This question touches on the core of the Fourteenth System of Causality-the nature of the causal chain and how logic works under this nature.
1. The nature of the causal chain: not linear, but network-like
A typical expression of linear causality is: A → B → C. Its implicit assumption is that the cause is single, the path is unique, and the result is certain. This view of causality applies to mechanical systems (such as pressing a switch and the light goes on), but it almost always fails for complex systems involving people, society, and life.
The causality of the fourteen systems is not linear, butnetwork-likeof. In network causality:
Doindoguo: One result is caused by multiple reasons, and one reason may also lead to multiple results.
feedback loop: B is not only generated by A, B will also affect A in turn (positive feedback or negative feedback).
nonlinear: Tiny inputs may produce huge outputs (butterfly effect), and huge inputs may also produce only small outputs (system saturation).
path dependence: The same cause, under different initial conditions, may lead to completely different effects.
emerged: New qualities at the overall level cannot be reduced to a single causal chain.
Therefore, in networked causality, there is no "clear, traceable, one-way" chain of causality-you cannot trace causality like you trace a line. What you can trace is a partial connection to a network.
2. In network-like causality, how does the logic of deduction work?
If cause and effect are network-like, can "logical deduction" still hold? Yes, but the way it works will change.
1. From "inevitable deduction" to "probability evaluation"
In linear causality, if you know A, you can deduce B (or with a high probability). In network-like causality, you don't know which path in the network A will trigger-A may trigger B, or it may trigger C, or it may trigger both B and C but their interaction produces D.
Therefore, logical deduction no longer produces "inevitable conclusions" but producesprobability distribution: What are the possible outcomes under given conditions? What are the respective probabilities?
This is not a weakening of logic, but a natural expansion of logic in complex systems. You can't be sure whether a particular causal chain will be activated, but you can identify which paths are more likely to be activated, which conditions are more likely to be triggered, and which feedback loops are more likely to be activated.
2. From "cause traceability" to "conditional identification"
In linear causality, the core of deduction is to "find the cause"-A occurred because there was X before it. In cyber-like causality,"looking for reasons" is often futile because there are too many and intertwined reasons. The occurrence of an event is not caused by a single cause, but is caused by the maturity of a set of conditions.
Therefore, the focus of logic starts fromTracing the chain of cause and effectsteeringidentification condition combination。The question is not "what caused it", but "under what conditions will it tend to happen."
3. Moving from "prediction results" to "recognition patterns"
In linear causality, the goal of logical deduction is to predict the outcome: knowing A and deriving B. In network-like causality, it is almost impossible to predict a single outcome (because the network is too complex and there are too many paths). But you can identify the system'sbehavior patternsWhat trends does the system exhibit under what conditions? In what direction does it tend to evolve? Where is its self-stabilizing mechanism?
This is not "prediction", but "pattern recognition". You are not predicting specific outcomes, you are identifying the current direction of evolution of the system-and then, after identification, decide where you are and what role you play in that direction, rather than determining where it will evolve before identification.
3. How does this logic work in the view of causality in the fourteen systems?
The 14th system's view of causality is a "non-deterministic view of origin"-it recognizes the laws of causality, but does not recognize linear determinism. The result is not a "pre-written script" but an "open probability cloud." Under this view of causality, the operating logic of deduction is as follows:
1. Conditional design takes precedence over outcome prediction
You can't predict the outcome, but you candesign conditionsto increase the probability of a favorable outcome. You invest resources in a certain direction, adjust the environment, introduce new variables-actions that change the shape of the probability cloud, but don't determine where it will end up.
The role of logic here is not to "deduce a definite result", but to "evaluate the change trend of the probability distribution after conditions change." You can't be sure where it will fall, but you can evaluate the areas of possible outcomes in all directions and find the most likely scenarios among them.
2. Feedback iteration replaces one-time deduction
In linear logic, you can draw a conclusion with one deduction. In network-like causality, the result of your deduction is itself a new input-it affects the system and changes the structure of the causal network. Therefore, logical deduction must be acycle: Assessment conditions → Take actions → Observe feedback → Update assessments → Adjust actions.
Each feedback helps you update the shape of the probability distribution and helps you check whether your previous assessment of the results is accurate. After confirmation, you can revise the prerequisites for the next deduction.
3. Identify hub nodes
In network-like causality, not all nodes are equally important. Some nodes are densely connected, and changing it can ripple affect a large number of paths. These nodes are the "hubs" of the network and your most valuable objects of concern. Part of the task of logical deduction isIdentify these hubs, rather than tracking all possible chains of cause and effect.
Once a hub node is identified, you can indirectly affect the entire network by changing the node without having to trace each path one by one. This structure-based deduction is more suitable for complex systems than path-based deduction, and is also closer to the way the fourteen systems understand the causal network.
4. Tolerance of incompleteness
Linear logic requires complete information to be deduced. The logical recognition of network-like causality: You can always only see part of the network, and there will always be unknown paths, unknown feedback, and unknown variables. Therefore, the deduction does not require completeness, only needs to be judged within the currently visible range.
This judgment may be the best judgment you could make at the time. If it turns out that it is not accurate enough afterwards, you can evaluate it again in the next feedback. You don't need to be accurate at once, you just need to revise it step by step after each feedback.
4. A core understanding: logic has not failed, but has only expanded the form
Linear logic "fails" in network-like causality not because the logic itself fails, but becauseApplicable conditions have changed。When you face a nonlinear, multi-feedback, emergent system, you can still use logic, but the form of logic changes: it no longer produces unique, definite conclusions, but an evaluation of probability distributions; it no longer traces a single cause, but identifies combinations of conditions; it no longer attempts to predict specific outcomes, but identifies the evolution of the system.
This logic is not "bad"-it just deducts from a more complex premise. If you can accept this premise, you will not doubt logic itself because you cannot find a definite conclusion, but will realize that definite results do not exist in the system, and logic can only tell you which directions are more likely to be closer to the state you want, not where it will stop. This in itself is a useful information. You can use it to make judgments, but your judgment needs to leave room for adjustment for its incompleteness-you can modify it later when you receive feedback.