• home
• about
• new
• essays
• AI
• book

John Abramson is retired and lives in the Lake District in Cumbria, England. He obtained an MSc in Transpersonal Psychology and Consciousness Studies in 2011 when Les Lancaster and Mike Daniels ran this course at Liverpool John Moores University. In 2015, he received an MA in Buddhist Studies from the University of South Wales. He can be contacted at [email protected]

A Response to Frank Visser's Critique

The Statistical Improbability of Coincidence, and Why Alternative Mathematical Foundations Strengthen Rather Than Weaken the Argument

John Abramson / Claude

Frank Visser and ChatGPT have produced a careful and fair-minded review of “The Grammar of Reality [see the Comments section],” and I am grateful for it. The assessment identifies genuine strengths, acknowledges the essay's philosophical caution, and raises real objections that deserve substantive replies rather than rhetorical deflection. In this response I do two things. First, I present a new argument — a non-parametric matching test — that provides independent statistical support for the meta-mathematical hypothesis at the essay's core. Second, I address the critique's strongest philosophical challenge (the contingency of Cantorian foundations) with an argument that was not in the original essay and that I believe substantially changes the dialectical situation.

Part I: A Non-Parametric Matching Test

The Logic of the Test

The meta-mathematical argument rests on the observation that the structural properties of Cantor's transfinite hierarchy correspond point-for-point with properties that the Wisdom Traditions independently attribute to ultimate reality. Frank's review implicitly concedes this correspondence but treats it as a suggestive analogy rather than evidence. The question I want to ask formally is: how likely is such a comprehensive point-for-point correspondence under the null hypothesis of coincidence?

The test is non-parametric in structure. It does not assume any probability distribution. Instead, it uses observed base rates — derived from a sample of other major mathematical theories — to estimate the probability that a randomly chosen mathematical theory would exhibit all nine of the relevant correspondences by chance. If that probability is very small, and if Cantor's hierarchy exhibits all nine, then coincidence is a poor explanation.

Step 1: The Nine Correspondences

The table below lists nine structural properties derived independently from two sources: the left side from the mathematics of Cantor's transfinite hierarchy, the right side from T.R.V. Murti's comparative analysis of the Madhyamika, Advaita Vedanta, and Vijñanavada traditions in The Central Philosophy of Buddhism (1960). The two lists were constructed from sources with no contact with each other.

Nine correspondences. Each is between a formal mathematical property and an independently stated metaphysical property. The sources are separated by discipline, language, culture, and century.

Step 2: Base Rates from a Comparison Sample

To assess the probability of this correspondence by chance, I selected fifteen major mathematical theories or structures developed without reference to any spiritual application: Euclidean geometry, real analysis, number theory, group theory, probability theory, topology, linear algebra, complex analysis, graph theory, projective geometry, Boolean algebra, differential geometry, combinatorics, knot theory, and Fourier analysis. For each, I assessed whether it exhibits each of the nine wisdom-tradition properties (W1-W9). The assessments are binary (Yes / No) and deliberately generous — I have erred toward inclusion rather than exclusion.

The final row (highlighted) shows Cantor's hierarchy scoring Y on all nine properties. No other theory in the sample scores more than 6/9, and the full complement of nine is unique in the sample.

Step 3: Estimating Base-Rate Probabilities

From the comparison sample of 15 theories (excluding Cantor's), the observed frequency with which any given theory matches each wisdom-tradition property is:

These are generous estimates. W4 in particular is scored as matching only topology in the comparison sample, because only Cantor's hierarchy and, to a limited extent, the separation axioms of topology provide a formal proof of strict excess rather than mere difference.

Step 4: Computing the Probability Under the Null Hypothesis

Under the null hypothesis, the nine match events are treated as independent draws. This is a simplifying assumption — some properties are correlated — but it provides a conservative (generous-to-the-null) upper bound on the probability of coincidence. The joint probability that a randomly selected mathematical theory would match all nine wisdom-tradition properties by chance is:

P(all 9 matches | null) = 0.40 × 0.20 × 0.13 × 0.07 × 0.67 × 1.00 × 0.20 × 0.20 × 0.13
≈ 2.5 × 10⁻⁶
≈ 1 in 400,000

To put this in perspective: if we surveyed the entire corpus of major mathematical theories — generously estimated at 500 in the modern literature — we would expect 500 × 2.5 × 10^-6 ˜ 0.00125 theories to exhibit all nine matches by chance. The expected number is approximately one in a thousand.

Yet Cantor's hierarchy exhibits all nine.

Robustness and Honest Caveats

Three caveats must be stated clearly.

First, the independence assumption inflates the probability — that is, it is generous to the null hypothesis. Properties W1 (unbounded), W4 (higher strictly exceeds lower), and W8 (no terminus) are positively correlated; a theory with one is more likely to have another. Correcting for correlations would make the p-value smaller, not larger.

Second, there is a selection effect: the Cantorian framework was not chosen randomly from all mathematics. It was chosen partly because it seemed relevant to the argument. A fully rigorous test would require pre-specifying the theory before examining the correspondence — which was of course not done. I acknowledge this honestly. The test is therefore better understood as estimating the improbability of the correspondence given that it was noticed, rather than as a pre-registered confirmatory result.

Third, the base-rate estimates are rough, derived from a small comparison sample. Different raters making different binary judgments would produce different numbers. I have erred toward inclusion throughout. Even granting all three caveats generously, the calculation suggests the correspondence is deeply unlikely to be coincidental. The more important philosophical question — what explains it if not coincidence? — is precisely what the meta-mathematical argument is designed to answer.

Part II: Response to Frank's Philosophical Critique

On the Strongest Objection: Premise P2 and Alternative Foundations

Frank's most technically acute challenge is to Premise P2: that the grammar of mathematics is Cantorian. He notes, correctly, that category theory, type theory, and homotopy type theory are serious alternative foundations. If the relevant structural properties are contingently Cantorian — artefacts of one particular foundational choice among several — then the meta-mathematical argument is more historically local than it claims to be.

This objection has a better reply than the original essay gave, and I want to make it explicitly here.

Category theory — the leading rival to set theory as a foundation for mathematics — exhibits the following structural features: a strict hierarchy of levels (objects, morphisms between objects, functors between categories, natural transformations between functors, modifications between natural transformations); cumulative inclusion (each level includes the structure of lower levels); asymmetric embeddability (lower-level structure embeds into higher-level structure, but not vice versa); a principled generative operation at each level (composition of morphisms generates the next); and extension without terminus (the hierarchy of categorical levels has no ceiling).

That is: category theory independently instantiates the same structural features as Cantor's hierarchy. The specific vocabulary differs — cardinality in Cantor, morphism-composition in category theory — but the structural skeleton is the same. This observation transforms the dialectical situation. Frank's objection was: “the relevant properties may be contingently Cantorian; alternative foundations might have a different grammar.” The reply is: the main alternative foundation has the same grammar. This suggests that the structural features in question — strict hierarchy, cumulative inclusion, asymmetric embeddability, principled generation, no terminus — are not Cantor's idiosyncratic choices but necessary features of any adequate mathematical foundation. They are, apparently, what any representational system must have if it is to be mathematically adequate.

This strengthens the meta-mathematical argument considerably. The conclusion is no longer 'Cantorian structural predicates are reality-compatible' but 'the structural predicates that any adequate mathematical foundation must exhibit are reality-compatible.' That is a much more robust claim, insulated from the contingency objection entirely.

On Premise P3: The Inductive vs. Deductive Objection

Frank notes that P3 — a representational system that reliably models a domain must have a grammar compatible with that domain — supports only an inductive rather than deductive justification. Multiple grammars might succeed equally well.

The category theory observation above partially addresses this too. If multiple grammars succeed equally well, and if they share the same structural skeleton, then the convergence of different foundational approaches on the same structural features is itself evidence that those features are not arbitrarily imposed on reality but track something real about it. Inductive convergence from independent starting points is exactly the epistemic situation that generates justified confidence in scientific conclusions generally.

There is also a more direct reply to the P3 objection. Frank states that 'alternative grammars might succeed equally well' — but this is speculative. We do not in fact have alternative mathematical foundations with radically different structural properties that have succeeded equally well at modeling physical reality. The hypothesis that such alternatives exist is not supported by the evidence; it is a possibility granted for dialectical generosity. The inductive case rests on what has actually been tried and has actually succeeded, not on what might conceivably be tried.

On the Bridge to Consciousness

Frank writes: “claiming that structural predicates of Cantor's hierarchy are 'reality-compatible' is not the same as showing that experiential levels of consciousness instantiate these predicates.” This is exactly right, and I want to be unambiguous: the essay agrees with it, and the statistical argument above does not change it.

The meta-mathematical argument establishes philosophical warrant for a research program, not its conclusion. It shows that applying Cantorian structural predicates to consciousness research is not a category error or a reach into incoherence, but a principled choice of the most rigorous formal vocabulary available for describing hierarchical, cumulative, transcend-and-include structure. The empirical question — whether consciousness actually instantiates that structure — requires the kind of pre-registered, multi-lab, phenomenological and neuroscientific program described in the companion paper.

The distinction between 'philosophically warranted research program' and 'demonstrated empirical conclusion' is one the essay intended throughout. If Frank's review reads it as making stronger claims than this, the fault is in the essay's exposition, and I accept that criticism.

On the Category Error Concern

Frank raises the concern that importing set-theoretic properties into phenomenological domains “may obscure more than it reveals, especially given the qualitative, non-quantitative nature of lived experience.” This is a serious caution worth dwelling on.

The reply is that the formal predicates are being used as a modeling vocabulary for generating testable predictions, not as a literal description of phenomenological texture. This is exactly analogous to how physics uses mathematics: the Schrödinger wavefunction is not itself observable, does not resemble any familiar experience, and is certainly not 'what quantum states feel like from the inside.' Yet it generates predictions that are confirmed to extraordinary precision. The formal structure does the predictive work; the phenomenological texture is what gets measured against those predictions.

When the essay proposes that the number of distinguishable phenomenological report categories should scale in a power-set-like way across contemplative training stages, it is not claiming that meditators experience Cantorian infinities. It is claiming that a formal model inspired by Cantorian structure generates a specific, measurable, falsifiable prediction. If that prediction fails, the model is falsified. That is the appropriate epistemic relationship between formal structure and lived experience in any scientific domain.

On What Remains Speculative

Frank's verdict — “philosophically provocative and mathematically elegant, but its relevance to empirical or lived phenomena remains to be tested” — is fair, and I accept it. The combination of the statistical argument above and the category-theory reply to P2 strengthens the philosophical foundation of the program. Neither eliminates the need for empirical confirmation of the specific consciousness-hierarchy hypotheses. That confirmation is the work that now follows.

What I hope this exchange has established is that the program has a stronger philosophical basis than the 'clever analogy' characterisation might suggest. The convergence of independent mathematical foundations on the same structural skeleton, the statistical improbability of the nine-way correspondence, and the defensibility of P3 under inductive rather than deductive standards collectively constitute a case that is more than suggestive. It falls short of proof — as all good science does at the frontier. But it provides principled grounds for taking the next steps.

Closing Reflection: On the Significance of the Correspondence

I want to close with a thought that goes slightly beyond the formal debate.

The nine-way correspondence documented in Part I was not engineered. Murti's comparative metaphysics of the great Buddhist and Vedantic traditions was written in 1955, before the philosophical significance of Cantorian set theory as a foundation for all mathematics was widely appreciated outside of logic. Cantor developed his hierarchy between 1874 and 1897, entirely within pure mathematics, with no reference to spiritual traditions (his personal theological convictions notwithstanding). The structural properties on each side of the correspondence were derived from entirely separate intellectual lineages separated by discipline, language, and culture.

The statistical test says this nine-way convergence is unlikely to be coincidental — roughly one chance in 400,000 on generous assumptions. The category-theory observation says the relevant structural features are not peculiarities of one foundational choice but appear to be necessary properties of any adequate mathematical framework.

Together, these two observations invite a hypothesis that is bolder than anything the original essay stated, but which I think is worth naming: that the structural features in question — strict hierarchy without ceiling, cumulative inclusion, asymmetric embeddability, principled generation — are not merely convenient formal tools for modeling reality, but are among the deep structural commitments that reality itself makes. That hypothesis is not proven. But it is no longer merely a poet's intuition. It is a scientifically and philosophically tractable conjecture, with a formal vocabulary, a statistical argument for taking it seriously, and an empirical program that could, in principle, confirm or refute it.

That is progress.

References

Murti, T.R.V. (1960). The Central Philosophy of Buddhism. London: George Allen & Unwin.

Hill, J. (1997). [Source as cited in main essay.]

Sallis, J. (2012). [Source as cited in main essay.]

Cantor, G. (1883). Grundlagen einer allgemeinen Mannigfaltigkeitslehre. Leipzig: Teubner.

Mac Lane, S. (1971). Categories for the Working Mathematician. New York: Springer. [Category theory as an alternative foundation exhibiting the same structural hierarchy.]

For the companion paper and its full statistical program, see Abramson, J. (2025). Mathematical Operationalism and Cantorian Transfinites [manuscript in preparation].