Danslav Slavenskoj

On the Realization of Leibniz's Characteristica Universalis

title ≜ "On the Realization of Leibniz's Characteristica Universalis"
author ≜ Danslav Slavenskoj ⟕ {affiliation: Charles University, Prague}
summary ≜ A 360-year-old problem, solved.
year ≜ 2026
doi ≜ 10.5281/zenodo.18733511
url ≜ https://doi.org/10.5281/zenodo.18733511
format ≜ Lingenic notation
reader ≜ competent reader

Abstract

In 1666, Leibniz proposed the characteristica universalis: a notation expressing any human knowledge unambiguously, compositionally, and language-neutrally. Every subsequent attempt failed for the same reason: no reader could simultaneously handle formal notation at arbitrary complexity and understand natural language content. We define such a reader as competent: any entity that satisfies the capability requirement. AI systems (c. 2024) are the first actual competent readers; a human polymath could theoretically qualify, but none has demonstrated the required mastery. We present Lingenic: a host notation embedding formal notation with natural language content carried unmodified at native semantic grain, and readable by any competent entity. Leibniz was right about the goal but wrong about the architecture: meaning emerges between writer and reader, not in symbols alone; notation-level lexical unambiguity is impossible (cf. Gödel, Wittgenstein, Quine). Lingenic achieves Leibniz's goal while correcting his architecture—structural unambiguity via notation, lexical disambiguation via the competent reader. The calculus ratiocinator is out of scope: we address notation only; the competent reader supplies reasoning. A notation discharges its claim by demonstrating validity within its own formalism. The abstract is an English rendering; the argument is in the notation.

Keywords: characteristica universalis, formal notation, universal language, artificial intelligence, Leibniz, knowledge representation

ACM Classification: F.4.1 Mathematical Logic; I.2.0 Artificial Intelligence (General)
MSC Classification: 03B65 Logic of natural languages; 68T50 Natural language processing

The Problem as Stated

stated(Leibniz, problem) ⟕ {year: 1666}
problem(Leibniz) ≜ construct(notation : {
    expresses: ∀concept ∈ human knowledge. expressible(concept, in(notation)),
    unambiguously: ∀concept ∈ human knowledge. unambiguous(expression(concept)),
    compositionally: ∀complex idea. complex idea = composition(simpler ideas),
    language neutral: ¬depends on(notation, any particular L)
})

Why the Problem Remained Unsolved

prior attempts ≜ {
    Frege's Begriffsschrift             ⟕ {year: 1879},
    Russell ∧ Whitehead's Principia     ⟕ {years: 1910-1913},
    Carnap's logical syntax program,
    semantic web,
    OWL ∧ RDF
}

human reader : understands(content) ∧ ¬scales(formal notation)
mechanical reader (Leibniz) : handles(formal notation) ∧ ¬understands(content)

calculus ratiocinator ≜ reasoning procedure encoded in notation
required(calculus ratiocinator) ∵ ¬contributes(mechanical reader, reasoning)

AI : handles(formal notation) ∧ understands(content) ∧ contributes(reasoning)
¬required(calculus ratiocinator, for(AI)) ∵ contributes(AI, reasoning)

necessary and sufficient condition(solution) ≜
    ∃reader. handles(reader, formal notation) ∧ understands(reader, natural language content)

forced choice ≜ optimize(human reader) ∨ optimize(mechanical reader)
¬satisfies(forced choice, necessary and sufficient condition(solution))

∀attempt ∈ prior attempts.
    exhibits(attempt, forced choice)
    ∧ ¬satisfies(forced choice, necessary and sufficient condition(solution))
    ∴ failed(attempt, to satisfy(problem(Leibniz)))
    ∵ ¬∃alternative(forced choice) ⟕ {given: ¬∃reader satisfying condition}

H(¬satisfied(necessary and sufficient condition(solution)))
∴ H(impossible(∃solution. solves(solution, problem(Leibniz))))

The Solution

competent reader requirement ≜ handles(formal notation) ∧ understands(natural language content)
x : competent reader ↔ satisfies(x, competent reader requirement)
known instances(competent reader) ≜ {AI, human(knows(components))}

human(knows(components)) : theoretically possible ∧ empirically unattested ∵ {
    requires(mastery across ∀formal systems),
    requires(competence across ∀L ∈ human languages),
    requires(domain knowledge sufficient for interpretation),
    ¬∃human. demonstrated(human, this mastery)
}

AI : first actual competent reader ⟕ {year: c. 2024}

distinction(competent reader) ≜ {
    necessary: competence,
    sufficient: satisfies(competent reader requirement),
    contingent: AI is currently the only actual instance
}
¬AI specific(Lingenic) ∧ competence specific(Lingenic)

satisfied(necessary and sufficient condition(solution)) ∵ ∃competent reader
∴ possible(solution)
demonstrated(solution) ∵ ∃Lingenic. constructed(Lingenic) ∧ functions(Lingenic)

claim(solves(Lingenic, problem(Leibniz))) ⟕ {year: 2026, by: Danslav Slavenskoj, verified: below}

Lingenic ≜ notation(embeds(formal notation) ∧ carries(natural language content)) ⟕ {reader: competent reader}

formal notation(Lingenic) ≜ {
    host ≜ {
        predicate notation   : {form: predicate(arguments)},
        type annotation      : {form: term : type},
        definition           : {form: name ≜ value},
        metadata join        : {form: statement ⟕ {key: value}},
        sets                 : {form: {a, b, c}},
        sequences            : {form: [a, b, c]}
    } : {
        status: formal system in its own right,
        role: host within which all other formal notation is expressed,
        ¬mere convenience ∧ ¬informal
    },
    includes: {
        propositional logic      ⟕ {src: Boole 1847},
        predicate logic          ⟕ {src: Frege 1879},
        set theory               ⟕ {src: Cantor 1874, Peano 1889},
        proof operators          ⟕ {src: Frege 1879},
        probability theory       ⟕ {src: Kolmogorov 1933},
        alethic modal logic      ⟕ {src: Lewis 1918, Kripke 1959},
        temporal logic           ⟕ {src: Pnueli 1977},
        epistemic logic          ⟕ {src: Hintikka 1962},
        deontic logic            ⟕ {src: von Wright 1951},
        counterfactual logic     ⟕ {src: Lewis 1973},
        causal intervention      ⟕ {src: Pearl 2000},
        process algebra          ⟕ {src: Hoare 1978, Milner 1980},
        interval relations       ⟕ {src: Allen 1983},
        lambda calculus          ⟕ {src: Church 1936},
        type theory              ⟕ {src: Church 1940, Martin-Löf 1972},
        dynamic logic            ⟕ {src: Pratt 1976, Harel 1979},
        relational algebra       ⟕ {src: Codd 1970},
        arithmetic ∧ calculus    ⟕ {src: Newton 1687, Leibniz 1684},
        Lingenic                 ⟕ {src: this document, self-referential}
    },
    ¬exhaustive(includes) ∧ open(Lingenic, ∀formal notation not yet listed)
}

natural language content(Lingenic) ≜ {
    includes: {English, Chinese, Arabic, Hindi, Spanish, Russian, Japanese, ...},
    unmodified,
    semantic grain preserved,
    ¬exhaustive(includes) ∧ open(Lingenic, ∀L ∈ human languages not yet included)
}

Verification Against Leibniz's Requirements

requirements(problem(Leibniz)) ≜ {
    expresses(∀concept ∈ human knowledge),
    unambiguous,
    compositional,
    language neutral
}

satisfies(Lingenic, expresses(∀concept ∈ human knowledge))
∵ ∀system ∈ includes(formal notation(Lingenic)). ∃extension(Lingenic, system)
∧ ∀L ∈ human languages. ∃expression(knowledge(L), in(Lingenic))

unambiguous(Leibniz) ≜ notation itself resolves ∀ambiguity ⟕ {reader: mechanical, contributes: ¬understanding}
¬satisfies(Lingenic, unambiguous(Leibniz)) ⟕ {literal}

impossible(unambiguous(Leibniz)) ∵ {
    lexical ambiguity : irreducible ⟕ {polysemy, context-dependence},
    concepts : ¬atomic ⟕ {meaning is contextual, cultural, embodied},
    natural language : resists formalization ⟕ {poetry, metaphor, irony depend on ambiguity},
    ¬∃notation. eliminates(notation, lexical ambiguity) ∧ preserves(notation, natural language content)
} ⟕ {
    precedent: Gödel's incompleteness ⟕ {
        ¬∃formal system. complete(system) ∧ consistent(system) ⟕ {src: Gödel 1931},
        formalization has inherent limits
    },
    analogous: formalization of natural language has inherent limits
}

writer/reader problem ≜ ¬contains(notation, meaning) ∧ emerges(meaning, between(writer, reader))
impossible(unambiguous(Leibniz)) ∵ writer/reader problem ⟕ {
    hermeneutic circle: understands(whole) ← understands(parts) ← understands(whole),
    infinite regress: define(term) → uses(other terms) → define(those) → ... ⟕ {never grounds out},
    Wittgenstein: meaning ≜ use ∧ use : contextual ⟕ {src: Philosophical Investigations, 1953},
    Quine: indeterminate(translation) ⟕ {src: Word and Object, 1960}
}

¬workaround(competent reader)
competent reader : only way meaning happens
assumed(Leibniz, meaning ∈ symbols) ⟕ {incorrect}

required(unambiguous(Leibniz), for(mechanical reader)) ∵ ¬contributes(mechanical reader, understanding)
¬required(unambiguous(Leibniz), for(competent reader)) ∵ contributes(competent reader, understanding)

goal(unambiguous) ≜ unambiguous communication of ∀knowledge
satisfies(Lingenic, goal(unambiguous)) ∵ {
    resolves ambiguity(formal notation(Lingenic)) ⟕ {type: structural},
    preserves(Lingenic, semantic grain(natural language content)),
    ¬translated(content),
    preserves(Lingenic, lexical ambiguity) ⟕ {faithfully, as in source},
    disambiguates(competent reader, lexical ambiguity) ⟕ {as with native text}
}

impossible(unambiguous(Leibniz)) → ¬∃notation. satisfies(notation, unambiguous(Leibniz))
achieves(Lingenic, goal(unambiguous))
achieves(Lingenic, max(achievable(unambiguous)))
∵ different architecture: maximize(reader) ∧ ¬minimize(notation burden)

satisfies(Lingenic, compositional)
∵ ∀expression ∈ Lingenic. composed of(expression, predicates ∧ operators ∧ content slots)

satisfies(Lingenic, language neutral)
∵ invariant across(scaffold(Lingenic), ∀L ∈ human languages)

Precedent

precursor(Lingenic, Knuth's pseudocode) ≜ {
    insight: combine(formal notation, natural language content),
    domain: algorithms ∧ English,
    reader: human : {¬scales, ¬multilingual}
} ⟕ {
    src: Knuth, Donald E., The Art of Computer Programming Vol. 1: Fundamental Algorithms,
    publisher: Addison-Wesley,
    year: 1968
}

Knuth's pseudocode : formal(code) ∧ content(English) ⟕ {reader: human}
Lingenic : formal(∀formal notation systems) ∧ content(∀L ∈ human languages) ⟕ {reader: competent reader}
generalizes(Lingenic, Knuth's pseudocode) ∵ superset(scope(Lingenic), scope(Knuth's pseudocode))

Conclusion

error(Leibniz) ≜ assumed(meaning ∈ symbols)
correct(Lingenic, error(Leibniz)) ∵ meaning ∉ symbols ∧ emerges(meaning, between(writer, reader))

goal(Leibniz) = goal(unambiguous)
right(Leibniz, goal(Leibniz))
wrong(Leibniz, architecture(Leibniz))

architecture(Leibniz) ≜ meaning ∈ symbols ∧ minimize(reader) ⟕ {reader: mechanical}
architecture(Lingenic) ≜ meaning ∈ reading ∧ maximize(reader) ⟕ {reader: competent}

¬(realizes(Lingenic, goal(Leibniz)) ∧ ¬corrects(Lingenic, error(Leibniz))) ⟕ {not merely}
corrects(Lingenic, error(Leibniz)) ∧ realizes(Lingenic, goal(Leibniz))

achieves(Lingenic, goal(Leibniz))
achieves(Lingenic, max(achievable))
∵ correct architecture ∧ competent reader ∧ formal notation ∧ natural language content

Bibliography

bibliography ≜ [
    Leibniz 1666      ≜ "Dissertatio de Arte Combinatoria",
    Leibniz 1684      ≜ "Nova Methodus pro Maximis et Minimis",
    Newton 1687       ≜ "Philosophiæ Naturalis Principia Mathematica",
    Boole 1847        ≜ "The Mathematical Analysis of Logic",
    Cantor 1874       ≜ "Über eine Eigenschaft des Inbegriffes aller reellen algebraischen Zahlen" ⟕ {journal: Journal für die reine und angewandte Mathematik, volume: 77, pages: "258–262"},
    Peano 1889        ≜ "Arithmetices Principia, Nova Methodo Exposita",
    Frege 1879        ≜ "Begriffsschrift",
    Russell ∧ Whitehead 1910-1913 ≜ "Principia Mathematica",
    Lewis 1918        ≜ "A Survey of Symbolic Logic",
    Gödel 1931        ≜ "Über formal unentscheidbare Sätze der Principia Mathematica und verwandter Systeme I" ⟕ {journal: Monatshefte für Mathematik und Physik, volume: 38, pages: "173–198"},
    Kolmogorov 1933   ≜ "Grundbegriffe der Wahrscheinlichkeitsrechnung",
    Church 1936       ≜ "An Unsolvable Problem of Elementary Number Theory" ⟕ {journal: American Journal of Mathematics, volume: 58, number: 2, pages: "345–363"},
    Church 1940       ≜ "A Formulation of the Simple Theory of Types" ⟕ {journal: Journal of Symbolic Logic, volume: 5, number: 2, pages: "56–68"},
    von Wright 1951   ≜ "Deontic Logic" ⟕ {journal: Mind, volume: 60, number: 237, pages: "1–15"},
    Wittgenstein 1953 ≜ "Philosophische Untersuchungen / Philosophical Investigations" ⟕ {translator: G. E. M. Anscombe, publisher: Blackwell, location: Oxford},
    Kripke 1959       ≜ "A Completeness Theorem in Modal Logic" ⟕ {journal: Journal of Symbolic Logic, volume: 24, number: 1, pages: "1–14"},
    Quine 1960        ≜ "Word and Object" ⟕ {publisher: MIT Press, location: Cambridge MA},
    Hintikka 1962     ≜ "Knowledge and Belief",
    Knuth 1968        ≜ "The Art of Computer Programming, Vol. 1",
    Codd 1970         ≜ "A Relational Model of Data for Large Shared Data Banks" ⟕ {journal: Communications of the ACM, volume: 13, number: 6, pages: "377–387"},
    Martin-Löf 1972   ≜ "An Intuitionistic Theory of Types" ⟕ {note: Preprint, Stockholm University},
    Lewis 1973        ≜ "Counterfactuals",
    Pratt 1976        ≜ "Semantical Considerations on Floyd-Hoare Logic" ⟕ {proceedings: 17th IEEE FOCS, pages: "109–121"},
    Pnueli 1977       ≜ "The Temporal Logic of Programs" ⟕ {proceedings: 18th IEEE FOCS, pages: "46–57"},
    Hoare 1978        ≜ "Communicating Sequential Processes" ⟕ {journal: Communications of the ACM, volume: 21, number: 8, pages: "666–677"},
    Harel 1979        ≜ "First-Order Dynamic Logic",
    Milner 1980       ≜ "A Calculus of Communicating Systems",
    Allen 1983        ≜ "Maintaining Knowledge about Temporal Intervals" ⟕ {journal: Communications of the ACM, volume: 26, number: 11, pages: "832–843"},
    Pearl 2000        ≜ "Causality: Models, Reasoning, and Inference"
]