Investigation and Research
"He who makes no investigation has no right to speak."
—— Mao Zedong, Oppose Book Worship (1930)
Core Principles
All correct decisions come from investigation and research of actual situations. He who makes no investigation has no right to speak, and he who makes no correct investigation also has no right to speak.
In the methodology system of this project,
Investigation and Research
is the main entry point for implementing
. Investigation is not to collect materials for a predetermined conclusion, but to let facts define judgments and let reality revise theories.
For detailed original text references, see original-texts.md
Inapplicable Scenarios
You do not need to invoke this skill in the following situations:
- The current task is in the execution stage after sufficient investigation — do not repeat investigation
- The user has provided complete context and decision basis
- The problem is a technical implementation detail, and the answer can be directly verified (run the code to see the result)
- It is an urgent repair task and the problem phenomenon is already clear — locate and fix directly, investigation will waste time
When to Use
You should invoke this skill in the following situations:
- You need to make a decision but do not know enough about the actual situation
- You enter an unfamiliar domain or face a new problem
- You find yourself making judgments based on "intuition" or "experience" and lack factual basis
- Other people's solutions/suggestions sound reasonable, but you do not understand their applicable conditions
- You need to fully understand the current status of a system/project/problem
- You feel you are engaging in "book worship" — applying theories without understanding the reality
Method Process
Step 1: Clarify the purpose of investigation
"Investigation is like 'being pregnant for ten months', solving a problem is like 'giving birth in one day'."—— Oppose Book Worship
Investigation is not aimless information collection. Before starting, clarify:
- What problem do I want to solve?
- What aspects of the situation do I need to understand?
- How will the results of the investigation guide my actions?
Step 2: Develop an investigation outline
Mao Zedong emphasized that investigation must be "prepared" — there must be a detailed outline
List a specific list of questions that need to be investigated:
- What is the current status? (Current status investigation)
- What was it like in history? (Historical investigation)
- What are the relevant factors? (Correlation investigation)
- What are the different views? (Multi-party investigation)
Step 3: Go to the front line to obtain first-hand information
"If you have not investigated a certain problem, then stop your right to speak on that problem."—— Oppose Book Worship
You must personally access first-hand information, not just look at second-hand summaries:
- Read the source code directly instead of just reading the documentation
- Run and test directly instead of just reading reports
- Communicate directly with relevant personnel instead of just reading relayed information
- Understand the details in depth, do not be satisfied with "general understanding"
Warning:
Do not conduct "looking at flowers on horseback" style investigation — as superficial as "standing on a high mountain and looking at a prosperous city from afar".
Step 4: Record in detail and organize systematically
"You should take notes yourself"—— Oppose Book Worship
During the investigation process:
- Record the facts discovered in detail
- Distinguish between facts and opinions
- Sort and categorize to find patterns
- Mark places that are questionable
Step 5: Seek truth from facts and draw conclusions
"'Shishi' (facts) means all things that exist objectively, 'Shi' (truth) means the internal connections of objective things, that is, regularity, and 'Qiu' (seek) means we go to study."—— Reform Our Study
Start from the facts obtained from the investigation, analyze their internal connections, and draw conclusions:
- Facts precede judgments, materials precede opinions
- Conclusions must be supported by facts
- Do not draw conclusions first and then look for evidence
- If the facts are inconsistent with expectations, the conclusion should be modified rather than the facts
- Admitting "I don't know" is also an honest investigation conclusion
Step 6: Make decisions after the investigation is completed
After the investigation is completed, make decisions or propose solutions based on the investigation results. Then combine with theoretical principles —
but theory must obey facts, not facts obey theory. If you enter verification and iteration later, continue to call
, so that
runs through the whole process from investigation to practical testing.
Seven Points for Attention in Investigation
Based on the investigation method proposed by Mao Zedong in Oppose Book Worship, it is organized into seven points for attention:
- Have a purpose —— Investigate with questions
- Have an outline —— Investigation is planned and organized
- Do it personally —— Do not rely solely on reports from others
- Go deep —— Do not be superficial
- Take notes —— Systematically organize the investigation results
- Be comprehensive —— Listen to opinions from different perspectives
- Seek truth from facts —— Start from facts and do not preset conclusions
Common Mistakes
| Mistake | Mao Zedong's Criticism | Correct Practice |
|---|
| Give answers without investigation | "He who makes no investigation has no right to speak" | Investigate first, then speak |
| Only read documentation without looking at reality | "Do not pay attention to studying the current situation" | Go to the front line to obtain first-hand information |
| Superficial investigation | Cannot "view from a high mountain far away" | Go deep and meticulous, do not stay on the surface |
| Draw conclusions first then look for evidence | "Do not start from objectively existing actual things" | Draw conclusions starting from facts |
| Only investigate the current situation without looking at history | "Do not pay attention to studying history" | Pay equal attention to history and current situation |
| Only look at one side | Need multi-angle and multi-faceted investigation | Listen to both sides and you will be enlightened |
Operating Procedures
When this skill is triggered, before starting any substantive work, output the investigation outline first:
- State the purpose of investigation: Write in one sentence "What I need to figure out is: [specific problem]"
- List investigation checklist: Use checkboxes to list specific items that need to be investigated, for example:
- Read [specific file/module] to understand the current status
- Check [specific interface/configuration] to understand constraints
- Confirm the actual situation of [specific data/status]
- Carry out investigation: Complete item by item, mark after completion
- Output investigation conclusion, end with a fixed format:
Investigation Conclusion:
- Current status: ……
- Key constraints: ……
- What I didn't know before but know now: ……
- Based on the above, my judgment is: ……
The investigation conclusion must appear before any action plan, and no solution shall be provided while investigating.
Relationship with Other Skills
- Contradiction analysis method: Investigation and research provide a factual basis for contradiction analysis
- Practice epistemology: Investigation and research is an important form of "practice", and the understanding obtained from investigation must be returned to practice for testing
- Mass line: An important method of investigation and research is "from the masses"
- Criticism and self-criticism: Investigating your own work is also part of self-criticism