🇩🇪 Deutsche Version: Münchhausen-Trilemma

The Münchhausen trilemma is Hans Albert’s epistemological diagnosis from the Treatise on Critical Reason (1968), formulated in the wake of Popper’s “Fries trilemma”: whoever seeks to provide ultimate justification for a proposition — final, unrevisable, resting on secure ground — necessarily arrives at one of three outcomes. The name cites the mendacious baron who pulls himself out of the swamp by his own hair: self-justification without an external point of support is exactly that — and logically impossible.

The three horns

  1. Infinite regress — every justification in turn requires a justification, ad infinitum. The chain never reaches load-bearing ground.
  2. Logical circle — the conclusion draws on the very premise it is meant to prove. The justification presupposes what it is supposed to show.
  3. Dogmatic termination — the chain is ended at an arbitrarily posited “Archimedean point” that itself remains unjustified and is withdrawn from criticism.

The three horns are mutually exclusive; a concrete attempt at ultimate justification ends in exactly one of them. The ancient precursor of the circle horn is the problem of the criterion (the diallelus) in Sextus Empiricus (Outlines of Pyrrhonism I 164 ff.): to recognize what is true, one needs a criterion; to justify the criterion, a meta-criterion — the circle of criteria.

Albert’s consequence — often overlooked

Albert does not draw scepticism from the trilemma but critical rationalism: the renunciation of ultimate justification in favour of fallibilism and systematic critical examination. In place of “secure ground” comes the revisable posit under permanent test — one posits a ground, but declares it fallible and keeps it open to scrutiny. This stands in the neighbourhood of Popper’s falsification principle (Popper falsification argument).

The grounded termination — Wittgenstein’s counter-point

In On Certainty (posthumously 1969), Ludwig Wittgenstein formulates the constructive answer to the third horn: “At the foundation of well-founded belief lies belief that is not founded” (§ 253); justification comes to an end, and the end is “our acting, which lies at the bottom of the language-game” (§ 204). On this view the regress does not end in arbitrariness but in grounded practice and the lifeworld. Such a termination differs categorially from the dogmatic one: it declares its ground, keeps it revisable — and it is carried out and answered for by a person. Because responsibility can be attributed only to persons, the legitimate endpoint of a chain of justification coincides with the locus of the attribution of responsibility.

Insight and archphenomenon — the realist-phenomenological answer

Realist phenomenology (Reinach, Hildebrand, Seifert) contests the completeness of the trilemma at its root: it captures only deductive justification — the derivation of propositions from propositions. Beside it, however, stands insight: the intellectual grasp of necessary essential states of affairs in essential intuition, in which the thing itself gives itself. Whoever sees that nothing can at once and in the same respect be and not be, or that a promise founds an obligation (Reinach), does not derive — he stands at the end of the chain of justification without any of the three horns obtaining: no regress (the insight needs no further support), no circle (nothing is presupposed that would have to be shown), no dogmatic termination (nothing is arbitrarily posited — the self-givenness of the thing justifies the end). In this sense insight is a genuine possibility of ultimate justification: not “backwards” through further propositions, but “downwards” into evidence.

The archphenomenon is the ontological locus of this answer: archphenomena — person, truth, cognition, being — cannot be further derived or reduced to something else, yet precisely not for that reason “unjustified”: they are accessible in immediate exhibition and are what carries all deriving in the first place. What Albert describes as the threatening arbitrariness of the termination is here, conversely, the foundation: at the bottom of the chain of justification lies not unfounded belief (Wittgenstein) but the self-given phenomenon. Albert replied that the experience of evidence is a psychic state and no guarantor of truth — evidence can deceive; the phenomenological counter-question is whether this reply can be uttered without itself laying claim to insight.

Significance for the AI debate

The trilemma provides the precise diagnosis for the question of whether an AI system can establish the validity of its own results: the attempt is the Münchhausen metaphor taken literally and falls prey to the circle — empirically confirmed three times over (cf. AI self-validation). Machine knowledge work becomes viable only through independent examining authorities (cf. verification authority): the regress becomes a finite chain of checks, the circle is avoided by separating generator and examiner, and dogmatism is replaced by revisable posits.

Demarcation: propositions, not being

The trilemma concerns the epistemic justification of propositions. The ontological question of the ground of being of the contingent — why there is anything at all that could also not be (contingency) — is categorially distinct from it: it asks not for the deductive derivation of a statement but for the ground of beings (Absolute Being). Albert himself contested this separation and turned the trilemma against cosmological and theological ultimate justification as well (Das Elend der Theologie, 1979) — and precisely here the counter-argument begins.

Counter-argument

Karl-Otto Apel and Vittorio Hösle object that the trilemma presupposes a deductivist concept of justification: justification is more than derivation from premises. Reflexive ultimate justification retrieves the non-circumventable presuppositions of arguing itself — whoever argues against them makes use of them and commits a performative self-contradiction. Moreover, fallibilism applies to itself: whoever declares everything revisable cannot exempt the validity of the logic on which the trilemma rests. The reach of this critique: it strikes Albert’s universalization of the trilemma to every ultimate justification — but not the circle diagnosis of self-validation, for Apel too demands a level of reflection distinct from the instance under examination.

Ontological classification: Subordinate concepts (pairwise distinct): infinite regress, logical circle, dogmatic termination; the grounded termination is distinct from the dogmatic one and is the legitimate form of the regress’s end.

Sources: Generated by querying the Personhood ontology. Research as of 14 July 2026 (research report The Münchhausen Problem in LLM-Assisted Ontology Engineering).

Further sources:

  • Albert, Hans (1968): Traktat über kritische Vernunft. Tübingen: Mohr Siebeck. (English: Treatise on Critical Reason. Princeton: Princeton University Press 1985.)
  • Albert, Hans (1979): Das Elend der Theologie. Hamburg: Hoffmann und Campe.
  • Sextus Empiricus: Outlines of Pyrrhonism I 164–169.
  • Wittgenstein, Ludwig (1969): On Certainty. Ed. G. E. M. Anscombe and G. H. von Wright. Oxford: Blackwell, §§ 166, 204, 253.
  • Popper, Karl R. (1935): Logik der Forschung. Vienna: Julius Springer. (English: The Logic of Scientific Discovery, 1959.)
  • Apel, Karl-Otto (1976): Das Problem der philosophischen Letztbegründung im Lichte einer transzendentalen Sprachpragmatik. In: Kanitscheider, Bernulf (ed.): Sprache und Erkenntnis. Festschrift für Gerhard Frey. Innsbruck, pp. 55–82.
  • Hösle, Vittorio (1990): Die Krise der Gegenwart und die Verantwortung der Philosophie. Munich: C. H. Beck.
  • Reinach, Adolf (1913): Die apriorischen Grundlagen des bürgerlichen Rechtes. Jahrbuch für Philosophie und phänomenologische Forschung 1, pp. 685–847. (English: The Apriori Foundations of the Civil Law, Aletheia 3, 1983.)
  • Hildebrand, Dietrich von (1976): Was ist Philosophie? Gesammelte Werke Vol. I. Regensburg: Josef Habbel. (English: What Is Philosophy? London: Routledge 1991.)
  • Seifert, Josef (1987): Back to ‘Things in Themselves’. A Phenomenological Foundation for Classical Realism. London: Routledge & Kegan Paul.

See also

Generated by querying the Personhood ontology.