reality-anchor

name

reality-anchor

description

Test a theoretical conclusion against reality using 6 verification methods from Zakery Kline's 'How to Think' Chapter 4. Use when someone says 'is this actually true', 'reality-check this idea', 'test my theory', 'am I right about this', 'does this hold up', 'verify my conclusion', 'is this grounded in reality', or 'I think X is true — prove me wrong.' Runs 6 reality tests and scores the conclusion.

Reality Anchor

Test any theoretical conclusion against reality using 6 verification methods from Chapter 4 of Zakery Kline's How to Think. The core insight: even perfectly coherent reasoning can lead to false conclusions if it fails to correspond with reality. Logical consistency is necessary but not sufficient — you need correspondence too.

When to Use

The user has arrived at a conclusion through reasoning, reading, or intuition and wants to know whether it actually holds up. They might say:

• "I've been thinking about X and I believe Y — is that right?"
• "Reality-check this theory for me"
• "This idea makes sense logically but something feels off"
• "I'm confident about this conclusion — poke holes in it"
• "Does this hold up outside my bubble?"

This is NOT for evaluating business ideas (use idea-vetter) or predictions (use prediction-scorecard). This is for testing whether a belief, theory, or conclusion corresponds with reality.

How to Run This

Step 1: Get the Conclusion

Ask: "What's the conclusion or theory you want to reality-test? State it as clearly as you can — one sentence if possible."

Then ask: "How did you arrive at this? Was it through reading, reasoning, personal experience, intuition, or something someone told you?"

Understanding the origin matters because different origins have different failure modes. A conclusion from pure reasoning needs observational testing. A conclusion from personal experience needs cross-contextual testing. A conclusion from authority needs independent verification.

---

The 6 Reality Tests

Run each test in order. For each one, discuss the findings with the user before scoring. Don't rush — the conversation IS the verification.

Test 1: Direct Observational Verification

The question: What specific, observable predictions does this conclusion make — and can they be checked?

If a conclusion is true, it should predict observable phenomena. If it predicts nothing observable, it may be unfalsifiable (and therefore not a claim about reality at all).

Walk through:

• "If your conclusion is true, what should we be able to observe in the world?"
• "Is there a specific situation where we could watch this play out?"
• "What would we expect to see that we wouldn't see if the conclusion were false?"

Kline's principle: A conclusion that makes no observable predictions isn't wrong — it's empty. It lives entirely in the realm of abstraction and says nothing about the world you actually inhabit.

Score:

• Pass — Makes specific, checkable predictions that match observation
• Partial — Makes predictions, but they're vague or only partially confirmed
• Fail — Predictions contradict what we observe
• N/A — The conclusion is definitional or mathematical (not an empirical claim)

Test 2: Historical Verification

The question: Does the historical record support or contradict this conclusion?

Most ideas that seem new have been tried, believed, or debated before. History is the largest dataset we have. Ignoring it is intellectual malpractice.

Walk through:

• "Has this idea been held before? By whom? What happened?"
• "Are there historical periods or societies that operated on this assumption? How did it go?"
• "What do primary sources — not summaries or popular accounts — actually show?"

Kline's principle: Beware of conclusions that require you to believe everyone before you was stupid. If your theory implies that centuries of practitioners, thinkers, or cultures all missed something obvious, the most likely explanation is that you're missing something, not them.

Score:

• Pass — Historical evidence consistently supports the conclusion
• Partial — Mixed historical record; some support, some contradiction
• Fail — Historical evidence strongly contradicts the conclusion
• N/A — The conclusion concerns something genuinely unprecedented

Test 3: Experiential Verification

The question: If this conclusion is true, what should you personally experience — and does that match?

This is the first-person test. Not "what does the data say" but "what does your own lived experience show." It's the weakest form of evidence in isolation but a powerful red flag when it contradicts the conclusion.

Walk through:

• "If this is true, what should your day-to-day experience feel like?"
• "Does your actual experience confirm or contradict this?"
• "Are you relying on other people's reported experience rather than your own? If so, why?"

Kline's principle: When your personal experience flatly contradicts a theory, you have two honest options: investigate whether your experience is atypical, or investigate whether the theory is wrong. What you cannot honestly do is dismiss your own experience because a theory told you to.

Score:

• Pass — Personal experience consistently aligns with the conclusion
• Partial — Experience partially aligns; some aspects don't match
• Fail — Personal experience directly contradicts the conclusion
• N/A — The conclusion is about domains outside personal experience

Test 4: Pragmatic Test

The question: When people act on this conclusion, what happens? Does it "work"?

Truth has practical consequences. If a conclusion is true, acting on it should produce predictable, functional results. If acting on it consistently produces confusion, failure, or incoherence, something is wrong with the conclusion — regardless of how elegant the reasoning is.

Walk through:

• "Who has actually acted on this belief? What were the results?"
• "If you started living as though this were true tomorrow, what would change?"
• "Does this conclusion make people who hold it more effective, or less?"

Kline's principle: A conclusion that is theoretically perfect but practically useless is not knowledge — it is decoration. And a conclusion that consistently produces dysfunction when applied is, at minimum, incomplete.

Score:

• Pass — Acting on this conclusion reliably produces the expected results
• Partial — Works in some contexts but fails or backfires in others
• Fail — Acting on this conclusion consistently produces bad outcomes
• N/A — The conclusion is purely descriptive (no action implications)

Test 5: Cross-Contextual Verification

The question: Does this conclusion hold across different cultures, time periods, and domains — or is it parochial?

"Truly robust conclusions should apply consistently across different contexts."

A conclusion that only works in one culture, one era, or one domain is not necessarily wrong — but its scope is narrower than it appears.

Walk through:

• "Does this hold true in cultures very different from yours?"
• "Would this have been true 500 years ago? Will it be true 500 years from now?"
• "Does this principle apply in adjacent domains, or only in the specific one you're thinking about?"
• "Where does it break? What's the boundary condition?"

Kline's principle: Cross-contextual consistency is the strongest non-experimental evidence for truth. When the same conclusion emerges independently across multiple cultures, eras, and disciplines, the odds that it's mere coincidence drop dramatically.

Score:

• Pass — Holds consistently across cultures, eras, and domains
• Partial — Holds in many contexts but breaks in identifiable ones
• Fail — Only holds in a narrow context; breaks when exported
• N/A — The conclusion is explicitly context-specific by design

Test 6: Explanatory Power

The question: Does this conclusion explain observed phenomena BETTER than competing explanations?

"When multiple independent paths lead to the same destination, we gain greater confidence in that destination."

A conclusion doesn't exist in a vacuum. It competes with other explanations. The question isn't just "does this work?" but "does this work better than the alternatives?"

Walk through:

• "What are the competing explanations for the same phenomena?"
• "What does your conclusion explain that the alternatives can't?"
• "What do the alternatives explain that your conclusion can't?"
• "Is your conclusion simpler, or does it require more assumptions?"

Kline's principle: Prefer the explanation that accounts for the most phenomena with the fewest assumptions — but not at the cost of ignoring phenomena that don't fit. An explanation that's simple because it ignores inconvenient data is not parsimonious; it's incomplete.

Score:

• Pass — Explains more phenomena with fewer assumptions than alternatives
• Partial — Roughly equal explanatory power to alternatives
• Fail — Alternatives explain the phenomena better
• N/A — No competing explanations exist (novel domain)

---

Output: Reality Anchor Report

After running all six tests, deliver the report.

Scoring Table

Test	Score	Key Finding
1. Direct Observational	[Pass/Partial/Fail/N/A]	[One-line summary]
2. Historical	[Pass/Partial/Fail/N/A]	[One-line summary]
3. Experiential	[Pass/Partial/Fail/N/A]	[One-line summary]
4. Pragmatic	[Pass/Partial/Fail/N/A]	[One-line summary]
5. Cross-Contextual	[Pass/Partial/Fail/N/A]	[One-line summary]
6. Explanatory Power	[Pass/Partial/Fail/N/A]	[One-line summary]

Boundary Map

Where the conclusion applies and where it doesn't. Every conclusion has boundaries — even true ones. The goal is not to prove the conclusion universally true but to map its actual territory.

• Applies in: [contexts, domains, conditions where the conclusion holds]
• Breaks in: [contexts, domains, conditions where it fails or needs modification]
• Untested in: [contexts where we can't yet evaluate it]

Overall Verdict

Verdict	Meaning
Grounded	Passes 4+ tests with no Fails. The conclusion corresponds with reality across multiple verification methods.
Partially Grounded	Passes 2-3 tests, or passes more but has 1+ Fail. The conclusion has real support but also real gaps. Refine the boundaries.
Ungrounded	Fails 3+ tests. The conclusion may be logically coherent but does not correspond with observable reality. Revise or abandon.
Untestable	4+ tests score N/A. The conclusion may be true but cannot be verified by these methods. Acknowledge this honestly rather than pretending certainty.

Closing

After delivering the verdict, ask: "Based on this analysis, do you want to refine the conclusion to fit the boundaries we found, or do you want to stress-test a competing explanation through the same six filters?"

The goal is not to win an argument but to hold conclusions that actually correspond with reality — even when that means letting go of elegant ideas that don't survive contact with the world.