adding premises, sub-conclusions, and a conclusion. Link nodes to show which claims support which.
Add nodes above, or load a template to get started. Each node represents a proposition in your argument.
■ Premise■ Sub-conclusion■ Conclusion
Animated Explainers
Step-by-step visual walkthroughs of key concepts. Click to start.
Read the explanation carefully before jumping to activities!
Further Support
Open these only if you need extra help or context
Mistakes to avoid before submitting
- Mistaking a vivid explanation for a correct one.
- Stopping the comparison as soon as one rival pulls ahead.
Where students usually go wrong
Settling for the first explanation that comes to mind.
Scoring by intuition rather than by named virtues.
Skipping the best-of-a-bad-lot check.
Treating a winning explanation as confirmed without further testing.
Historical context for this way of reasoning
Peter Lipton
Lipton argued that 'inference to the best explanation' is a disciplined comparison of rivals, not a vote for the first plausible idea. The capstone puts that discipline into practice.