Chapter 2 — Kahneman–Tversky: The Private-Diagnostic Pole

Kahneman and Tversky do not reduce the human being to selfishness. Their contribution is more precise and more powerful: they relocate rationality from the abstract economic agent to the psychologically bounded decision-maker. The human being is no longer treated as a transparent optimizer. He becomes a private decision-maker exposed to framing, loss, probability distortion, intuitive judgment, and systematic bias.

This relocation is one of the major intellectual achievements of modern decision theory. It challenges the assumption that human agents simply maximize expected utility in stable and coherent ways. It does not destroy the concept of rationality; it changes the question. Instead of asking only what a rational agent would choose under ideal conditions, it asks how actual human beings judge and choose under conditions of cognitive limitation.

Prospect Theory begins as a critique of expected utility theory as a descriptive model of decision-making under risk. Kahneman and Tversky argue that choices among risky prospects exhibit systematic effects that are inconsistent with the basic assumptions of expected utility theory. People evaluate outcomes relative to reference points, weigh losses differently from gains, and do not treat probabilities in a linear way.

This is a decisive move. The classical model assumes an agent whose preferences are sufficiently stable and coherent to be represented through utility. Kahneman and Tversky shift attention from formal coherence to psychological structure. The individual does not simply compare final states of wealth. He experiences gains and losses relative to a reference point. He is often more sensitive to losses than to equivalent gains. He may treat certainty differently from probability, and small probabilities differently from moderate or large ones.

In this sense, Prospect Theory does not abolish rationality. It diagnoses the limits of a particular model of rationality. The rational economic agent becomes psychologically bounded. Rationality remains present as a normative background, but the actual human being is shown to depart from it in patterned ways.

The claim is not that Kahneman–Tversky fail at a task they set for themselves. Their work does not aim to build a theory of family, community, or objective reason. The claim is narrower: when their framework is used as a general image of rationality, it remains private-diagnostic and does not by itself supply a communal or objective-critical account of reason.

The Prior Question of Risk and Loss

The keyword here is risk. If the reply is that Kahneman and Tversky studied decision-making under risk rather than family, community, or objective reason, the next question must be: risk to what, and to whom? Risk is never a free-floating category. It presupposes a bearer of value and a horizon within which loss matters. If risk is modeled primarily as risk to the private decision-maker, then the framework has already selected an anthropological unit. It has not merely described choice; it has located choice in the individual subject.

A theory of risk is also a theory of loss. It must presuppose what counts as loss, who bears the loss, and within what horizon the loss becomes meaningful. A loss is not merely a negative outcome on a formal scale. It is a loss for someone, or for something: a person, a family, a community, a rule, a future, a continuity, or a form of life.

A human being, however, does not exist as a purely isolated bearer of gain and loss. Human risk may be risk to family, community, continuity, trust, obligation, loyalty, or a shared form of life. This does not invalidate Prospect Theory as a theory of decision under risk. It clarifies its conceptual boundary: the theory is powerful precisely where risk can be represented through the private decision-maker, but it does not by itself explain how risk is transformed when the bearer of value is familial, communal, or universal.

This matters for the present essay because the diagnostic structure remains private. The decision-maker appears as an individual confronted with risk, uncertainty, gain, loss, and probability. Family, community, tradition, loyalty, and cosmopolitan obligation are not absent from human life; they are simply not constitutive elements of rationality within the basic architecture of Prospect Theory.

The critical point is not that Kahneman and Tversky deny family, community, or moral obligation. The critical point is that their framework does not build these elements as independent rational layers. A family interest can appear in a decision problem. A communal norm can influence preference. A religious or cultural commitment can shape framing. A social identity can affect risk perception. But these elements enter as variables in the psychology of choice. They do not become a separate definition of rationality.

This is the conceptual limitation that matters for the essay. When a person gives priority to family or community over immediate private material gain, the Kahneman–Tversky framework can describe the decision, but it does not necessarily classify the family or community interest as a higher form of rationality. It may describe the choice as preference, framing, affect, identity, norm, or bias. The framework has no internal need to say that family or community expands rationality itself.

This does not make the framework false. It makes it partial. Its strength lies precisely in the diagnosis of bounded judgment. But the more successful the diagnostic language becomes, the more important it becomes to ask what image of the human being it leaves behind.

The diagnostic achievement of Kahneman–Tversky is enormous. They show that human beings are not transparent calculators. They show that judgment is context-sensitive, intuitive, unstable, and often systematically mistaken. They make visible a hidden architecture of decision-making that classical rational choice theory had flattened.

But the risk of this achievement is conceptual reduction. Once the human being is described mainly through bias, loss, framing, intuitive error, and probability distortion, the richer image of the human being can recede. The person may begin to appear less as a bearer of ends and more as a mechanism of response. He becomes a system to be predicted, corrected, nudged, managed, or optimized.

This is not necessarily what Kahneman and Tversky intended. It is a risk internal to the reception and use of their framework. A diagnostic theory of bounded rationality can become an instrumental grammar of human management. The human being as a private decision-maker can become the human being as a behavioral mechanism.

The strongest version of the argument is not that Kahneman–Tversky are irrationalist. That would be false.

The stronger and more accurate argument is that Kahneman–Tversky expose the boundedness of private decision-making, but they do not restore a concept of reason that can judge the ends themselves.

A theory of decision under risk always contains, implicitly or explicitly, a theory of the bearer of risk.

This places them precisely within the first axis of the essay. They represent the private-diagnostic pole of subjective-instrumental reason: rationality as the analysis of judgment and choice under conditions already structured by options, risks, losses, gains, probabilities, and reference points.

Chapter 3 — Aumann: The Strategic-Communal Pole

Aumann enters the argument as a correction of the private-diagnostic picture. If Kahneman–Tversky analyze the bounded individual under risk, uncertainty, framing, loss, and bias, Aumann analyzes the agent within games, incentives, rules, repetition, conflict, cooperation, family, and community.

This correction is not external to subjective reason. It does not yet restore objective reason in the Frankfurt sense. Aumann does not begin by asking whether the agent’s ends are worthy of universal rational legitimacy. He begins by asking how an agent, given his information and interests, may act rationally within a strategic environment.

Aumann’s definition of rationality is crucial: a person’s behavior is rational if it is in his best interests, given his information. This definition does not identify rationality with moral goodness, cosmopolitan legitimacy, or objective truth. It identifies rationality with the agent’s best interests under informational conditions.

The move from Kahneman–Tversky to Aumann is a move from the private decision-maker to the strategic agent. In the Kahneman–Tversky framework, the central question is how the individual evaluates gains, losses, probabilities, risks, and frames. In Aumann’s framework, the central question is how an agent acts when others are also acting, anticipating, reacting, signaling, threatening, cooperating, defecting, punishing, remembering, and repeating.

This difference matters because game theory does not treat action as an isolated choice. It treats action as part of an interaction structure. A choice is not merely a psychological event; it is a move within a game. The meaning of the move depends on expectations, incentives, information, and the anticipated behavior of others.

The decisive point for the present essay is that Aumann’s framework can model family and community as part of rational agency itself. If rationality means acting in one’s best interests given one’s information, then the content of “interest” is not limited to immediate private material gain. An agent’s interests may include family, community, religious commitment, loyalty, reputation, long-term survival, group continuity, or adherence to a rule. These are not necessarily irrational intrusions into the calculus. They may be part of the calculus.

The crucial point is not that Aumann develops a moral theory of family or community. The point is that his framework can represent such commitments as internal to the agent’s utility function rather than as irrational disturbances of choice.

This is the internal correction Aumann offers against the narrower behavioral picture. A person who sacrifices private gain for family is not necessarily irrational. A person who acts for community is not necessarily captured by bias. A person who follows a long-term rule rather than maximizing in a single episode is not necessarily defective as a rational agent. The relevant question becomes: what are the agent’s goals, what does he know, what game is he in, and what rule of action is he following?

Aumann’s work on rule-rationality strengthens this point. Rule-rationality does not assume that people maximize utility in every isolated act. It suggests that people often follow rules or modes of behavior that usually, though not always, maximize utility. This makes Aumann’s model more hospitable to social forms of rationality: family, community, loyalty, and long-term codes of conduct can be understood as rule-governed structures. They may appear irrational from the standpoint of one isolated act, but rational from the standpoint of repeated interaction and long-term orientation.

Repetition is central here. In a one-shot decision, immediate gain may dominate. In repeated games, the future enters the present. Reputation, punishment, reward, trust, threat, cooperation, memory, and deterrence acquire rational force. A rule that seems costly in a single act may become rational over time because it stabilizes expectations and preserves cooperation.

This logic is especially important for understanding family and community. A family is not a single decision problem. A community is not a single utility calculation. Both are continuing frameworks of expectation, obligation, memory, sanction, and trust. In this sense, Aumann gives rational dignity to forms of life that cannot be reduced to immediate private gain.

Aumann’s correction of Kahneman–Tversky has three main strengths:

First, he prevents rationality from collapsing into private instinctive egoism.

Second, he prevents “irrationality” from becoming an explanatory shortcut.

Third, he restores the agent’s world: interaction, time, repetition, rule, conflict, cooperation, and social embedding.

But Aumann’s strength is also his boundary. He widens the subject, but he does not transcend the subject. He expands rationality from the private chooser to the strategic agent, and from the strategic agent to family, community, loyalty, and long-term rule. But he does not introduce an objective-critical criterion above the agent’s ends.

The family can be rational for the agent. The community can be rational for the agent. A religious or national commitment can be rational for the agent. But the question remains: rational in relation to what higher standard?

Aumann corrects private rationality by adding community, but he does not correct communal rationality by adding objective critique.

This is the decisive boundary. Aumann restores the social thickness of the agent, but he does not restore humanity as the highest criterion of reason.

Chapter 4 — Grillparzer’s Motif of Decline

Grillparzer’s line is not introduced here as literary ornament. It is introduced as a structural formula of decline: when humanity loses authority as the highest reference point of reason, rationality can pass first into communal identity and then into instrumental management of the human being as object.

Der Weg der neuern Bildung geht

Von Humanität

Durch Nationalität

Zur Bestialität.

The way of modern cultivation, education, or formation goes from humanity, through nationality, to bestiality. The present essay does not use this line as a historical proof. It uses it as a heuristic schema: a conceptual diagram, not a deterministic law. The terms Humanität, Nationalität, and Bestialität name three moments in the decline of reason.

Humanität

Humanität corresponds to objective-critical reason. Humanity means that the human being is not first a private decision unit, not first a player in a game, not first a family member, not first a member of a community, and not first a national subject. The human being is first a bearer of universal significance. This does not eliminate family, community, nation, or historical identity. It places them under a higher criterion.

Nationalität

Nationalität should not be read too narrowly. In the context of this essay, it does not refer only to modern nationalism in a political sense. It names a wider stage: the stage in which the universal human standpoint is replaced by collective belonging. This collective belonging may appear as family, community, people, religion, nation, tradition, historical identity, or group loyalty. It is wider than the private individual, but narrower than humanity.

Aumann’s framework corresponds structurally to this middle term insofar as it gives rational standing to communal belonging without yet restoring humanity as the highest criterion. This does not make Aumann a nationalist thinker. It means that his model gives family, community, loyalty, rule, repeated interaction, and long-term commitment a rational place inside subjective reason.

But Grillparzer’s middle term also marks a danger. Community is wider than the individual, but it is not yet humanity. A community may discipline egoism, but it may also absolutize itself. A family, group, people, nation, or tradition can become the final limit of reason. When that happens, the communal correction of private egoism becomes a new closure.

Bestialität

Bestialität is the most dangerous term and must be handled carefully. In this structural use, it does not refer to biological animality. It refers to the degradation of the human being into an object of calculation, mobilization, prediction, and control.

This is the possible endpoint of instrumental reason when it is detached from objective-critical reason. The human being is no longer viewed as a bearer of universal dignity. He is no longer even primarily a member of a moral community. He becomes a mechanism.

At this stage, the language of rationality changes. It no longer asks whether ends are justified. It no longer asks whether the community has a human limit. It asks how behavior can be predicted, shaped, optimized, controlled, incentivized, nudged, or administered.

Kahneman–Tversky’s framework can be placed near this diagnostic-instrumental risk, not because of its intention, but because of its possible reception as a language of behavioral management. Their work shows that people are subject to systematic bias. This is a major intellectual achievement. But once the human being is described mainly through bias, loss aversion, framing, probability distortion, and intuitive error, a managerial grammar becomes possible. The person can be treated as a behavioral system to be corrected or exploited.

The central problem is the disappearance of Humanität. If humanity remains the highest criterion, then family, community, nation, and decision systems remain limited by the universal human standpoint. The agent may pursue interests, but not every interest becomes legitimate merely because it is his. A community may pursue goals, but not every communal goal becomes rational merely because it preserves the community. A behavioral system may predict action, but prediction does not exhaust the meaning of the human being.

Grillparzer provides the motif; Horkheimer provides the philosophical analysis. The movement from Humanität to Nationalität to Bestialität can be translated into Horkheimer’s terms as the decline from objective reason into subjective reason and then into instrumental domination. Humanity corresponds to reason capable of judging ends. Nationality corresponds to subjective reason expanded into collective belonging. Bestiality corresponds to instrumental reason emptied of universal human restraint.

Chapter 5 — Horkheimer and Adorno: Objective Reason and the Critique of Ends

Horkheimer and Adorno do not merely add a moral objection to behavioral decision theory or game theory. They change the level of the question. The issue is no longer whether the individual chooses consistently, or whether the agent acts according to his interests. The issue is whether reason has retained the power to judge the ends it serves.

This is why the Frankfurt School cannot be placed beside Kahneman–Tversky and Aumann as a third parallel theory of rationality. Kahneman–Tversky and Aumann remain within the analysis of rationality relative to given ends. Horkheimer and Adorno ask whether those ends themselves can claim rational legitimacy.

Subjective Reason

Subjective reason is reason understood from the standpoint of the subject. It asks what is useful, effective, adaptive, coherent, efficient, or advantageous for a given agent or system. It does not primarily ask whether the end itself is true, good, just, humane, or rationally legitimate.

Instrumental Reason

Instrumental reason is not separate from subjective reason in Horkheimer’s sense; it is the operational form of subjective reason when reason is reduced to usefulness, adaptation, self-preservation, and effective means. This is not wrong by itself. Human beings must choose means, compare strategies, anticipate consequences, and act under constraints. A society that cannot calculate means cannot survive.

The problem begins when instrumental reason becomes the whole of reason. When rationality is reduced to the efficient selection of means, the question of ends disappears. At that point, any goal can become “rational” if the means chosen for it are coherent and effective.

Objective Reason

Objective reason is different. It does not ask only whether a means is suited to an end. It asks whether the end itself has rational standing.

Objective reason therefore contains a normative demand. It implies that reason is not merely an instrument of the subject, the family, the group, the state, the market, or the institution. Reason has the authority to judge these entities. It can ask whether their goals are true, humane, just, universalizable, or destructive.

This is why objective reason corresponds to Humanität in the Grillparzer structure. Humanity is not only compassion. It is the claim that the human being, as human, has priority over private interest, communal loyalty, national identity, technical efficiency, and administrative control.

Without objective reason, there is no stable basis from which to criticize the ends of a subject or group. There may still be preferences, incentives, strategies, institutions, identities, and calculations. But there is no higher rational tribunal before which those ends must answer.

From this standpoint, both Kahneman–Tversky and Aumann are incomplete. Kahneman–Tversky diagnose the private individual. Aumann expands the agent into family, community, and strategy. But neither framework, by itself, restores the critique of ends.

A decision may be psychologically understandable and still serve an unworthy goal.

A strategy may be coherent and still serve a destructive end.

A community may be rational for its members and still violate humanity.

A rule may be stable and still be wrong.

The connection to Dialectic of Enlightenment sharpens the point. Horkheimer and Adorno do not argue that reason simply disappears in modernity. They argue that Enlightenment reason can turn against itself when it becomes domination through calculation, classification, administration, and control. The danger is not irrationality alone; it is rationalized domination.

For Horkheimer and Adorno, the catastrophe of modern reason is not that reason disappears, but that reason survives in a narrowed form: as calculation without reconciliation, administration without truth, and domination without self-critique.

This is the central lesson of the Frankfurt School for the meaning of rationality. A society may become more rationalized and less reasonable at the same time. It may improve technique while losing the capacity to judge ends. It may optimize administration while degrading humanity. It may refine calculation while abandoning critique.

This is why the Frankfurt critique is not merely moral. It is conceptual. It asks whether a form of rationality that cannot judge its own ends still deserves the name reason in the full sense.

Kahneman–Tversky diagnose the private subject under bounded judgment.

Aumann expands the subject into family, community, rule, and strategic interaction.

Horkheimer and Adorno ask whether the ends served by both forms of rationality deserve rational legitimacy.

Conclusion — Aumann Corrects Kahneman, Frankfurt Corrects Both

The argument of this essay has not been that Kahneman–Tversky, Aumann, and the Frankfurt School offer three competing theories of rationality. Its argument has been that Kahneman–Tversky and Aumann occupy two poles of subjective-instrumental reason, while Horkheimer and Adorno recover the missing question of objective reason: whether the ends themselves are worthy of rational legitimacy.

Kahneman–Tversky represent the private-diagnostic pole. They show that the individual decision-maker is not the transparent rational agent assumed by classical economic theory. Human judgment is bounded, framed, intuitive, loss-sensitive, probability-distorting, and exposed to systematic bias. Their achievement is to make the boundedness of actual judgment visible within the language of rational analysis.

Aumann represents the strategic-communal pole. He corrects the narrowness of the private decision-maker by restoring the agent’s world: games, incentives, rules, repetition, cooperation, conflict, family, community, loyalty, and long-term conduct. For Aumann, a person who acts for family or community is not necessarily irrational. Family and community can belong inside the rational structure of the agent’s interests.

But this correction remains inside subjective reason. Aumann expands the subject; he does not place above the subject an objective-critical criterion that judges the subject’s ends. That is the task of Frankfurt.

Grillparzer’s line gives the structure a compact historical and poetic form. The movement from humanity, through nationality, to bestiality names the danger of decline when universal humanity loses priority. Aumann’s framework corresponds structurally to the middle term, not as crude nationalism, but as the communal expansion of subjective reason. Kahneman–Tversky’s framework corresponds structurally to the diagnostic-instrumental risk, not because it advocates dehumanization, but because its language can be absorbed into systems that treat the human being primarily as a behavioral mechanism.

Subjective reason asks what serves the subject.

Instrumental reason asks what works.

Objective reason asks whether the purpose itself deserves rational authority.

Kahneman–Tversky reveal the boundedness of the individual decision-maker. Aumann corrects that narrowness by expanding rationality into family, community, rule, and strategic interaction. Horkheimer and Adorno then reveal the limitation of both: rationality cannot be complete unless it can judge the ends themselves.

This formula preserves the value of each position. Kahneman–Tversky are necessary because rationality cannot ignore psychological boundedness. Aumann is necessary because rationality cannot be reduced to private decision mechanics. Frankfurt is necessary because rationality cannot remain a servant of given ends.

Aumann corrects Kahneman. Frankfurt corrects both.

The meaning of rationality cannot be exhausted by decision quality, behavioral diagnosis, strategic success, family interest, communal loyalty, or rule-following. These are important forms of rationality, but they remain incomplete when they do not submit their ends to criticism.

The central danger of modern reason is not irrationality alone. It is rationality without objective judgment: calculation without humanity, strategy without universal limit, community without critique, and management without a rational account of ends.

The question of rationality must therefore be returned to its highest form:

not merely how the individual chooses,

not merely how the agent succeeds,

not merely how the community preserves itself,

but whether the ends served by choice, agency, and community are worthy of reason.

​

This essay positions rationality as a layered problem rather than a single technical standard. Kahneman–Tversky clarify the bounded structure of individual judgment; Aumann expands rationality into games, rules, repetition, family, community, and long-term strategy; the Frankfurt School restores the decisive question of whether the ends served by rational systems are themselves worthy of reason. In this sense, the article contributes to the broader RATIUM.AI project by distinguishing decision, strategy, and the critique of ends as separate but interconnected levels of rational analysis.

​

References

Adorno, Theodor W., and Max Horkheimer. Dialectic of Enlightenment. 1944/1947.

Aumann, Robert J. War and Peace. Nobel Prize Lecture, 2005.
Nobel Prize lecture PDF

Aumann, Robert J. Rule-Rationality versus Act-Rationality. 2008.

Grillparzer, Franz. “Der Weg der neuern Bildung geht...” Epigram, 1849.
Zeno.org source text

Horkheimer, Max. Eclipse of Reason. 1947.
Stanford Encyclopedia of Philosophy: Max Horkheimer

Kahneman, Daniel. Maps of Bounded Rationality: Psychology for Behavioral Economics. Nobel Prize Lecture / American Economic Review, 2003.
Nobel Prize lecture page

Kahneman, Daniel, and Amos Tversky. “Prospect Theory: An Analysis of Decision under Risk.” Econometrica, 1979.
Prospect Theory PDF