Foundational Limits: Why BFO’s Aristotelian Framework Cannot

Model Modern Science

Michael DeBellis

michaeldebellis.com, San Francisco, CA, U.S.A.

Keywords: Basic Formal Ontology (BFO), Philosophy of Science, Physics, Quantum Mechanics, Wave-Particle Duality.

Abstract: The Basic Formal Ontology (BFO) has gained widespread adoption in the biomedical domain and is

increasingly promoted as a domain-neutral upper ontology suitable for all branches of science, engineering,

and business. Its design reflects a commitment to metaphysical realism rooted in Aristotelian distinctions,

particularly between Continuants (entities that persist through time) and Occurrents (entities that unfold over

time). While this approach has demonstrated utility in domains such as biology and medicine, it encounters

significant limitations when applied to more complex or foundational areas of science, such as quantum

physics. Using the example of the electron, whose ontological status defies classical categorization, I argue

that the BFO framework lacks the flexibility to accommodate the indeterminacy, contextuality, and non-

locality inherent in quantum theory. Bell’s theorem and the incompatibility between general relativity and

quantum mechanics further highlight the fragmented and model-dependent nature of contemporary science.

These challenges suggest that the search for a single upper model for all domains is based on a mistaken

assumption: that science shares a single unified ontology. I conclude that ontology design must acknowledge

the methodological and conceptual pluralism of science, and that attempts to enforce a single top-level

ontology risk obscuring rather than clarifying the structure of scientific knowledge.

INTRODUCTION

The Basic Formal Ontology (BFO) presents itself as

a realist, domain-neutral upper ontology a framework

intended to serve as a universal foundation for all

scientific and engineering knowledge (Arp, Smith, &

Spear, 2015). It has been adopted in numerous applied

ontology projects, especially in the biomedical

domain, and is widely viewed as a principled

alternative to ad hoc modeling approaches. But BFO

is not merely a technical tool; it is a metaphysical

commitment. It is rooted in a classical philosophical

tradition based on the Aristotelian worldview that

assumes the world consists of discrete, identifiable

entities with essential properties, organized in a

coherent, hierarchical structure.

This paper challenges the assumption that such a

framework can serve as a universal foundation for

science. Specifically, I argue that BFO’s core

metaphysical assumptions are incompatible with

modern physics, and by extension, with any ontology

that seeks to model scientific knowledge in a general

https://orcid.org/0000-0002-8824-9577

and foundational way. Using examples from quantum

mechanics such as the electron, I demonstrate that

BFO cannot faithfully represent key aspects of

modern science. This is not a minor technical

limitation; it is a categorical mismatch between the

ontology’s conceptual apparatus and the ontological

structure of the phenomena it aims to model. While I

focus on BFO, this critique applies to any approach

that claims one model can be the foundation for all

science and engineering.

My conclusion is that the entire concept of one

unified model that is a foundation for all of science is

neither possible, nor consistent with modern science.

Overview of BFO’s Conceptual Framework

BFO’s architecture is grounded in a two-tier

metaphysical division between continuants and

occurrents:

• Continuants are entities that persist through

time while maintaining their identity. They

are wholly present at any moment. Examples

DeBellis, M.

Foundational Limits: Why BFO’s Aristotelian Framework Cannot Model Modern Science.

DOI: 10.5220/0013771200004000

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 17th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2025) - Volume 2: KEOD and KMIS, pages

151-155

ISBN: 978-989-758-769-6; ISSN: 2184-3228

151

include physical objects, organisms, and

anatomical structures.

• Occurrents are entities that unfold over

time and have temporal parts. These include

processes, events, and activities.

BFO further refines these categories into subtypes

such as independent continuants (e.g., a cell),

specifically dependent continuants (e.g., a colour or

role), and processes (e.g., digestion or cell division).

It presumes that all scientific and engineering entities

can be consistently categorized within this

framework, and that the distinctions it encodes are

universal across all domains of scientific inquiry and

engineering design.

QUANTUM MECHANICS: A

CHALLENGE FOR BFO

Quantum mechanics presents an instructive challenge

to BFO’s metaphysics. On the surface, one might be

tempted to classify concepts such as the electron as an

independent continuant: a physical entity with well-

defined properties such as mass and charge, which

participates in various physical processes. However,

quantum theory resists this simplification in several

ways.

2.1 Identity and Persistence

BFO assumes that continuants possess identity

conditions: they are individuated entities that persist

over time. But in quantum mechanics, electrons are

fundamentally indistinguishable (Griffiths, 2017). In

entangled systems, electrons do not retain individual

identity, and it is often meaningless to ask which

particle is which (Saunders, 2003) (French & Krause,

2006). More fundamentally, the electron’s state is

represented by a wavefunction, which encodes a

probability distribution over all possible

configurations. The electron does not have a

determinate position or trajectory, or even a

determinate set of properties, independent of

measurement (Adams, 2013).

This violates BFO’s assumption that scientific

entities have clearly defined identity conditions

independent of observational context.

2.2 Wave-Particle Duality

In the double-slit experiment, electrons display wave-

like interference when not observed, and particle-like

behavior when measured. This is not merely a

limitation in our instruments; it reflects a fundamental

fact about how electrons exist (Feynman, Leighton, &

Sands, 1965). There is no single ontological category

— “wave” or “particle” — that captures what an

electron is. It behaves as both, depending on context.

BFO has no apparatus for modeling context-

dependent modality. It requires that entities have a

consistent ontological status regardless of

observation. Modeling an electron as a process or as

a role does not resolve the issue; it merely shifts the

problem without addressing the underlying

incompatibility.

2.3 Superposition and Measurement

Quantum superposition allows an electron to exist in

multiple states simultaneously. It is not simply that

we don’t know the state — the electron has no definite

state until measured. The act of measurement causes

a collapse of the wavefunction, producing a

determinate outcome from an indeterminate

condition.

This challenges the core BFO assumption that all

entities have definite properties at all times. BFO

assumes ontological realism in the classical sense:

that properties like location, energy, and identity exist

whether or not we observe them. Quantum theory

shows that this is not the case.

2.4 Bell’s Inequality and the Limits

of Classical Logic

One of the most striking demonstrations of the

disconnect between classical metaphysics and

modern physics is Bell’s Inequality. First formulated

by John Bell in 1964, the inequality was designed to

test whether the probabilistic nature of quantum

mechanics could be explained by so-called hidden

variables—underlying, unobserved properties that

determine the outcomes of measurements. Crucially,

Bell assumed that these properties would be local

(influencing only nearby events) and realist (existing

independently of observation). Under these

assumptions, Bell derived a mathematical

constraint—an inequality—that the correlations

between the outcomes of measurements on entangled

particles must obey.

KEOD 2025 - 17th International Conference on Knowledge Engineering and Ontology Development

152

However, quantum mechanics predicts, and

experiments have consistently confirmed that

entangled particles violate Bell’s Inequality. Their

correlations are stronger than any local hidden

variable theory can explain. This implies that no

model that simultaneously respects both locality and

realism can account for quantum phenomena. The

implication is profound: the world is not structured

according to classical expectations about separability

and pre-existing properties. The correlations

observed in entangled systems appear to arise non-

locally, or at the very least, in ways that are not

reducible to underlying logical constraints like

conjunction, subset inclusion, or definable identity

conditions.

To build intuition for what Bell’s Inequality rules

out, consider a common analogy used in teaching: if

we count the number of people in a room who are

from Massachusetts and also blonde, that number

must be less than or equal to the number of people

who are simply from Massachusetts (Adams, 2013).

This is a basic principle of set theory: the conjunction

of two properties always applies to a subset of those

to which either one applies individually. Bell’s result

shows that nature does not conform to this kind of

logic at the quantum level. The correlations between

measurements on entangled particles cannot be

explained by assuming they had pre-existing, local

values for each observable. The violation of the

inequality reflects a structural failure of classical

logic when applied to physical reality.

This poses a fundamental challenge to ontologies

like BFO, which assume a realist, logic-based model

of the world. BFO depends on the idea that entities

have intrinsic properties, that those properties exist

independently of observation, and that they can be

composed and constrained using classical logical

operators. But Bell’s Theorem—and the physics it

represents—shows that this model of the world is

untenable in the quantum domain. The assumptions

that underlie classical ontology are not just

incomplete; they are demonstrably incorrect when

applied to phenomena that are now central to our

scientific understanding of matter, energy, and

causality. As such, BFO’s framework is not merely

limited. It is metaphysically misaligned with some of

the most rigorously verified aspects of modern

science.

Of course, these issues apply to any particle in quantum

physics. I focused on an electron to make things simple.

CONCLUSION: THE LIMITS

OF A UNIVERSAL ONTOLOGY

AND AN ALTERNATIVE

APPROACH

At this point, one might reasonably ask: if BFO works

well for biological and medical ontologies, why does

its inability to model quantum mechanics matter?

3.1 The Fallacy of a Universal

Upper Model

The answer lies in BFO’s own ambition. It is

described as more than a practical tool for biomedical

data modeling. It presents itself as a formal upper

ontology. A universal foundation for science,

engineering, and business, applicable across all

science and engineering domains (Arp, Smith, &

Spear, 2015), (International Standards Organization

(ISO), 2021). For example, in (Arp, Smith, & Spear,

2015): “[the focus of BFO is] directed at providing a

description and explanation of the kinds of objects

and relations that are common to all scientific

domains.” And a bit further down (bold italics

added): “Where the biologist studies cells, the

chemist studies molecules, and the physicist studies

energy and electrons; the philosophical ontologist, in

contrast, is interested in giving an account of what is

common to cells, molecules, and electrons”

If BFO cannot model the most basic constituents

of matter in modern physics such as electrons

, then

its claim to ontological generality is fundamentally

undermined.

This is not just a failure of fit between a specific

model (e.g., electron) and a particular upper ontology.

It points to a deeper problem: modern physics reveals

ontological structures that are incompatible with the

assumptions of classical logic and metaphysics (aka

common sense). The central role of superposition,

entanglement, and contextual measurement in

quantum mechanics breaks the foundational

principles of classical First-Order Logic (FOL). For

example, the Kochen–Specker Theorem shows that it

is impossible to assign consistent truth values to all

quantum propositions in a non-contextual way. Bell’s

Theorem and its experimental validations

demonstrate that no local hidden-variable theory, i.e.,

no classical theory with deterministic, observer-

independent properties, can reproduce the predictions

of quantum mechanics. Even the logical structure of

Foundational Limits: Why BFO’s Aristotelian Framework Cannot Model Modern Science

153

quantum theory differs: the lattice of propositions in

quantum mechanics is non-distributive, unlike the

Boolean logic that underlies FOL and ontologies like

BFO (French & Krause, 2006).

In short, classical logic and the ontological

commitments it encodes, cannot serve as a universal

framework for all scientific theories. Quantum theory

requires a fundamentally different logical and

metaphysical approach. This point is not merely

philosophical: it has been rigorously demonstrated

through decades of theoretical and experimental work

(French & Krause, 2006).

Furthermore, even physics lacks a unified

ontological framework. General relativity and

quantum mechanics are both successful in their

respective domains, but they are not compatible

(Smolin, 2001). Attempts to unify them — such as

string theory or loop quantum gravity — remain

speculative. The structure of modern science is

pluralistic, not unified. Different domains require

different assumptions about space, time, identity, and

causality.

In this light, the idea of a single, coherent upper

ontology that can serve as a foundation for all of

science is deeply questionable. It reflects an outdated

philosophical worldview in which all knowledge

could be ordered into a single hierarchy of universals.

But the reality of scientific practice reveals a more

fragmented, contextual, and often contradictory

landscape. Modern science often consists of diverse

irreconcilable models.

This is elegantly expressed by Stephen Hawking:

“a [scientific] theory is just a model of the

universe, or a restricted part of it, and a set of rules

that relate quantities in the model to observations that

we make. It exists only in our minds and does not have

any other reality (whatever that might mean). A

theory is a good theory if it satisfies two

requirements: It must accurately describe a large

class of observations on the basis of a model that

contains only a few arbitrary elements, and it must

make definite predictions about the results of future

observations.”

As an example outside of physics, both

population dynamics and evolutionary game theory

have been used extensively to model the evolution of

traits in populations, yet they rely on different

mathematical frameworks and make different

assumptions about causality and interaction.

Population dynamics traditionally employs

differential equations to model aggregate population

changes (Murray, 2002), while evolutionary game

theory uses payoff matrices and concepts like

Environmentally Stable Strategies (ESS) to analyse

the behaviour of organisms as strategies in a game

theory model (Smith, 1982). Despite addressing

highly overlapping domains, these models are

currently not reconcilable much less capable of

reducing one to the other.

An elegant summary of how actual science works

comes from Alan Adams description of quantum

mechanics (Adams, 2013):

“it is a fact that, if you take this expression and

you work with the rest of the postulates of quantum

mechanics… you reproduce the physics of the real

world. You reproduce it beautifully. You reproduce it

so well that no other models have even ever vaguely

come close to the explanatory power of quantum

mechanics. OK? It is a fact. It is not true in some

epistemic sense. You can't sit back and say, ah a

priori starting with the integers we derive that p is

equal to -- no, it's a model. But that's what physics

does. Physics doesn't tell you what's true. Physics

doesn't tell you what a priori did the world have to

look like. Physics tells you this is a good model, and

it works really well, and it fits the data. And to the

degree that it doesn't fit the data, it's wrong. OK? This

isn't something we derive. This is something we

declare. We call it our model, and then we use it to

calculate stuff, and we see if it fits the real world.”

3.2 An Alternative Approach

A position paper is not the place to provide an

alternative upper model; however, I will at provide

some suggestions. For business ontologies, there are

usually ontologies such as FIBO (EDM Council,

2024) that are the consensus of leaders in the domain

and are the best foundation for that domain. For

business domains that don’t have such a curated

foundation, the Gist (Blackwood, 2020) upper model

from Cambridge Semantics is a good foundation.

For scientific ontologies a good starting point is

to look at what practitioners across various disciplines

have done and to the extent that there are

commonalities, extract relevant entities from other

curated ontologies from organizations such as the

W3C and Dublin Core. This is something I have done

with an ontology called Basic Reusable Ontology

(BRO) (DeBellis, 2025). These are properties such as

dct:creator and skos:altLabel that I routinely add to

any ontology that I develop. I created this ontology

for my own use and only offer it as an example. I think

a very worthwhile project would be to have a

standards group that defines such a basic upper model

that focuses primarily on meta data as well as a few

classes that are in most ontologies. The best approach

would be a layered set of ontologies, starting with the

KEOD 2025 - 17th International Conference on Knowledge Engineering and Ontology Development

154

most basic metadata concepts and adding additional

basic concepts such as :has_part and :Person to larger

models where each larger model is a superset of the

entities in the previous model. Such a group could

provide a leaner alternative to Gist. E.g., by including

classes from Prov-O such as Agent and Organization

(Prov W3C Working Group, 2013).

ACKNOWLEDGEMENTS

Thanks to Robert Rovetto for valuable feedback on

early versions of this paper. Thanks to Dr. Robert

Neches for discussions that led to this paper.

REFERENCES

Adams, A. (2013). MIT 8.04 Quantum Physics I: Lecture 2:

Experimental Facts of Life. (MIT Open Courseware)

Retrieved 7 1, 2025, from https://www.youtube.com

/watch?v=U6fI3brP8V4&t=2060s

Arp, R., Smith, B., & Spear, A. D. (2015). Building

Ontologies with Basic Formal Ontology. Cambridge,

MA, USA: MIT Press.

Blackwood, P. (2020, November 3). A Brief Introduction to

the Gist Semantic Model. (Semantic Arts) Retrieved 2

19, 2025, from semanticarts.com: https://www.sema

nticarts.com/a-brief-introduction-to-the-gist-semantic-

model/

DeBellis, M. (2025, 8 6). Basic Reusable Ontology (BRO).

Retrieved from github.com: https://github.com/mde

bellis/Basic_Reusable_Ontology

EDM Council. (2024). The Financial Industry Business

Ontology. (EDM Council) Retrieved April 30, 2024,

from https://spec.edmcouncil.org/fibo/

Feynman, R. P., Leighton, R. B., & Sands, M. (1965). The

Feynman Lectures on Physics (Vol. 3: Quantum

Mechanics). Boston, MA, USA: Addison-Wesley.

French, S., & Krause, D. (2006). Identity in Physics: A

Historical, Philosophical, and Formal Analysis.

Oxford, England: Oxford University Press.

Griffiths, D. J. (2017). Introduction to Quantum Mechanics

(3rd ed.). Cambridge, MA, USA: Cambridge

University Press.

International Standards Organization (ISO). (2021).

ISO/IEC 21838-2:2021 Information technology — Top-

level ontologies (TLO) Part 2: Basic Formal Ontology

(BFO). (International Standards Organization (ISO))

Retrieved 7 1, 2025, from https://www.iso.org/stand

ard/74572.html

Murray, J. D. (2002). Mathematical Biology I: An

Introduction (3rd ed.). New York, NY USA: Springer.

Prov W3C Working Group. (2013, April). PROV-

Overview. (P. Groth, L. Moreau, Editors, & W3C)

Retrieved May 8, 2023, from www.w3.org/:

https://www.w3.org/TR/2013/NOTE-prov-overview-

20130430/

Saunders, S. (2003). Physics and Leibniz’s Principles. In

Symmetries in Physics: Philosophical Reflections.

Cambridge, MA, USA: Cambridge University Press.

Smith, J. M. (1982). Evolution and the Theory of Games.

Cambridge, UK: Cambridge University Press.

Smolin, L. (2001). Three Roads to Quantum Gravity. New

York, New York, USA: Basic Books.

Foundational Limits: Why BFO’s Aristotelian Framework Cannot Model Modern Science

155