Notes on Skepticism, Eddington's Tables, and the Simulation Hypothesis

Skepticism and the Matrix Idea

Skeptical scenario: brains in vats / dream world / matrix-like simulation where our brains in vats are fed signals by other brains; we appear alive but may not be
The skeptic’s challenge: if this is true, most of our beliefs about the world (grounded in perception) would be undermined because perception would be feeding us deceptive representations
Chalmers’ defense: this skeptical scenario does not actually undermine ordinary beliefs; he uses an analogy to illustrate why higher-level truths can remain true even if there is a deeper, surprising layer

Eddington's Two Tables (Chalmers’ analogical device)

Ordinary table (everyday, perceptual):
- Described as solid, tangible, color, texture, permanence (though nothing is truly permanent)
Table in physics (fundamental description):
- Made of atoms; atoms are mostly empty space; protons, neutrons, electrons in constant motion; atoms are not solid at a fundamental level
- Color is not intrinsic to fundamental physics; color emerges from perceptual systems
Key contrast: higher-level properties (solid, color, texture) vs lower-level, fundamental physical description (atoms, quantum properties, etc.)
Point of the analogy: discovering surprising properties at the fundamental level does not falsify the ordinary table’s reality; it’s a different description at a lower level, not a denial of the table’s reality
Takeaway: the surprising nature of fundamental physics does not undermine the reality of everyday objects; there is a hierarchy of descriptions at different levels

If we are in a simulation: a parallel analogy to Eddington’s tables

If hands (and other phenomena) are ultimately bits in a computer program (on/off switches, binary states), that would still be real in the sense that they are the output of a functioning system
Even if the most fundamental level is computational (bits), that does not negate the reality of the structured world experienced within the simulation
The simulation would have its own levels: physics, chemistry, biology, etc., potentially with their own laws and emergent properties
The external world that runs the simulation would still exist; a creator or outside substrate is needed to run the simulation
The existence of a lower, computational level does not negate the reality of the inside-world physics and emergent phenomena within the simulation

The reality of the levels: hierarchy of the sciences

Common picture of scientific organization:
- Social sciences (highest level) -> psychology -> biology -> chemistry -> physics (lowest level)
In this view, higher-level states (beliefs, desires, selves) may be constituted by lower-level states (neural states, biochemical processes, atomic interactions)
Example: belief that there is water in this cup vs belief that there is milk in this cup; the difference is encoded in slightly different brain states
Brains are biological systems; chemistry underpins neural processes; chemistry itself is governed by physics (atoms, subatomic particles)
The hierarchy is a description framework: higher-level descriptions do not cease to be true or useful when we learn their lower-level explanations
The key idea: the existence of a more fundamental layer (even if simulation-based) does not invalidate higher-level truths; it just reframes them at a different level of description

The broader claim: if we’re in a simulation, ordinary beliefs can still be true

Even if we are in a simulated world, the physics within the simulation would be real for us and would ground our beliefs
The possibility of evolution by natural selection inside the simulation is plausible; there could be a self-consistent causal structure within the simulation that yields ordinary empirical regularities
If the base layer is “bits” (0s and 1s), that does not automatically negate the reality of objects as experienced inside the simulation
The creators or the outside physical substrate would be real; there would be an external physical world where the simulation runs
There would still be a physical world that grounds causation, meaning, and reference within the simulated environment

Meaning, reference, and language in a simulation (causal theory of reference)

Chalmers draws on a causal theory of reference from philosophy of language
Core claim: the meaning of a word (e.g., "cat") is anchored by its causal connections to the things in the world with which we interact
Inside a simulation, the word "cat" would refer to simulated cats because of the causal interactions within that simulated reality
If one’s life is entirely within a simulation and they learn the term "cat" in that context, the meaning is indexed to simulated cats; the word still picks out the right kind of thing in that world
A thought experiment: imagine being non-simulated, then placed into a simulation without realizing it; the content of beliefs like "cat" could become false because the real-world referent has changed, yet many past beliefs could still survive if their content was anchored to the prior reality
The content of concepts depends on the reality with which one has causal contact; crossing from one ontological regime to another could disrupt the content of terms that were learned in a different regime
If you grew up in a non-simulated world and are moved to a simulation, your word meanings may fail to correctly track the simulated referents; this is a problem for a purely naive mapping of meaning to referents across radically different environments
The causal theory of reference provides a plausible mechanism for how words could maintain content in a simulation, but it also highlights potential failures if the causal history changes radically

Strengths and limits of Chalmers’ position (devil’s advocate and critical edge)

Chalmers acknowledges the following limitations of his proposal:
- It's possible to imagine alternative skeptical scenarios beyond the matrix (e.g., being in a dream within a dream, or waking up in a different reality), but the core idea remains: our beliefs can remain true even within a skeptical scenario because they track a real structure within the scenario
- The meaning of words can be sensitive to the causal history; a shift in reality (simulation vs non-simulation) can threaten the content of long-standing beliefs if those beliefs depend on the original causal network
The higher-level beliefs grounded in our perceptual experiences can still be true in their own right within the correct causal framework, even if the deepest level of reality is different from what we traditionally assume

Metaphysical beliefs vs empirical beliefs; Wittgenstein and Moore on knowledge

Distinction between different kinds of beliefs:
- Empirical beliefs anchored in perceptual experience can be true even in a simulated world; they can be true within the internal logic and physics of the simulation
- Metaphysical beliefs (e.g., existence of a creator, God) are more sensitive to the underlying ontology; a simulation would imply a creator/external substrate, altering the truth-value of certain metaphysical assumptions
Wittgenstein’s idea (as invoked): you cannot challenge everything at once; the knowledge-claim game involves focusing on specific doubts and showing that you can rule them out with targeted reasons
Moore’s claim (in contrast): he presents arguments claiming to justify knowledge claims even against skeptical scenarios by arguing that the premises of the skeptical argument are less certain or more questionable than his proposed claims
The position in the lecture: no one thinks you can definitively prove you’re not in a skeptical scenario; what can be done is to present stronger, more plausible justification for our ordinary beliefs than the skeptical arguments do

Practical and exam-related implications

Even if there were a simulation, there would still be a physically described world (the simulation’s own physics and its outside substrate)
Skeptical scenarios may affect the status of metaphysical and scientific beliefs (e.g., about the ultimate nature of reality), but they do not automatically destroy ordinary empirical knowledge
There are no plausible skeptical scenarios that undermine all of our beliefs at once; a Wittgenstein-esque move suggests keeping beliefs that are necessary for navigating the world while treating skeptical arguments as special cases rather than general refutations
Real-world takeaways: our everyday beliefs about objects, causal relations, and perceptual experiences remain practically trustworthy within the framework we operate in, even if the deeper metaphysical status of those beliefs is complex or uncertain

Class discussion prompts and assessment themes (from the session)

Moore vs Wittgenstein on knowledge vs certainty: knowledge requires being able to rule out specific doubts, not merely having a feeling of certainty
Chalmers on the possibility of proving we aren’t in a skeptical scenario: the claim is that such proof is unlikely; instead, we rely on robust justification of ordinary beliefs
The matrix/simulation discussion: if a simulation exists, there would be an outside physical reality; the meanings of our terms are anchored to causal relationships, which may differ across realities
Hypothetical test questions discussed in class:
- Moore thinks our most basic beliefs can be justified by an argument; Wittgenstein challenges that idea; Chalmers offers a different approach arguing about the persistence of empirical beliefs under skepticism
- True/false prompts about what Wittgenstein or Moore would say; what makes a belief count as knowledge; how to evaluate knowledge claims against skeptical scenarios

Resources and next steps

Instructor notes mention additional resources and a link to a folder/page with references used in class discussions
Students are encouraged to review the cited discussions on skeptical scenarios, the matrix analogy, and the philosophy of language (causal theory of reference) for a deeper understanding
Consider exploring further readings on Eddington’s two tables, Chalmers’ arguments on simulation skepticism, and the causal theory of reference to strengthen comprehension for the exam

Key takeaways to memorize for the exam

Skepticism about reality (brain in a vat, matrix) challenges perception-based knowledge but does not automatically undermine ordinary empirical beliefs
Eddington’s two tables illustrate that a more fundamental description of objects can differ drastically from common-sense descriptions, yet both can be true at different levels
If we are in a simulation, a real external substrate and internal simulation physics could still ground our beliefs; the outside world remains a valid causal source
The causal theory of reference ties meanings to causal interactions with the world; a switch between realities (sim vs non-sim) can complicate word meanings and reference
There are limits to skeptical scenarios; Wittgenstein’s insight emphasizes not attempting to invalidate everything at once and focusing on rule-governed knowledge claims
In philosophy of mind and metaphysics, empirical beliefs and analytic claims can persist even under skeptical challenges; metaphysical beliefs may be more susceptible to revision

Quick references (as a study aid)

Eddington’s two tables: ordinary table vs physics-based table
Relationship between higher-level and lower-level scientific descriptions
Simulation hypothesis: external substrate, internal laws, potential for evolution within simulation
Causal theory of reference: meaning = reference through causal interactions
Wittgenstein: knowledge as ability to rule out specific doubts; not merely certainty
Moore vs Chalmers: different strategies for handling skepticism; none offer an absolute-proof denial of skeptical scenarios