dialysis-prism

Dialysis Prism

R — Original Text (Reading)

"Problem" is: the gap between expectations and the current situation. Suppose the expected value of something is B and the current situation is B′, then this gap (B′→B) is the problem. ... To solve this problem, you cannot stare at B′→B; instead, you must look through B′ to see ABC, which I call the "Dialysis Prism." ... Step one, calibrate goal B: the goal must conform to the SMART principles, and at the same time avoid treating means as the goal. Step two, reconstruct method A: to change the current situation, do not start from the current situation and add a new solution; instead, go back and reconstruct the original method system. Step three, eliminate variable C: find it through the basic framework of "phenomenon, data, reason," and use the 5 Whys questioning method to dig out the true cause. — Chapter 17, "Dialysis Prism"

I — Methodological Skeleton (Interpretation)

Definition of problem:Problem = expectation (B) - current situation (B'). If you cannot find the gap, you cannot find the problem; "no problem" often means you know neither the goal nor the current situation
The surface problem is symptom B'→B; the root cause is hidden in the three factors ABC: A (method), B (goal), C (variable)
Prism B (calibrate goal): first check whether the goal itself is correct. Common errors — vague goals (such as "happiness," which cannot be measured), treating means as the goal (such as "earning 1 million," which may be a means rather than a goal)
Prism A (reconstruct method): the current situation B' is caused by the original method A. Treating symptoms is finding another new path from B' to B; treating the root cause is reconstructing the original method system itself
Prism C (eliminate variables): if both A and B are fine but B' still exists, there must be a hidden variable C. Use the "phenomenon, data, reason" framework to find it: phenomenon → data → reason. Use 5 Whys to ask continuously and dig out the root cause
Key principle: do not give advice too early — "we love giving advice too much, yet often have not figured out what the problem actually is"
The essential difference in methodology: the Self-Strengthening Movement was patching the old system (treating symptoms), while the Meiji Restoration was reconstructing the entire system (treating the root cause)

A1 — Application in the Book (Past Application)

Self-Strengthening Movement vs. Meiji Restoration: the Qing government, under the premise of keeping the feudal system unchanged, only learned Western technology (patching the old method A), and ultimately failed; Japan reformed both the system and technology at the same time (reconstructing the entire method A), and in one move joined the ranks of the great powers — the same problem, a world of difference between treating symptoms and treating the root cause
Factory capacity decline during the World Cup: the phenomenon is a decline in capacity (phenomenon), the data is that the night-shift defect rate soared by 3 times (data), and using 5 Whys revealed: workers stayed up late watching football → poor mental state → operational errors → increased defects (reason). The improvement action landed on "things you can change yourself": prohibit bringing mobile phones into the factory
Analysis of rising complaint rates: do not use subjective expressions like "I feel there are more complaints," but instead use data — "last month 2‰ vs. this month 6.7‰, a 3-fold increase," shifting from perceptual cognition to rational analysis
The error of taking "happiness" as a goal: Happiness cannot be measured → no gap can be found → you don't even know what the problem is → you cannot begin to solve it

A2 — Trigger Scenarios (Future Trigger)

When to use:

The same problem keeps recurring; after each "solution," it resurfaces after a while
You've tried multiple solutions with no effect and are starting to suspect the problem definition itself may be wrong
You need to do root cause analysis (retrospectives, incident investigations, performance improvement)
When making an improvement plan, you want to make sure you're grasping the essence rather than the surface
Team discussions fall into a "treat the symptom, not the root cause" loop

Language signals:

"Why has this problem come up again?"
"Why can it never be solved?"
"I've tried so many methods and none of them worked"
"What exactly is the root of the problem?"
"Where exactly is the problem?"
"Are we just treating the symptoms and not the root cause the whole time?"

Distinguish from adjacent skills:

vs. brain-unseal: Brain unsealing isa prerequisite step for restoring the brain's ability to function, while the Dialysis Prism is aproblem analysis toolused after the brain is running. Order: first unseal → then use the Prism
vs. structured-thinking: Structured thinking isa tool for breaking down and categorizing, while the Dialysis Prism isa tool for penetrating the surface to find the root cause. You can first use structuring to break out all the elements, then use the Prism to investigate them one by one
vs. mirror-world: The Prism applies todeterministic problems(where ABC can be found), while the mirror world applies touncertain problems(where ABC itself is uncontrollable)

E — Executable Steps (Execution)

Step 1: Calibrate goal B — first confirm whether the problem definition is correct

Write down the problem you are currently facing, expressed in the format "B - B' = problem": What is the expectation (B)? What is the current state (B')? Where is the gap?
Check whether goal B conforms to the SMART principle: Specific (S), Measurable (M), Achievable (A), Relevant (R), Time-bound (T)
Check whether you have mistaken a "means" for a "goal": ask yourself, "What will I get after achieving B?" If that answer is the real goal, then replace B
Completion criteria: Goal B can be expressed with a specific number or a clear verifiable state, and is not an intermediate means but the ultimate end

Step 2: Reconstruct method A — trace back the existing system rather than patching it

Sort out theoriginal method Athat led to the current state B' and what links it includes (people, processes, tools, rules)
Do not start from B' and add a new solution; instead return to A itself and ask, "Which link in the original method caused B' to deviate from B?"
Perform asystemic reconstructionof that link rather than incremental patching: replace rather than add on
Completion criteria: You can draw a comparison diagram of "method A before reconstruction" and "method A' after reconstruction," and A' is not a patch layered onto A but changes one core link

Step 3: Eliminate variable C — use image-number-principle + 5 Whys to find hidden variables

If the problem still persists after refactoring A, start variable investigation:
- Xiang(Phenomenon): What specifically observed abnormal manifestations are there? List all abnormal phenomena
- Shu(Data): Find quantitative data to replace "I feel"—what is the data? How large is the magnitude of change?
- Li(Reason): What is the underlying reason for the data change? Use 5 Whys to ask continuously for at least 5 layers until you findwhat you can change's root cause
Land improvement actions on "things you yourself can change," not external uncontrollable factors
Completion standard: Produce a data-driven problem analysis report (phenomenon → data → 5 Whys chain → root cause → executable improvement actions), and the improvement actions are 100% within your own controllable scope

B — Boundary

Scenarios not to use:

A completely new field with no historical method A to trace back to—no old method to refactor, requiring design from scratch
Purely uncertain events (stock market crash, sudden disaster)—the ABC model does not apply; switch to mirror-world's four-domain strategy
Emergencies requiring immediate action—no time for in-depth analysis; stop the bleeding first, then review
The problem itself is simple enough that the cause can be seen at a glance—over-analysis wastes time

Failure modes:

Skipping B and looking directly for A: the goal itself is wrong, and all subsequent analysis spins idle on an incorrect foundation
Patching A instead of refactoring it: adding a rule/process every time a problem occurs, making the system increasingly bloated while the problem still persists
The 5 Whys stop at the first layer: for example, "production capacity decline → worker error → done," without asking why the worker made an error
Finding the root cause but it is uncontrollable: for example, "because the World Cup is being held"—the analysis is correct but cannot lead to action, equivalent to analyzing in vain
Using "I feel" instead of data: without quantitative analysis, the prism degrades into guesswork

Author blind spots:

Lack of priority guidance on "how to determine whether to start investigating from B, A, or C"; in practice all three directions are possible, requiring substantial time and resources to investigate one by one
The 5 Whys method depends on the questioner's experience; different people may ask completely different causal chains for the same phenomenon—there is no calibration mechanism
"Refactoring method A" in practice may involve complex factors such as organizational restructuring and redistribution of interests, far more difficult than described in the book
The Xiang-Shu-Li framework comes from the I Ching but is simplified into "phenomenon → data → reason"; in actual analysis, determining the "reason" often requires support from domain expertise

Related skills

This skill is related to the following skills:

structured-thinking(Prerequisite dependency): The analysis process of the dialysis prism (calibrating B, refactoring A, eliminating C) requires structured thinking as a foundational tool; first use the MECE principle to break down all problem elements, then use the prism to investigate root causes one by one.