Reason is a methodical, causal progression from reality to knowledge to action. It begins with what exists, not with wishes, and it moves by logic, not by feelings. Below is the full sequence, with each step’s function and its place in the hierarchy.
-
Choice to focus (the precondition)
You must choose to direct your mind to reality. This volitional act—sustained attention to facts—is the root of all subsequent cognition and the basic exercise of free will. Without focus, there is no reasoning. [1][2] -
Observation (perception as the base)
Percepts are the given; the senses are valid. You register entities, their attributes, actions, and relationships. No inference is drawn yet; you simply grasp what is there. [1][2] -
Isolation and description (attentional selection)
You isolate relevant units in the field of perception, identify distinguishing features, and name or ostensively point to them. You are preparing the material for abstraction. [2] -
Concept-formation (abstraction by essentials)
You differentiate and integrate the observed units, omitting measurements within a range to form a concept with a unit perspective (e.g., “length,” “metal,” “market”). This is how the mind condenses many concretes into one mental unit. [2][4] -
Definitions by essentials
You define each concept by genus and differentia, capturing its fundamental distinguishing characteristic(s) in the present context of knowledge. Definitions are objective and may be refined as context expands; referents do not change. [2][3] -
Propositional formulation (statement of facts)
You connect concepts in declarative form to identify facts: subject–predicate, cause–effect. Logic is the law of non-contradictory identification; you reject package-deals, equivocation, and stolen-concept fallacies. [3] -
Induction (generalization from cases to principles)
You move from observed concretes to universal principles by identifying causal connections that explain and necessitate the cases; you use experiment, controlled observation, and measurement to distinguish essentials from accidentals. [6][3] -
Deduction (implications from principles)
From validated principles, you derive implications for new cases, preserving logical necessity and checking for contradiction. Deduction without prior induction is groundless; induction without subsequent deduction is blind. [3] -
Reduction (validation back to the perceptual level)
You justify higher-level claims by tracing them stepwise back to first-hand observations; this enforces the primacy of existence and guards against floating abstractions. [3][2] -
Measurement and quantification
Where appropriate, you assign numbers to magnitudes, establish units, and relate quantities functionally; this tightens explanation and prediction. [6] -
Causal explanation (the “why”)
You integrate laws and mechanisms that account for observed regularities. Explanation is not a slogan; it is a demonstration of how an entity’s identity necessitates its actions. [6][3] -
Context-keeping and integration (the safeguard)
You integrate each new conclusion with the full context of your knowledge, updating definitions as needed and rejecting any claim that clashes with established facts. Knowledge is hierarchical and contextual; certainty is contextual. [2][3] -
Validation standards (evidence, burden, and the arbitrary)
You accept only that for which evidence exists; the burden of proof is on the asserter; the arbitrary—claims offered without evidence—is to be dismissed, not refuted. Degrees of support include possibility, probability, and certainty, all defined by evidence within context. [3][2] -
Error detection and correction
You actively test for contradictions, check premises, replicate observations, and refine methods. When errors appear, you identify their roots and correct them by reduction to facts. [3][6] -
Evaluation (fact–value integration)
With facts established, you appraise them by reference to an objective standard: the requirements of life proper to a rational being. Values are facts about what promotes or harms life; there is no dichotomy between the “is” and the “ought” when ethics is grounded in reality. [5] -
Value-judgment by principle (virtues)
You judge and choose by rational virtues—rationality, independence, integrity, honesty, justice, productiveness, pride—because they are principles of successful action in reality, not social conventions. [5] -
Deliberation: ends–means integration
You set goals, rank alternatives by causal efficacy and long-range consequences, and select means consistent with rights and the non-initiation of force. Short-range gains that undercut long-range life are rejected. [5][6] -
Decision and action
You act on your best rational judgment, implementing plans in reality. Action is the point of knowledge; knowledge that does not guide action is a pretense. [5] -
Feedback and continuous improvement
You observe outcomes, compare them to predictions and values, and iterate: update hypotheses, retune plans, and reintegrate the results into your knowledge hierarchy. Reason is a continuous, self-correcting process. [6][3]
Operational rules that apply at every step
- Primacy of existence: facts over feelings, reality over consensus. [1]
- Logic as method: identity, non-contradiction, excluded middle. [3]
- Objectivity: adhere to reality by method; keep context; integrate without contradiction. [2]
- Evidence first: observation and experiment under logical control; explanation in causal terms. [6]
- Moral sovereignty: no sacrifice of reason to emotion; no surrender of the mind to authority. [5]
Condensed checklist
- Focus → 2) Observe → 3) Isolate/describe → 4) Form concepts → 5) Define by essentials → 6) Form propositions → 7) Induce principles → 8) Deduce implications → 9) Reduce to facts → 10) Measure/quantify → 11) Explain causally → 12) Integrate contextually → 13) Validate (evidence/burden/arbitrary) → 14) Detect/correct error → 15) Evaluate by objective standard → 16) Judge by virtues → 17) Deliberate ends–means → 18) Decide/act → 19) Feedback/iterate. [1][2][3][5][6]
This is the complete process of reason: a disciplined chain from existence to evaluation to action, with every link tied back to reality and every step governed by logic. Evasion at any point breaks the chain. [1][3]
Sources
Here is additional, more technical guidance that fills in methods, safeguards, and working checklists for every stage—so you can move from raw observation to validated knowledge to rational action without gaps or floating abstractions.
- Architecture of a full reasoning chain
- Base: metaphysically given facts, grasped by perception; the senses are valid and are the cognitive starting point. [1][2]
- Method: logic—non-contradictory identification—applied through induction, deduction, and reduction, with strict context-keeping. [3]
- Validation: tie every claim to evidence; assign proper status (possible/probable/certain) by the weight and integration of facts; dismiss the arbitrary. [2][3]
- Application: ends–means integration guided by objective values and the virtues; act, observe outcomes, and iterate. [5][6]
- Advanced concept-formation: extracting essentials
- Contrast method: to form/clarify a concept, select units and close “foils” (what it is vs. what it isn’t); isolate the characteristic that explains the greatest number of similarities/differences. [2]
- Measurement-omission: identify a measurable attribute shared by the units and omit its particular measurements within a range; this yields a unit perspective (e.g., any specific length qualifies as “long” within a context). [2][4]
- Definitions by essentials: define by genus and differentia; choose the most fundamental distinguishing characteristic, given current context; update wording as knowledge expands without changing the referents. [2][3]
- Unit-economy: prefer the smallest number of concepts needed to cover the widest range of facts without contradiction; purge redundant or package-deal terms. [2][4]
- Induction that yields necessity, not mere habit
- Causal focus: treat induction as discovering how an entity’s identity necessitates its actions; run controlled variations to separate essentials from accidentals. [3][6]
- Methods of difference and agreement: vary one factor while holding others constant; track what changes and what remains invariant to expose causal drivers. [6]
- Measurement discipline: quantify where possible; discover functional relations (linear, exponential, threshold) that connect attributes and actions. [6]
- Generalization criterion: a universal principle is warranted when (a) the causal mechanism is identified, (b) counter-cases are shown to lack the causal conditions, and (c) predictions succeed under novel tests. [3][6]
- Reduction: tying higher-level ideas back to the perceptual base
- Procedure: take a high-level claim → identify its immediate premises → repeat until you reach first-hand or reproducible observations; state each link explicitly. [3]
- Test: if you cannot complete the chain to observations without gaps or equivocations, you are dealing with a floating abstraction. [2][3]
- Example pattern: principle → underlying law → operational definition of terms → measurement method → direct observations. [6][3]
- Standards of evidence and the status of propositions
- Burden of proof: on the asserter; absence of disproof is not evidence. [3]
- The arbitrary: a claim offered without evidence/context is to be dismissed, not refuted, because it asserts nothing cognitively. [2][3]
- Possibility: a claim is possible only if it identifies a specific causal route consistent with known facts; “not-contradicted” is not enough. [3]
- Probability: measured by the proportion and quality of independent evidence, integrated without contradiction to the rest of knowledge. [2]
- Certainty: contextually absolute when the total available evidence, integrated with established principles, leaves no unresolved alternatives. [3]
- Context-keeping: integration without contradiction
- Tree of knowledge: map higher-level conclusions to the lower-level nodes from which they derive; update the tree when any node changes. [2]
- Collision protocol: when a new finding seems to clash with a validated principle, check for (a) mismeasurement, (b) context-bound scope conditions, (c) equivocation of terms, (d) necessity to refine the principle’s statement. [3][6]
- Definition policing: never switch definitions mid-argument; mark contextual qualifiers in definitions to prevent stolen-concept fallacies. [2][3]
- Fallacies and how to detect them methodically
- Stolen concept: using a concept while denying or ignoring its logical roots (e.g., denying causality while asserting scientific “law”); fix by reduction. [2][3]
- Package-deal: smuggling different things under one term (e.g., equating rational self-interest with brute predation); fix by redefinition by essentials. [2]
- Reification of the zero: treating a non-existent as an existent (e.g., “the average family has 2.3 children” as a real entity); fix by identifying units and referents. [2]
- Equivocation: shifting meanings across a chain of reasoning; fix by explicit, stable definitions. [3]
- From fact to value without the “is–ought” gap
- Standard: life proper to a rational being; evaluate facts by their causal impact on that standard. [5]
- Virtues as action-principles: rationality, independence, integrity, honesty, justice, productiveness, pride—each names a policy that reliably achieves long-range values in reality. [5]
- Decision rule: choose the alternative whose causal consequences best advance your hierarchical values without initiating force or contradicting principles you must rely on tomorrow. [5]
- Practical checklists you can use
A. Investigative reasoning (science/engineering)
- State the problem as a factual question; define all terms ostensively or operationally. [6]
- Gather baseline observations; instrument and calibrate. [6]
- Generate causal hypotheses; list predicted invariances and failure conditions. [3][6]
- Design controlled tests; quantify; record. [6]
- Induce a generalization; deduce novel predictions; test again. [3][6]
- Reduce the final claim to observational chains; publish definitions and methods. [3][6]
B. Concept work (analysis/definitions)
- Identify the referents; gather close contrasts; ask “what makes these the same kind?” [2]
- Name the omitted measurements; state the genus and differentia. [2][4]
- Test against borderline cases; refine for essentials; eliminate package-deals. [2]
- Record the context of the definition; update wording as knowledge expands. [3]
C. Decision-making (policy/product/business)
- Specify the goal in factual terms; rank it within your value hierarchy. [5]
- List alternatives; map causal pathways and long-range effects; quantify trade-offs. [6]
- Exclude any means that require evasion or initiating force; they are self-contradictory in a rational life. [5]
- Choose, act, measure results; feed back into the knowledge base. [6]
- Handling anomalies and counterexamples
- Diagnose measurement error first; then check scope conditions; then re-inspect definitions; only then consider principle revision. [3][6]
- A true counterexample must match the concept’s definition and context; otherwise it is a misuse of terms. [2][3]
- Keeping emotions in their proper place in method
- Emotions are consequences of appraisals, not tools of cognition; treat them as data about your value premises, not about external facts; they neither validate nor invalidate propositions. [2][3]
- Miniature end-to-end example (schematic)
- Observation: a device shuts down at high load; temps spike. [6]
- Concept/definition: “thermal throttling” defined by temperature-triggered performance reduction. [2]
- Induction: tests show failure above 85°C at VRM; mechanism identified: voltage regulator derates with temperature. [6]
- Deduction: adding heatsink surface area and airflow should extend stable load to X amps. [3][6]
- Reduction: thermocouple logs and manufacturer derating curves confirm mechanism. [6]
- Decision/action: redesign cooling; verify via stress test; record results and update design guidelines. [6]
- Summary principles to enforce at every step
- Primacy of existence: facts over wishes; evidence over authority or consensus. [1]
- Logic as method: identity, non-contradiction, excluded middle; no gap between theory and observation via reduction. [3]
- Objectivity: context-keeping and integration; definitions by essentials; reject the arbitrary. [2]
- Ethics of cognition: independence and honesty—first-handedness in observation and inference; justice in crediting evidence. [5]
Use this as a disciplined workflow. It is not optional. Evade any link in the chain—perception, concept-formation, induction, reduction, integration, evaluation—and you sever knowledge from reality and action from causes. [1][2][3][5][6]
Sources
Finally:
Here is a concrete, end-to-end example of the process of reason applied to a real production problem—turning a collapsing yield on a CNC machining line into a solved, validated improvement, step by step, from observation to action to feedback.
- Choice to focus
- The manager refuses guesses and demands facts; he commits to sustained, first-handed inquiry as the precondition of reasoning. [1][3]
- Observation (perception as base)
- He gathers the immediate data: hourly yield reports, part measurements, machine sensor logs, tool-change records, ambient temperature, and material lot IDs, with calibrated instruments and timestamps. [1][6]
- Isolation and description of units
- He isolates the defect class: edge burrs exceeding tolerance at the same chamfered feature on parts from Mill #3 during the night shift. [2]
- Concept-formation and definitions by essentials
- He defines “burr” operationally as “raised material height > 0.10 mm measured via profilometer at the chamfer edge after pass two,” and “yield” as “conforming parts per hundred produced,” excluding rework. [2][4]
- Hypothesis generation (causal candidates)
- Plausible causes are listed by identity-bound mechanisms: excessive tool wear, feed rate too high for the carbide grade, coolant flow intermittency, or harder-than-normal material lots. [3][6]
- Controlled tests (methods of difference/agreement)
- Holding material lot constant, he varies feed rate in fixed steps and records burr height; then, holding feed rate constant, he swaps in new vs. worn tools; finally, he logs coolant pressure vs. time to check for intermittent drops. [6]
- Measurement and quantification
- Data show burr height rises sharply once measured flank wear exceeds 0.20 mm and rises further with feed above 0.30 mm/rev; coolant pressure fluctuations show no correlation with defect spikes. [6]
- Induction (identifying the causal principle)
- He generalizes: above a wear threshold W*, the cutting edge plastically deforms the surface instead of shearing it, and the effect is amplified by higher feed; therefore, burr formation is necessitated by the worn-edge geometry at given loads. [3][6]
- Deduction (deriving implications)
- If burrs are necessitated when wear > W* and feed ≥ F, then (a) reducing feed below F will temporarily reduce burrs even with current tools, and (b) a tool-change schedule enforcing wear < W* will eliminate the defect at standard feed. [3]
- Reduction (validation back to the perceptual base)
- He traces each premise to measurements: profilometer traces, microscope wear gauges, feed logs, and time-synced parts data; no link rests on authority or floating abstraction. [3][2]
- Context-keeping and integration
- He checks other mills and shifts; where operators followed the same carbide grade but had earlier tool changes, yields remained normal—confirming scope and guarding against contradiction. [2][3]
- Status assignment and burden of proof
- With convergent evidence and no residual alternatives consistent with the facts, the principle is contextually certain; arbitrary counter-claims without data are dismissed. [3][2]
- Evaluation by an objective standard
- He evaluates alternatives by the requirements of long-range production and profit—objective values rooted in productiveness—not by feelings or consensus. [5]
- Ends–means integration and decision
- He selects means that causally achieve the end: change to a tougher carbide grade, set a predictive tool-change interval at measured wear = 0.15 mm, restore the standard feed, and add an inline burr sensor for early warning. [6][5]
- Action under objective rules
- He implements the SOP changes, updates definitions and gauges, trains operators in the measurement method, and removes any procedure that invited ad hoc guessing. [5][6]
- Feedback and continuous improvement
- Over two weeks, yield returns from 91% to 99.6%; the inline sensor flags tools nearing the threshold; the data are folded back into maintenance planning and cost models; definitions and charts are archived. [6][3]
What this illustrates about the process of reason
- Primacy of existence: facts and measurements govern; wishes are irrelevant. [1]
- Logic as method: every step is non-contradictory identification—induction to find causes, deduction to project implications, reduction to anchor conclusions in observations. [3]
- Objectivity and context: clear concepts, operational definitions, burden of proof on the asserter, and integration of all relevant facts before assigning certainty. [2][3]
- Ethics in cognition and action: independence, honesty, and productiveness guide choices; the result is practical success because virtue is the policy of adhering to reality. [5]
Condensed chain (mapped to the case)
Focus → Observe/measure → Isolate/define → Hypothesize causes → Test/quantify → Induce principle → Deduce implications → Reduce to data → Integrate context → Validate status → Evaluate by objective standard → Decide and act → Measure outcomes → Iterate. [1][2][3][6][5]
This is reason in action: a disciplined chain from reality to knowledge to effective action, with every link tied back to facts and governed by logic. Evasion is not an option. [1][3]
Sources
----------------------------------------------------------------------------------------
Use the process of reason any time you need to move from facts to reliable conclusions and action. It is a general-purpose method for understanding, designing, making decisions, and correcting.
Where it applies (with what it does there)
- Philosophy and worldview
- Clarify concepts and definitions; integrate knowledge without contradiction; reduce claims to evidence; build a coherent metaphysics–epistemology–ethics that guides life.
- Science
- Observe, hypothesize, test, quantify, explain causally, generalize; predict novel results and check them.
- Engineering/design
- Translate needs into specifications; model causes; run trade‑offs; prototype, measure, iterate; validate against real performance.
- Medicine/diagnosis
- Gather symptoms and history, generate differentials, test discriminators, identify mechanism, choose treatment, and track outcomes.
- Psychology (scientific and personal)
- For research: operationalize constructs, run controlled studies, and analyze mechanisms. For self-management: treat emotions as data about appraisals, test beliefs against facts, run small behavioral experiments, measure change.
- Ethics and day-to-day decision-making
- Define goals by objective values, map causal pathways and long‑range effects, rule out means that contradict principles you depend on, choose and act, then update from results.
- Business/strategy/operations
- Diagnose bottlenecks, quantify customer value, test pricing and positioning, model unit economics, design SOPs, monitor KPIs, correct variances.
- Law and policy
- Define terms precisely, trace principles to individual rights and causal social outcomes, evaluate proposals by long‑range effects and non-initiation of force.
- Education and skill-building
- Set learning objectives, chunk concepts by essentials, practice with feedback loops, test transfer to new contexts.
- Art and criticism
- Make standards explicit (theme, style, craft), relate choices to intended values/effects, compare against exemplars; refine by audience feedback.
About “creating a rational religion”
- If “religion” means claims about supernatural beings/events: reason requires evidence tied to reality. Without testable evidence and causal mechanisms, such claims are arbitrary; reason cannot validate them.
- What reason can do: evaluate any religion’s moral teachings and life-practices by their real causal effects on human flourishing; extract what is true/useful, and reject what conflicts with facts. You can build a rational, secular life‑philosophy (worldview, ethics, rituals of reflection) grounded in evidence and logic.
When to invoke the full process
- High stakes or irreversible choices (career move, surgery, policy change)
- Conflicting claims or expert disagreement
- Novel situations or innovations
- Recurring failures or anomalies
- Moral questions with long-range consequences
- Any time your feelings, peers, or authority are pulling one way and facts are unclear
What the process will not do
- Validate assertions offered without evidence (the arbitrary)
- Substitute for missing data; it will assign only possibility/probability where evidence allows
- Deliver certainty beyond the context of established facts
- Justify short‑range gains that undercut long‑range life or rights
A compact everyday template
- Focus: commit to facts over feelings.
- Observe: gather the relevant evidence.
- Define the problem and terms.
- Generate causal hypotheses/options.
- Evaluate by causes, probabilities, trade‑offs, and principles; quantify where you can.
- Decide and act.
- Measure outcomes; correct and iterate.
Example in 30 seconds: considering a career change
- Observe: market data, your skills/performance, financial runway.
- Define: goal = more impact + 20% income growth within 2 years.
- Options: stay and negotiate; switch firms; retrain for an adjacent role.
- Evaluate: model income paths, risk, skill gaps; check references and placement rates; rule out any option requiring dishonest signaling.
- Decide/act: choose the highest expected value consistent with principles; set milestones; review at 90 days.
Bottom line: the process of reason is the universal, disciplined workflow for going from reality to knowledge to action—use it wherever truth and consequences matter.
In addition:
The senses are valid, but the map is not the territory
Properly understood, “the map is not the territory” names a trivial but crucial distinction: concepts, propositions, and models (maps) are human abstractions, while reality (the territory) exists independent of consciousness. Objectivism affirms the primacy of existence, the validity of the senses, and the possibility of objective, logically validated maps that faithfully correspond to reality. The senses are our direct contact with the territory; logic and objective method are how we build accurate maps from that base. [3] [1] [2] [4]
- Territory first: primacy of existence
- Reality exists independent of anyone’s wishes, fears, or decrees. Consciousness is the faculty of perceiving what is, not of creating it. This is the primacy of existence; it is the anchor that makes “map vs. territory” meaningful rather than an excuse for skepticism. [3] [4]
- Why the senses matter
- Perception is the given. The senses are valid—they put you in direct contact with entities and their attributes. They do not “manufacture” reality; they report it, within the range and form of each sense. Apparent “illusions” are cases of misinterpretation at the conceptual level or of unusual conditions, not evidence that perception is unreliable in principle. Instrumentation extends the senses without altering their basic cognitive status. [1] [2]
- What the “map” is
- The “map” comprises your concepts, definitions, generalizations, and theories: an economy of units formed by differentiation and integration, with measurement-omission identifying essentials while omitting non-essential measurements. Proper definitions are by genus and differentia and must be context-kept and updated as knowledge grows, without changing the referents. [2] [4] [6]
- The relationship: from territory to map without a “veil”
- There is no “veil of ideas” cutting you off from reality. You start on the ground—perception—and then abstract. Validation means reducing claims back to perceptual data via a chain of definitions and logical steps. Knowledge is contextual and hierarchical; certainty is contextual, earned when a conclusion is logically integrated and no counter-evidence exists within the full context. [2] [3] [1]
- How to make the map correspond to the territory (objective method)
- Observe: Gather perceptual data (directly and via instruments). Facts first. [1] [5]
- Conceptualize: Form and define concepts by essentials; use measurement-omission and maintain unit-economy. [4] [6]
- Integrate without contradiction: All identifications must cohere under the laws of identity and non-contradiction. Reject package-deals and stolen concepts. [2]
- Explain causally: Seek causal connections—causality is identity in action. Prefer theories that identify mechanisms, not just correlations. [3]
- Validate by reduction: Trace claims back to perceptual evidence; place the burden of proof on the asserter; dismiss the arbitrary as neither true nor false. [2] [3]
- Test and measure: Derive testable implications; use measurement to check magnitudes and limits; refine the map when data demand it. [5]
- Keep context: Update definitions and conclusions when the context expands, integrating new facts without discarding valid prior knowledge. [2]
- Calibrate language: Use precise, non–package-dealed terms; ban anti-concepts that smear distinct phenomena into one blob. [2]
- What this rejects
- Skepticism that treats maps as inescapably detached from reality. When concepts are formed and validated objectively, a map can be true of the territory. [2] [4]
- Subjectivism or consensus as standards. Reality—not feelings or votes—is the court of final appeal. [3]
Bottom line: The territory (reality) is primary; the senses are our means of contact with it; objective logic and method turn that contact into a reliable map. If you adhere to perception, logic, causality, integration without contradiction, measurement, and context-keeping, your map will correspond to the territory—and when it doesn’t, reality will tell you, and reason will show you how to correct it. [1] [2] [3] [4] [5] [6]
Sources
Finally: