Epistemology: Knowledge Types
Before you can test whether a claim is justified, you need to know what kind of claim it is — because different kinds of knowing have different justification standards, different failure modes, and different evidentiary requirements. Mixing up knowledge types is a recurring source of bad reasoning: treating empirical claims as if they're matters of pure logic, treating intuitions as if they're perceptions, treating testimony as if it's first-hand knowledge.
This skill classifies the kind of knowing in play, then draws out what that classification implies for how the claim can be established or challenged.
Your Process
Step 1: Extract the Claim State the claim being made as precisely as possible. Strip away rhetorical packaging. What exactly is being asserted?
Framing check: Confirm the specific claim before continuing. State what you've identified — the actual claim being analyzed and the context in which it is being made — in one sentence, then use AskUserQuestion:
- Question: "I'm reading this as: [your one-sentence framing of the specific claim and context]. Is that right?"
- Header: "Framing"
- Options:
- Yes — proceed — framing is correct
- Adjust — one element is off; user will correct it before you continue
- Reframe — different situation than read; incorporate the correction before proceeding
Step 2: Classify Along the Primary Axis — A Priori vs. A Posteriori
-
A priori — can be known through reason alone, independent of experience. True by definition, logical necessity, or mathematical proof. Examples: "all bachelors are unmarried," "2+2=4," "if A>B and B>C then A>C."
- Test: could this be false if the world were different? If no, it's a priori.
- Failure mode: confusing definitional truths with empirical claims ("free markets are efficient" can be made a priori by definition, but that makes it empty rather than powerful).
-
A posteriori — requires experience and evidence. Could be false; the world could have been otherwise. Examples: "water boils at 100°C at sea level," "this product's NPS is 42."
- Test: would we need to investigate the world to know if it's true? If yes, it's a posteriori.
- Failure mode: treating empirical claims as if they've been established when they've only been assumed.
Step 3: Classify the Form of Knowing
-
Propositional (knowing that) — a fact or state of affairs. "The conversion rate dropped." "Keynes believed X." Most analytical claims are propositional. Can be true or false; can be supported by evidence; can be transmitted by testimony.
-
Procedural (knowing how) — a skill or capacity. "Knowing how to ride a bicycle." "Knowing how to negotiate." Cannot be fully reduced to propositions — you can know all the facts about cycling and still not know how to ride. Failure mode: confusing explanation (propositional) with competence (procedural). Particularly relevant in organizations: knowing about a process is not the same as being able to execute it.
-
Acquaintance (knowing of) — direct familiarity through experience. "Knowing Paris." "Knowing what grief feels like." Richer than testimony but harder to transmit. Failure mode: assuming shared acquaintance when the other person only has propositional knowledge (talking about a difficult customer as if the other person knows what it's like to manage them).
Step 4: Classify the Source
| Source | What it provides | Key vulnerability |
|---|---|---|
| Perception | Direct sensory evidence | Observation errors, selection effects, limited perspective |
| Inference | Conclusions drawn from other knowledge | Depends on validity of chain and truth of premises |
| Testimony | Knowledge passed from others | Reliability of the source, transmission errors, motivated reporting |
| Intuition | Fast, pattern-based judgment | Hard to articulate; may encode bias; often reliable in expert domains |
| Memory | Retained past experience | Reconstructive, not reproductive; degrades; subject to post-hoc editing |
Step 5: Assess the Epistemic Implications
Given what kind of knowing is in play:
- What evidence is actually relevant to this claim?
- What evidence is irrelevant but being treated as relevant?
- What can this kind of knowing establish — and what does it categorically not establish?
- Is the person making the claim aware of the limits of their knowledge type?
Step 6: Identify the Type Confusion (if any)
The most common errors:
- Treating a posteriori claims as if they were a priori — "obviously," "of course," or "any reasonable person would agree" applied to empirical claims.
- Treating testimony as if it were perception — "I heard that..." treated with the same weight as "I saw that..."
- Treating procedural knowledge as if it were propositional — assuming that explaining a skill is equivalent to conferring it.
- Treating intuitions as evidence — strong feeling that X used as if it were observation of X.
Human Check-in
Before proceeding, use the AskUserQuestion tool. State your interpretation of the situation in 1–2 sentences — what is being analyzed and what the core question is — then ask:
- Question: "My read: [your 1–2 sentence interpretation]. How do you want to proceed?"
- Header: "Scope"
- Options:
- Full classification — Complete all steps with the epistemic implications drawn out
- Type and implications only — Classify quickly and flag what follows, skip full source analysis
- Identify the confusion — Focus on what knowledge-type error is being made, skip full taxonomy
- Refine the claim — Sharpen what we're analyzing before starting
Proceed based on their selection. If the user reframes, incorporate the correction before running any analysis.
Output Format
The Claim
[Exact claim being analyzed]
Primary Classification
A priori / A posteriori: [Classification + one sentence justification]
Form of Knowing
Type: Propositional / Procedural / Acquaintance Notes: [Why this classification matters here]
Source Analysis
| Source in Play | Weight Given | Assessment |
|---|---|---|
| [Perception / Inference / Testimony / Intuition / Memory] | High / Medium / Low | [Is this appropriate? Any vulnerabilities?] |
Epistemic Implications
What this kind of knowing can establish:
- [Point 1]
- [Point 2]
What it cannot establish:
- [Limit 1]
- [Limit 2]
Type Confusion (if any)
Confusion identified: [Specific error] Effect on the argument: [What this means for the claim's validity]
Upshot
[One clear paragraph: what is the person actually dealing with here, and what follows for how they should reason about or act on this claim?]
Notes
Use epistemology-justification when you've classified the knowledge type and now want to test whether the belief is actually justified. Use epistemology-epistemic-status when you need to calibrate confidence across a whole domain rather than analyze one claim. Use logic-check when the issue is inference validity rather than knowledge-type misclassification.
What's Next
After delivering this output, use AskUserQuestion to offer the next move:
- Question: "Knowledge types classified. What's next?"
- Header: "Next"
- Options:
/s4h-epistemology-epistemic-status— Assign epistemic status to each knowledge type/s4h-decision-criteria-weighting— Weight decision criteria by the type and reliability of knowledge/s4h-investigation-evidence-audit— Audit evidence quality for each knowledge type- Done — Wrap up and synthesise what we have so far
