This is the draft of a paper to be published. It develops the ideas already presented in the paper called "The Sceptical Deal with the Concept of Reality", published in the book called LINES OF THOUGHT (CSP 2014)
Two Concepts of Reality and the Dissolution of Scepticism
Summary:
This paper argues that external
world scepticism arises from an unnoticed ambiguity in the notion of reality.
Once we distinguish inherent reality – the status of experience as genuine from within a world – from adherent reality – the status a world has relative to other possible worlds – the sceptical demand for demonstrative proof is revealed as conceptually
misguided. Inherent reality is demonstrable, since it is given in the
structural features of experience and extends, via causal integration, to both directly
and indirectly accessible entities. Adherent reality, by contrast, is not
demonstrable, because establishing it would require comparative access to
multiple worlds, a standpoint our epistemic situation rules out. The classical
sceptical dialectic collapses because it presupposes a univocal notion of
reality and attempts to meet an incoherent demand. What remains is the Realist Presupposition (RP): in the absence of evidence for a sceptical scenario, we are entitled to
assume that the external world is both inherently and adherently real. RP is
not a metaphysical thesis but a rational orientation grounded in explanatory
economy and the conditions under which doubt is intelligible.
Keywords: reality; dissolution of
scepticism; rational presupposition; external world scepticism; explanatory
economy; hinge epistemology; conceptual analysis.
The problem of external‑world scepticism has long been regarded as one of
the deepest and most persistent challenges in epistemology. Its force derives
from a simple but powerful thought: our experience is compatible with radically
different underlying realities, and nothing within experience itself seems
capable of ruling out these alternatives. From this, the sceptic concludes that
we cannot know that the external world is real (Descartes 1641, Meditation I).
By contrast, the classical anti‑sceptical tradition (Reid 1764; 1785) maintains
that we do know this and that sceptical alternatives are not genuine
epistemic possibilities but artefacts of a mistaken theory of perception.
Despite centuries of reformulations, refinements, and attempted dissolutions,
the dialectic continues to exert pressure on our epistemological self‑understanding.
This paper argues that the dialectic persists only because it rests on a
conceptual mistake. The sceptical problem arises from a confusion in the
underlying concept of reality. It depends on an unnoticed ambiguity: sceptical
arguments presuppose that the attribution of reality constitutes a single
epistemic status that the world either possesses or fails to possess, and anti‑sceptical
responses typically accept this assumption while disputing the sceptic’s
conclusion. I argue that this presupposition is false. The notion of reality is
not univocal. It divides into two fundamentally different concepts – inherent reality and adherent reality – each with its own epistemic profile, criteria, and conditions of
assessment. Once this distinction is made explicit, the sceptical problem is
transformed. The sceptic’s demand for demonstrative proof is revealed as
conceptually misguided, and the familiar anti‑sceptical responses are shown to
inherit the same mistake. In this way, the entire dialectic collapses.
The argument of this paper is not a variation on familiar anti‑sceptical
strategies. It does not rely on denying closure principles (Dretske 1970;
Nozick 1981), on contextualist manoeuvres (DeRose 1995; Williams 1996; Zhao
2026), on disjunctivist accounts of perceptual knowledge (McDowell 1994), on
externalist theories of warrant (Britt 2026), on modal dissolutions of
sceptical reasoning (Bangu 2018), or on transcendental arguments (Strawson
1959). Nor does it attempt to defeat scepticism by claiming that sceptical
hypotheses are probabilistically negligible (Vogel 1987), explanatorily
inferior (Lycan 2001), semantically incoherent (Putnam 1981), or pragmatically
irrelevant to our epistemic practices (Stroud 1984). The approach developed here
is of a different kind.
Instead, it proceeds by conceptual analysis. It shows that the sceptical
problem arises only because we have failed to distinguish two fundamental
senses of reality that play different roles in our epistemic lives. Once the
distinction is in place, the sceptical demand for demonstrative certainty is
exposed as a demand for something that cannot be provided – not because our
epistemic powers are limited, but because the demand itself is conceptually
incoherent.
The distinction between inherent and adherent reality reflects a
structural feature of our epistemic situation. Inherent reality concerns the status of experience within a world. It is marked by vividness, involuntariness, intersubjective
checkability, law‑governed regularity, and contextual coherence. These features
converge so strongly that the attribution of inherent reality is forced upon us
by the very conditions of intelligible experience. Their scope extends beyond
direct perception: through causal integration, they apply equally to indirectly
accessible entities such as remote nebulae, bacteria, molecules, and subatomic
particles. Properly understood, these criteria yield a straightforward proof of
the external world; a result that stands in contrast with Carnap’s defeasible
view that the existence of the external world (the “thing world”) is a
pragmatic fiat (1950: §2). Inherent reality is demonstrable because it is given in the structure of
experience itself.
Adherent reality, by contrast, concerns the status of a world relative to other
possible worlds. It is a comparative notion: a
world is adherently real when it stands in the appropriate relation to other
worlds that might exist. But adherent reality cannot be demonstrated because
demonstration would require access to at least one additional world with which
ours could be compared. Such access is conceptually impossible, except in
hypothetical sceptical scenarios whose very possibility remains indemonstrable.
This conceptual point has profound epistemological consequences. It
shows that the sceptical demand for proof of external reality in the adherent
sense is not merely unmet; it is misconceived. The sceptic asks for a kind of
proof that the concept of adherent reality itself excludes. The anti‑sceptic,
in attempting to meet this demand, inherits the same mistake. Both sides of the
traditional dialectic assume that reality has a single epistemic status. Both
sides fail for the same reason.
Once the conceptual confusion is removed, we can articulate the rational
stance that remains. Call it the Realist Presupposition (RP):
RP: In the absence of any evidence for a sceptical scenario, we are rationally
entitled to assume that the external world is both inherently and adherently
real.
RP is not a metaphysical thesis. It is not a claim about the ultimate nature of reality. It is a rational
orientation forced by the structure of our epistemic situation. It replaces the
misguided demand for demonstrative proof with a stance that is neither dogmatic
nor inferential. It is grounded in explanatory economy (Russell 1912:
26–27) and in the hinge‑epistemological insight that doubt requires
evidential motivation (Wittgenstein 1969). RP dissolves the sceptical
problem rather than refuting it.
The structure of the paper is as follows. Section I articulates the
distinction between inherent and adherent reality and shows how it reorganises
the sceptical dialectic. Section II demonstrates the collapse of the classical
sceptical and anti‑sceptical arguments once the distinction is in place.
Section III clarifies the Realist Presupposition. Section IV vindicates the
Realist Presupposition. Section V concludes by drawing out the broader
implications of the view.
The resulting picture is architectonic. It reorganises the
epistemological landscape rather than merely adjusting its contours. It shows
that scepticism is not a deep epistemic threat but a product of (deep)
conceptual confusion. Once the confusion is removed, the problem disappears.
I. Two Concepts of Reality
The sceptical problem arises only
because we treat reality as having a single
epistemic status. This assumption is so deeply entrenched in philosophical
discourse that it is rarely questioned. Yet once we examine the role the concept plays in our epistemic practices,
it becomes clear that reality is not univocal. It
divides into two fundamentally different notions – inherent reality and adherent reality – each with its own criteria, epistemic profile, and conceptual function. The
failure to distinguish these notions is the true source of the sceptical problem.
This section articulates the distinction and shows how it reorganizes
the epistemological landscape. The distinction is not merely terminological. It
reflects a structural feature of our epistemic situation. It is the conceptual
key that unlocks the sceptical dialectic.
1. Inherent Reality
We begin with the notion of inherent reality. This is the sense of reality operative from within a world; it concerns the status of experience as genuine rather than
illusory. Inherent reality is marked by a cluster of features that jointly
articulate the ordinary conception of “the real world.” These features may be
called criteria of reality, and they are constitutive of what we ordinarily mean by external reality. The
fundamental list is the following:
1. Vividness: perceptual
experiences present themselves with maximal intensity and richness, typically
sufficient to distinguish them from imagination or hallucination.
2. Involuntariness: the world imposes
itself upon us; its deliverances are not under our control.
3. Intersubjective
checkability: distinct subjects can poaaibly verify
the same states of affairs.
4. Law‑governed regularity: the world exhibits stable patterns that support prediction, explanation,
and scientific inquiry.
5. Contextual coherence: external reality displays synchronic and diachronic coherence with our
life histories and expectations; objects stand in recognisable relations to one
another, and their temporal trajectories are stable enough to sustain
intelligibility, continuity, and foreseeability.
These criteria need not be
continuously manifest in every moment of experience. A subject may temporarily
lack one or more of them while still living among inherently real things. A
drunken or half‑comatose person may fail to enjoy vivid experiences of the
objects before him; a castaway may live for years without any possibility of
intersubjective verification; someone abruptly transported into an utterly
unfamiliar environment may momentarily lack contextual coherence. These
absences, however, are local and contingent. They occur against a broader
experiential background in which the criteria are jointly satisfiable and, in
ordinary circumstances, jointly satisfied. Their occasional suspension does not
undermine inherent reality; they presuppose it. Such lapses are intelligible
only within a world whose general structure allows the criteria to converge,
reinforce one another, and persist across time.
Crucially, the criteria of reality are not independent. They must be jointly satisfied. They form
a mutually reinforcing pattern that persists across time, even across the
entire span of a human life. Their convergence is so
robust that the attribution of inherent reality is not optional; it is forced
upon us by the very conditions of intelligible experience. To deny inherent
reality would require rejecting the structural features that make experience
possible.
The necessity of treating the criteria jointly becomes clear once we
observe that each can be satisfied in isolation without warranting any
attribution of reality. Consider vividness: subjects
suffering from alcoholic hallucinosis may experience hallucinations whose
sensory force and clarity are indistinguishable from ordinary perception.
Intersubjective checkability can likewise be mimicked: a dream may present
itself as shared with other agents, each apparently confirming the same events.
Law‑governed regularity and contextual coherence admit similar counterexamples:
one may have a monotonously orderly dream in which nothing violates familiar
constraints of nature, or a dream that unfolds with perfect contextual
continuity.
These cases show that the satisfaction of any single criterion, or even
several in combination, does not compel the attribution of reality. What is
impossible, however, is the convergence of all five criteria in a way that
yields a stable and intelligible experiential structure. The point is not that
such convergence must persist for an entire lifetime. Rather, what matters is
that the criteria come together with sufficient stability, depth, and
explanatory coherence to form a unified pattern. Once the world supplies an
integrated account of the subject’s experiences – one in which vividness,
involuntariness, intersubjective checkability, regularity, and contextual
coherence mutually reinforce one another – the attribution of inherent reality
becomes unavoidable. In such circumstances, we are entitled, indeed forced, to
say: it is real; it was real: inherently real.
The criteria of inherent reality supply a reliable demonstration of the
existence of the external world, precisely what Kant lamented as the “scandal
of philosophy” in its absence. These criteria apply first to the things around
us: objects, properties, states of affairs, events, and processes. But their
scope is not confined to what is directly perceptible. We also apply them, indirectly,
to entities accessible only through instrumentation: remote nebulae, bacteria,
molecules, subatomic particles, and the forces of nature. Modern telescopes allow
us to detect primordial cosmic phenomena that would otherwise remain
inaccessible to observation. The trajectories
of electrons, protons, muons, and positrons, for example, become visible in a
cloud chamber through the condensation of supersaturated vapor. Such extensions
are justified because these entities are causally connected to observable
phenomena. If A causes B, and B invariably exhibits property R, it is
reasonable to extend R to A, mutatis mutandis.
A more decisive argument for this extension lies in the arbitrariness of
the human sensory range. To restrict reality to what human beings can perceive
would be mere anthropological chauvinism. Some animals see ultraviolet; others
see infrared. Migratory birds “see” magnetic fields. Bats and dogs perceive
ultrasound; blue whales detect infrasound that travels across oceans. It is
easy to imagine beings whose sensory modalities allow them to perceive
bacteria, molecules, or atoms directly—something we ourselves do, to some extent,
through olfaction. There is therefore no principled reason to confine inherent
reality to direct perception. Scientific realism is simply the recognition of
this fact.
By the way, once we attend solely to the criteria of inherent external
reality and understand that they may be satisfied either directly or indirectly,
we can even construct a straightforward proof of the external world:
1. The things presently
experienced around us satisfy the criteria of inherent external reality.
2. Many things experienced in the past satisfied these criteria and satisfied
them again when re‑experienced.
3. (Inductively from 1 and 2) Many things experienced in the past, though not
presently experienced, certainly satisfy the criteria of external reality.
4. We have continually
encountered new things that satisfy the criteria of external reality.
5. (Inductively from 4)
There must be many things never experienced that nonetheless satisfy the
criteria of external reality.
6. Testimony is, in general, a reliable means of attaining knowledge.
7. There is abundant
testimony concerning things we have never personally experienced, but that
satisfy the criteria of external reality.
8. (Deductively from 6
and 7) Many things we have not personally experienced are known to be real
through testimony.
9. (From 1, 3, 5, and 8)
There is a totality of things: (a) some presently experienced, which satisfy
the criteria of external reality; (b) many not presently experienced, but known
to satisfy those criteria insofar as they satisfied them in the past; (c)
others unknown, but which would satisfy them, given the continual appearance of
new things that do satisfy the criteria; (d) many never experienced by us, but
known to satisfy the criteria through testimony.
10. “Our external world”
is the sum total of things recognized as satisfying the criteria of external
reality.
11. (From 9 and 10) Our
external world satisfies the criteria of external reality; it exists; it is
(inherently) real.
This is why common sense regards the reality of the
external world as obvious, and why only a metaphysician, or someone out of his
wits, would doubt it. My suggestion is that this line of reasoning has, at some
point, been tacitly performed by each of us. Even a cave dweller, if asked
whether the world exists, would likely respond with a bored “yes.” Trivial as
it may seem, this sort of analysis can be philosophically useful as a critique
of metaphysical excess.
Inherent
reality is demonstrable. This point is crucial. Demonstrability here does not
mean that inherent reality is established by argument. It means that inherent
reality is given in the structure of experience itself. It is not inferred, hypothezised, or assumed; it is recognised. The sceptic
cannot coherently deny inherent reality, for the denial would undermine the very
conditions under which sceptical reasoning is possible.
This is the first major result of the distinction: inherent reality is
epistemically secure.
2.
Adherent Reality
The second notion is adherent reality. Whereas inherent reality concerns the status of experience from within a world, adherent reality concerns the status of a world relative to other
possible worlds. A world is adherently real when
it stands in the appropriate relation to other worlds that exist or might
exist—roughly, when it is the “actual” world rather than a simulation,
deception, or derivative construct.
A crucial point must be marked at the outset. Adherent reality cannot
be assessed from within a single world. Under ordinary circumstances only
one world is accessible to us, and no comparison is possible. The concept
becomes operationally meaningful only against the background of a sceptical
scenario; only when two inherently real worlds are simultaneously available for
comparison. This is not an epistemic limitation but a conceptual feature of the
notion itself.
It is conceivable that the joint satisfaction of the inherent criteria
of external reality is not immune to radical sceptical hypotheses. Dreams,
Cartesian demons, and brain-in-a-vat scenarios (Brueckner 1986) are at least logically conceivable. Suppose, for instance, that I awake in a strange place, surrounded by
unfamiliar beings who nonetheless resemble my newly implanted body. They explain, in fluent English,
that I had been a brain in a vat, running the program Philosophy Professor
on Planet Earth. This, they tell me, is a standard pedagogical procedure
on the planet Omega, designed to cultivate cognitive diversity among its
inhabitants. My brain has now been implanted into a healthy Omega body, and I
am free to live among them, equipped with the intellectual resources acquired
during the Earth‑program.
Let us further suppose that, after appropriate therapeutic adjustment, I
come to inhabit this new world without difficulty. It satisfies all the
inherent criteria of external reality: stable objects, coherent causal
structure, interpersonal interaction, and the absence of epistemic anomalies.
They show me the supercomputers, the vats, the pedagogical apparatus; I meet
others who were once brains in vats and who recount their own trajectories.
Over time, as I become acquainted with the Omega world and, by experiencing its
increasing contextual coherence, I come to regard it not merely as inherently real
but also as adherently real. The Earth world, though inherently real,
was fictional, its coherence being explained by its origin in the Omega world.
The choice of which world is adherently real in such a scenario is comparative and
explanatory. It is grounded not in the
inherent criteria of reality, which are shared by both worlds, but in the
greater explanatory coherence of treating one world as the source of the
other. The adherently real world is the one that contains the other as its
fictional product. In the skeptical scenario, the Earth world is thus explained
as a construct generated within the Omega world. Omega is therefore both
inherently and adherently real, whereas the Earth world—despite my former
conviction—was merely inherently real. Their shared inherency explains why, in
the comparison case, we are entitled to call both of them “real”: each
satisfies the internal criteria of external reality, even though only one
occupies the appropriate explanatory position relative to the other.
As long as no radical sceptical scenario arises, we assume that both
inherent and adherent reality obtain. We apply the Presupposition of Reality
(PR) – the assumption, grounded in parsimony and the absence of grounds for
doubt, that the world we inhabit is both inherently and adherently real – to
the Omega world, and not to the slowly forgotten Earth world. Only upon the
discovery of a sceptical scenario are we forced to compare worlds and determine
which is adherently real, based on the explanatory superiority of treating the
sceptical‑scenario world as the originating one.
The criteria
for the adherent irreality are not stable. Nothing guarantees that the Omega
world is the only adherently real world. Perhaps I am being manipulated by the inhabitants
of planet Zeta, and Omega is itself fictional. Or perhaps everything is an
extraordinarily vivid dream, and I awake again as a philosophy professor on
Earth, concluding that Earth is the only inherently and adherently real world. The situation resembles the film The Real Thing, in which characters move from one world to another and then to another.
In the final scene, after a shootout in a Western saloon, a tall black man
enters and asks: “Could someone tell me whether this is the real world?” The
question is intelligible only because multiple inherently real but adherently
unreal worlds have been traversed.
We can even imagine a world in which one lives as a brain in a vat and
knows it. Suppose that, for humanistic reasons, Omega punishes irredeemable
criminals by condemning them to spend their lives as brains in vats, where they
may continue their misdeeds, though somewhat unpleasantly aware that they
inhabit an adherently unreal world. The point is not empirical but conceptual: adherent irreality is a
comparative explanatory notion, applicable only when two or more inherently
real worlds are available for comparison.
These thought-experiments serve to show that adherent reality is a relational
concept which cannot be assessed from within a single world. Defending adherent
reality/irreality of a world would require access to at least one additional
world with which ours could be compared. Such access exists only in the
counterfactual situation in which a sceptical scenario becomes actual. Absent
that, our epistemic situation provides no standpoint from which its supposed
adherent reality/irreality could be evaluated. We inhabit one world, and only one world is ever accessible.
This point is not epistemic but
conceptual. It is not that we lack evidence or cognitive powers needed to
demonstrate adherent reality/irreality. It is that adherent reality is not the kind of thing that can be defended
under ordinary circumstances. The concept itself excludes
the possibility of demonstration. To demand proof of adherent reality as a
relational concept is to ask for something incoherent, not because our
epistemic powers are limited, but because the demand violates the concept’s
conditions of application.
This is the second major result of the distinction: as a relational
concept, adherent reality is not demonstrable, except in the hypothetical case
of a sceptical scenario, where it can be rationally justified.
II. Evaluating the Argumentative
Structures
Once the distinction between inherent and adherent reality is in place, the familiar sceptical demand for proof is exposed as
conceptually misguided. The sceptic asks for a kind of demonstration that the
very concept of adherent reality excludes. Under ordinary epistemic conditions,
adherent reality is not merely undemonstrable; it is inapplicable, because its assessment requires comparative access to more than one
world, and only one world is ever accessible. The anti‑sceptic inherits the
same confusion by attempting to provide such proof. The entire dialectic
collapses because its central requirement is incoherent.
This collapse is not a rhetorical manoeuvre. It follows from a deeper
point: the sceptical dialectic presupposes that reality names a single epistemic status. Without this presupposition, neither the
sceptic’s modus tollens nor the anti‑sceptic’s modus ponens can be convincingly
formulated. The distinction between inherent and adherent reality, therefore,
does not merely clarify the dialectic; it reorganizes it. It shows that scepticism is not a deep epistemic threat but a product
of conceptual confusion. To make this point precise, we must examine the
argumentative structures themselves and identify the implicit commitments
concerning reality on which they depend.
1. The Sceptic’s Modus Tollens
The familiar sceptical argument
is usually presented as:
1. If we know that the
external world is real, then we can rule out sceptical alternatives.
2. We cannot rule out sceptical alternatives.
3. Therefore, we do not know that the external world is real.
Its intuitive force derives from
a widely accepted epistemic requirement: knowledge demands the ability to
exclude relevant alternatives. If we cannot exclude the possibility that we are
dreaming, deceived, or simulated, then we cannot know that the external world
is real.
But the argument trades on an unnoticed ambiguity. The sceptic’s
conclusion is ambiguous between:
Inherent unreality (our world is dream‑like,
illusory, unreal), and
Adherent unreality (our world is not the “real”
world relative to other possible worlds).
Ancient scepticism typically
aimed at the first conclusion; modern scepticism aims at the second. Once we
make explicit the commitments concerning reality that the argument presupposes,
the argument aiming at the first conclusion becomes:
1. If we know that the external world is adherently real, then we can
rule out adherently real sceptical alternatives.
2. We cannot rule out adherently
real sceptical alternatives.
3. Therefore, we do not know that the external world is inherently
real.
This version is visibly invalid.
It shifts illicitly from adherent reality in the premises to inherent reality
in the conclusion. The sceptic’s conclusion does not follow.
Modern sceptics typically avoid this mistake by aiming only at adherent
unreality. Their argument must then be rendered as:
1. If we know that the external world is adherently real, then we can
rule out adherently real sceptical alternatives.
2. We cannot rule out adherently
real sceptical alternatives.
3. Therefore, we do not know that the external world is adherently
real.
This version is deductively
valid. And indeed, we cannot prove that our world is adherently real, since
adherent reality is not demonstrable. But this does not matter. The sceptic’s
conclusion is epistemically idle, since we have good reason to accept that our
world is adherently real through the Realist Presupposition (RP). What we do not have
reason to do is to reject the possibility of adherent unreality. RP licenses acceptance, not demonstrative exclusion.
The sceptic’s argument therefore fails to generate any epistemic threat.
It merely restates the conceptual point that adherent reality is not
demonstrable; a point that follows from the definition of adherent reality
itself.
2. The Anti‑Sceptic’s Modus Ponens
The anti‑sceptical response
simply reverses the sceptic’s reasoning:
1. If we know that the
external world is real, then we can rule out sceptical alternatives.
2. But we do know that
the external world is real.
3. Therefore, we can rule out sceptical alternatives.
At first glance, this seems
compelling. It captures the intuition that sceptical hypotheses are idle: we do
not seriously entertain the possibility that we are dreaming or deceived, and our
epistemic practices proceed without hesitation.
Its fully explicit form is:
1. If we know that the external world is adherently real, then we can
rule out adherently real sceptical alternatives.
2. But we do know that the external world is inherently real.
3. Therefore, we can rule
out adherently real sceptical alternatives.
The first premise must employ
adherent reality, since sceptical scenarios concern the comparative status of
worlds. The conclusion must also concern adherent reality, for the anti‑sceptic
aims to exclude the possibility that our world is adherently unreal. But the
second premise invokes inherent reality instead. The argument equivocates between the two notions of reality, and the
conclusion does not follow.
We can make the argument deductively
valid:
1. If we know that the external world is adherently real, then we can
rule out adherently real sceptical alternatives.
2. But we do know that the external world is adherently real.
3. Therefore, we can rule
out adherently real sceptical alternatives.
This argument is valid, but the
second premise is flawed. RP licenses acceptance of adherent reality, but it is
not strong enough to rule out the possibility that our world is not the
adherently real world in sceptical scenarios. The kind of knowledge vindicated
by RP is not conclusive knowledge.
The anti‑sceptic’s intended conclusion, that we can demonstratively rule
out sceptical alternatives, it is not merely unsupported; it is conceptually
meaningless. It presupposes a framework in
which adherent reality can be demonstrated from within a single world. But
adherent reality is not the kind of status that can be defended under ordinary
circumstances. The anti‑sceptic’s argument inherits the sceptic’s lack of
conceptual scrutiny.
3. The Dialectical Collapse
The sceptical and anti‑sceptical
arguments collapse for the same reason: both presuppose that reality is a
single epistemic status. Both demand demonstrative proof of reality. Both
assume that knowledge of reality requires the ability to exclude sceptical
alternatives. Both are mistaken.
This collapse is not a refutation
of scepticism but a dissolution. It shows that the sceptical problem arises
only because we have failed to distinguish two senses of reality that play
different roles in our epistemic lives. Once the distinction is in place, the
sceptical demand for demonstrative certainty of adherent reality loses its
force, and the anti‑sceptical attempt to provide such proof is revealed as
equally misguided.
III. A Deepened View of the Realist
Presupposition
The collapse of the classical
sceptical dialectic leaves a residual question: how should we rationally
proceed once we recognise that adherent reality cannot be demonstrated and, in
ordinary circumstances, cannot even be assessed? The sceptic concludes that we must suspend judgment. The anti‑sceptic
concludes that we must affirm the reality of the external world. Both
conclusions inherit the conceptual mistake that generated the sceptical problem
in the first place. Both assume that reality has a single epistemic status and
that knowledge of reality requires demonstrative proof. Once the distinction
between inherent and adherent reality is in place, neither conclusion can be sustained.
This section articulates the
rational stance that remains once the conceptual confusion is removed. The
stance is not inferential, evidential, or metaphysical. It is a presupposition
forced by the structure of our epistemic situation. As introduced earlier, it
is the Realist Presupposition (RP):
RP: In the absence of any evidence for a sceptical scenario, we are rationally
entitled to assume that the external world is both inherently and adherently
real.
RP is not a thesis about the
ultimate nature of reality. It is not a claim that the external world is
metaphysically fundamental, ontologically privileged, or uniquely actual. It is
not a hypothesis to be tested or a conclusion to be defended. RP is a rational
orientation, a stance that structures our epistemic practices once we recognize
the limits of demonstrative proof and the conceptual inapplicability of
adherent reality in ordinary circumstances.
1. What RP Is Not
To understand RP, we must first
clarify what it is not. These clarifications pre‑empt several familiar objections and prevent RP
from being misclassified.
RP is not dogmatic. It does not assert that sceptical alternatives are impossible, incoherent,
or self‑defeating. It does not claim that sceptical scenarios violate logical
or metaphysical constraints. RP does not foreclose sceptical possibilities; it
merely refuses to treat them as live in the absence of evidence.
RP is not metaphysical. It does not claim that our world is metaphysically fundamental or uniquely
actual. It does not attempt to settle the metaphysics of modality or the ontology
of possible worlds.
RP is not inferential. It does not rely on usual inferences to the best explanation,
probabilistic justification, or abductive reasoning. It is not a conclusion
properly drawn from evidence.
RP is not psychological. It is not a description of how
human beings happen to think. It is a normative stance
grounded in the conceptual structure of rational doubt.
These clarifications block the
objection that RP is merely pragmatism, merely dogmatism, or a disguised
metaphysical thesis. RP is none of these. It is a structural requirement of
rational epistemic practice once the inherent/adherent distinction is properly
understood.
2. The Conditional Form of RP
RP has a distinctive conditional
form:
In the absence of any evidence for a sceptical scenario…
This clause is essential. RP does
not assert that sceptical scenarios are impossible or incoherent. It asserts
only that sceptical hypotheses require evidential motivation. Without such
motivation, sceptical doubt is empty.
The second clause is equally
essential:
…we are rationally entitled to assume that the external
world is both inherently and adherently real.
This clause does not assert
that the external world is adherently real. It asserts only that we are rationally entitled to assume that it is.
The entitlement is not inferential, evidential, or metaphysical. It is a
presupposition allowed by the structure of our epistemic situation, more specifically,
by the fact that adherent reality cannot be applied unless a sceptical scenario
actually occurs.
RP, therefore, has a distinctive epistemic profile. It is neither a belief nor a hypothesis. It is a presupposition. It structures our epistemic practices without being
the conclusion of any argument.
This clarification blocks the
objection that RP is too weak (merely pragmatic) or too strong (dogmatic). RP
is neither. It is structurally allowed.
IV. Grounds for the Realist
Presupposition
RP, the realist presupposition,
is supported by two independent considerations that arise naturally once the
distinction between inherent and adherent reality is in place. The first
concerns explanatory economy. The second concerns the structure of rational
doubt. These considerations do
not prove RP – nothing could
prove RP, given the conceptual limits imposed by adherent reality. Rather, they show that RP is the rational stance to adopt once we recognize
those limits. They pre‑empt the objection that RP is arbitrary or unjustified.
1. Explanatory Economy
The first ground for RP concerns
explanatory economy. This point has historical roots in Russell’s observation
that positing an external world is the simplest explanation of our experience
(1912, chap. II). Jonathan Vogel (1990) later recast this insight using the
machinery of inference to the best explanation (IBE): the ordinary hypothesis, that
our world is the adherently real one, is explanatorily superior to any
sceptical alternative, which would require ad hoc stipulations to
preserve the appearances.
Our conceptual distinction clarifies the precise force of this point.
IBE does not prove adherent reality.
It cannot prove adherent reality, because
adherent reality is not demonstrable and, in ordinary circumstances, not even
applicable. But IBE does support the rational assumption that our world is
adherently real. It shows that the ordinary hypothesis is explanatorily
preferable to sceptical alternatives because sceptical hypotheses require
additional assumptions that are not motivated by experience, whereas the
ordinary hypothesis is simpler, more direct, and more economical.
This clarification blocks the familiar objection that IBE is
insufficient to refute scepticism. RP does not use IBE to refute scepticism. It
uses IBE to justify a presupposition.
2. The Structure of Sceptical
Hypotheses
This point can be made more
precise. A sceptical hypothesis must preserve the appearances. It must explain
why our experience exhibits vividness, involuntariness, intersubjective
checkability, and law‑governed regularity. It must explain why our epistemic
practices succeed, why our scientific theories are predictively successful, and
why our perceptual judgments are reliable. To do so, it must posit a complex
system of deception, simulation, or manipulation. It must posit a structure
that mirrors the structure of the external‑world hypothesis. It must posit
additional mechanisms to preserve the appearances.
The ordinary hypothesis that our
world is not only inherently but also (ceteris paribus) adherently real does
not require these additional assumptions. It posits an external world that
exhibits the structural features of experience and no deception, simulation, or
manipulation. It posits no additional mechanisms. This makes it explanatorily
superior. IBE therefore supports the rational
assumption that our world is adherently real without proving adherent reality or
meeting the sceptical demand for demonstrative certainty. It supports a
presupposition, not a conclusion.
This clarification blocks the objection that RP overreaches by
converting explanatory preference into knowledge.
3. The Structure of Rational Doubt
The second ground for RP concerns
the structure of rational doubt. This point has roots in Wittgenstein’s
observation that doubt is intelligible only against a background of certainty
(1969). A sceptical hypothesis requires evidential motivation. Without such
motivation, the doubt is empty. It cannot be meaningfully entertained.
Our conceptual distinction clarifies the precise force of this point.
Sceptical hypotheses concern adherent reality; they concern the status of our
world relative to other possible worlds; they require evidential motivation.
Without such motivation, they cannot be meaningfully entertained. They are not
live possibilities. They are idle.
This clarification blocks the objection that sceptical hypotheses can be
entertained without evidence.
4. The Inapplicability of Adherent
Reality
This point can be made more
precise. To doubt that our world is adherently real, we would need evidence
that our world is a simulation, deception, or inferior alternative, that is, we
would need the evidence that our world is not the “actual” world. We would need evidence that our world stands in the wrong relation to other
possible worlds. But we possess no such evidence:
no evidence of deception, simulation, or manipulation. Without such evidence, sceptical doubt is empty. It
cannot be meaningfully entertained. RP therefore remains undefeated. The
recognition that sceptical hypotheses require evidential motivation, and that
we possess none, is crucial. It shows that RP is a rational stance forced by
the structure of rational doubt.
This clarification blocks the objection that RP is optional or
arbitrary.
5. The Convergence of the Two Grounds
These two grounds – explanatory
economy and the structure of rational doubt – converge. They support RP from
different directions. They show that RP is not arbitrary, dogmatic, or
metaphysical. RP is the rational orientation that structures our epistemic
practices once the conceptual confusion is removed.
The convergence is not accidental. It reflects a structural feature of
our epistemic situation. We inhabit one world, and
only one world is ever accessible. We cannot compare our world with other
possible worlds. We cannot determine whether our world is the “actual” world
rather than a simulation or deception, except in a logically conceivable
(though perhaps physically impossible) radical sceptical scenario. Adherent reality is not demonstrable and, in ordinary circumstances, not
even applicable. Sceptical hypotheses require evidential motivation. Without
such motivation, they are empty. Explanatory economy supports the rational
assumption that our world is adherently real. The structure of rational doubt
supports the rational entitlement to assume adherent reality.
The two grounds converge. RP is the stance that remains once the
conceptual confusion is removed. It is the rational orientation that structures
our epistemic practices once we recognise the limits of demonstrative proof.
V. Conclusions
The argument of this paper has
proceeded by conceptual clarification rather than empirical investigation or metaphysical speculation. The
dissolution of scepticism is not achieved by showing that sceptical hypotheses
are improbable, explanatorily inferior, pragmatically irrelevant, or logically
inconsistent. It is achieved by recognizing that the sceptical demand for
demonstrative certainty rests on a category mistake. The sceptic asks for proof of adherent reality, but adherent reality has no demonstrable status. In our ordinary
circumstances, it is not even applicable, because its assessment requires
comparison across worlds, and only one world is ever accessible, except in
logically conceivable sceptical scenarios whose empirical possibility remains
unknown. The anti‑sceptic inherits the same mistake by attempting to provide
such proof. Both sides of the traditional dialectic misfire because both assume
that reality has a single epistemic status.
By contrast, inherent reality is demonstrable. Section I showed that inherent reality is given in the structure of
experience itself. The criteria of vividness, involuntariness, intersubjective
checkability, law‑governed regularity, and contextual coherence through time form
a mutually reinforcing pattern that makes the attribution of inherent reality
unavoidable. These criteria apply not only to directly perceived objects but
also to indirectly accessible entities, from remote nebulae to subatomic
particles, whose causal integration with observable phenomena justifies their
inclusion within the domain of inherent reality. Once these criteria are
properly understood, they yield a straightforward proof of the external world:
the totality of things that satisfy the criteria of external reality, whether
directly perceived, indirectly detected, or known through testimony. This is
why inherent reality is epistemically secure.
This result stands in sharp contrast to Carnap’s view that the existence
of the external world is a pragmatic fiat. Our analysis shows that the
existence of the external world is not a matter of linguistic choice or
practical convenience. It is demonstrable because inherent reality is given in
the very conditions of intelligible experience. The sceptic cannot coherently
deny inherent reality, for the denial would undermine the conditions under which
sceptical reasoning is possible.
Once the conceptual confusion between inherent and adherent reality is
removed, we must adopt a stance that structures our epistemic practices. The Realist Presupposition
(RP) is the stance that remains. RP
is supported by two independent considerations. Explanatory economy shows that the ordinary hypothesis – that our world is the adherently real
one – is explanatorily superior to sceptical alternatives, which require ad
hoc stipulations to preserve the appearances. The structure of rational
doubt shows that sceptical hypotheses
require evidential motivation, and that without such motivation, sceptical
doubt is empty. These considerations do not prove RP; they vindicate it. They show that RP is the rational stance to adopt once we recognize
the limits of demonstrative proof and the conceptual inapplicability of
reasoning for or against adherent reality in ordinary circumstances.
The conceptual distinction between inherent and adherent reality has
profound epistemological consequences. It clarifies what can and cannot be
demonstrated, the limits of rational doubt, the structure of epistemic
entitlement, the role of presuppositions in epistemic practice, the nature of
the sceptical problem, and the rational stance that remains once the conceptual
confusion is removed.
Most importantly, it clarifies why the sceptical dialectic cannot be repaired. Once reality is recognized as conceptually bifurcated, the sceptic cannot
reformulate the problem in terms of adherent reality alone, for adherent
reality is not demonstrable and not applicable in ordinary circumstances. Nor
can the anti‑sceptic reformulate the response in terms of inherent reality
alone, for inherent reality is not the status at issue in sceptical scenarios.
The dialectic collapses because its presupposition – the univocity of reality –
is untenable. What remains is not a refutation of scepticism but a dissolution: the sceptical problem cannot be coherently formulated once the conceptual
distinction is in place.
References
Bangu, S. (2018). The modal
fallacy in sceptical arguments. Synthese, 195(3), 1203–1222.
Berkeley, G. (2003 [1710]). A Treatise Concerning the
Principles of Human Knowledge. Dover Publications.
BonJour, L. (2010). Epistemology: Classic Problems
and Contemporary Responses. Rowman &
Littlefield Publishers.
Bouwsma, O. K. (1949). Descartes’ evil genius. Philosophical Review, 58, 141–151.
Britt, K. (2026). In defense of
defeat. Synthese, 128(208).
Brueckner, A. (1986). Brains in a
vat. Journal of Philosophy, 83(3), 148–167.
Carnap, R. (1950). Empiricism,
semantics, and ontology. Revue Internationale de Philosophie, 4, 20–40.
DeRose, K., & Warfield, T. A.
(1999). Scepticism: A Contemporary Reader. Oxford University Press.
DeRose, K. (1995). Solving the
skeptical problem. Philosophical Review, 104, 1–52.
Descartes, R. (1641). Meditationes de prima
philosophia. Paris: Michaelem Soly. (French translation: Méditations Métaphysiques, 1647.)
Dretske, F. (1970). Epistemic
operators. Journal of Philosophy, 67, 1007–1023.
Frege, G. (1918–1919). Der
Gedanke: Eine logische Untersuchung. Beiträge zur Philosophie des
deutschen Idealismus, 1, 58–77.
Hume, D. (2007 [1739]). A Treatise of Human
Nature: A Critical Edition. Oxford University
Press.
Kant, I. (2004 [1783]). Prolegomena to Any Future
Metaphysics. Cambridge University Press.
Locke, J. (1975 [1689]). An Essay Concerning Human
Understanding. P. H. Nidditch (Ed.). Clarendon
Press.
Lycan, W. (2001). Moore against
the new skeptics. Philosophical Studies, 103, 35–53.
Mill, J. S. (1889). An Examination of Sir
William Hamilton’s Philosophy. Longmans, Green &
Co.
Mill, J. S. (2009 [1889]). A System of Logic. Oxford University Press.
Moore, G. E. (1962). Proof of an
external world. In Philosophical Papers (pp. 144–148). Collier Books.
Moore, G. E. (1953). The meaning
of real. In Some Main Problems of Philosophy. George Allen & Unwin.
Nozick, R. (1981). Philosophical
Explanations. Oxford University Press.
Putnam, H. (1981). Reason, Truth, and
History. Cambridge University Press.
Reid, T. (1997 [1764]). An Inquiry into the Human
Mind on the Principles of Common Sense. Edinburgh University Press.
Reid, T. (2002 [1785]). Essays on the
Intellectual Powers of Man. Edinburgh University
Press.
Russell, B. (1912). The Problems of
Philosophy. Williams and Norgate.
Strawson, P. F. (1963). Carnap’s
views on ontology. In P. A. Schilpp (Ed.), The Philosophy of Rudolf
Carnap (Vol. 11). Open Court.
Stroud, B. (1984). The Significance of
Philosophical Scepticism. Oxford University
Press.
Vogel, J. (1990). Cartesian
skepticism and inference to the best explanation. Journal of Philosophy, 87(11), 658–666.
Vogel, J. (1987). Tracking,
closure, and inductive knowledge. In S. Luper-Foy (Ed.), The Possibility of
Knowledge: Nozick and His Critics (pp. 197–215). Rowman & Littlefield.
Williams, M. (1996). Unnatural Doubts. Princeton University Press.
Wittgenstein, L. (1969). On Certainty. G. E. M. Anscombe & G. H. von Wright (Eds.). Blackwell.
Wittgenstein, L. (1958). The Blue and the Brown
Books. Basil Blackwell.
Zhao, B. (2026). Contextualism, sensitivity, and
skepticism. Acta Analytica, 41, 521–528.
These criteria were
already partially considered in the philosophical tradition by (at least) Locke
(1689), Berkeley (1710), Hume (1739), Kant (1781), J. S. Mill (1889), Gottlob Frege (1918) and G. E. Moore (1953).
Experiments with artificial reality, such as the integration
of neural‑input devices like the Mudra wristband with AR glasses, reinforce the
central point of the distinction. These systems already generate environments
that are, to a significant extent, inherently real: they
exhibit vividness, independence, virtual intersubjectivity, perceptual
coherence, causal responsiveness, and experiential stability from within the
user’s standpoint. Nevertheless, these environments remain adherently
unreal because their ontological status is derivative: they are
produced, sustained, and delimited by the physical world, which is inherently
and presumably adherently real as well. The artificial environments depend on
our world as their generative substrate and cannot exist independently of it.
Their reality is therefore comparative and subordinate.
Although definitionally very different, the present
distinction has its origins in reflections on Carnap’s (1950) contrast between
internal and external questions of existence, where he suggested that the
existence of the external world as a whole cannot be proved. In my view, he was
confusing his own external question of existence with the adherent question of
reality. On our account, the inherent reality of the world as a whole can be
established as a summation of its items, whereas adherent reality, though not
provable, can still be shown to be the most rational stance.
