Cognition

• What sets human beings apart from other animals is their capacity for sophisticated thought.

• Only humans can perform complex cognition such as building airplanes, writing Hamlet, composing symphonies, propounding scientific theories like Relativity, or discovering DNA.

• Humans possess voluminous knowledge of the world, much of which is beyond the conception of other animals.

• Epistemology is the theory of knowledge, seeking to understand how humans accomplish such sophisticated thought and acquire extensive knowledge.

• The goal is to understand rational thought, from the routine to the sublime.

• We want to know how humans accomplish epistemic tasks that other creatures find impossible.

• Cognitive psychology studies aspects of human cognition scientifically but includes both rational and irrational thought.

• Our interest lies specifically in rational cognition—what it means for thought to go right.

• Understanding what makes humans rational and enables great intellectual achievements is a philosophical challenge.

• Psychologists presuppose a pre-theoretic understanding of rationality but do not clarify it deeply.

• We seek a general theory of rationality: what it means to be a rational cognizer, and how rationality enables acquisition of broad world knowledge.

• This is a philosophical question with a long history.

• Rational cognition includes more than pursuit of knowledge; knowledge has a purpose—to help us navigate the world.

• Knowledge guides us in deciding how to act, so rational cognition also involves cognitive processes in action decisions.

• The focus here is on the purely intellectual aspects of cognition involved in pursuing knowledge.

• A key conclusion of contemporary epistemology: rational thought behind great achievements is not different in kind from rational thought in routine epistemic tasks.

• Routine epistemic tasks include determining the color of an object, remembering facts, understanding gravity, and basic arithmetic.

• Complex discoveries like DNA arise from stringing together many routine epistemic problems.

• The extraordinary nature of human thought is already present in our ability to solve mundane epistemic problems.

• The study begins with the mundane to ultimately understand the sublime.

Skeptical Problems

• Epistemologists traditionally ask, “How is knowledge possible at all?” rather than focusing on specific discoveries like curing cancer or discovering DNA.

• Knowing how to do simple epistemic tasks is different from knowing how these tasks are done in a way that can be precisely explained or replicated.

• For example, we know how to pick up a cup, but explaining it well enough for a robot to do it took years of engineering effort.

• It is extremely difficult to explain why what we do yields knowledge.

• Philosophers have historically used skeptical arguments to show that even simple epistemic tasks seem impossible.

• The tale of Mike and Harry illustrates a skeptical problem: Mike suspects he is a brain in a vat connected to a computer simulating his experiences.

• If the brain-in-a-vat scenario were true, Mike—and by extension anyone—could not know if their sensory experiences are real or simulated.

• Since there is no way to disprove the brain-in-a-vat hypothesis, one cannot be sure that the material world exists as perceived.

• This leads to the skeptical conclusion that knowledge of the external world is impossible because it could all be an illusion.

• Such skeptical problems challenge even the most basic, mundane knowledge we take for granted.

• They also challenge sophisticated human knowledge, undermining what makes human cognition unique.

• Historically, epistemologists like Descartes aimed to find indubitable beliefs to justify all others, while Hume struggled with skepticism about induction.

• Kant acknowledged the problem that the existence of the external world must be accepted merely on faith, and no satisfactory proof can counter doubt.

• Contemporary epistemology views refuting radical skepticism as impossible in principle because skeptics doubt all relevant premises.

• The skeptic’s refusal to accept any starting point means there is no basis for reasoning with them.

• Therefore, the entire project of refuting the skeptic is considered ill-founded because it grants the skeptic no grounds to argue against doubt.

• Proper treatment of skeptical arguments requires viewing them differently: we start with a large set of beliefs initially regarded as knowledge.

• These beliefs can conflict and form an inconsistent set, meaning they cannot all be true simultaneously.

• Skeptical arguments start from premises we initially confidently accept and conclude that certain kinds of knowledge are impossible, conflicting with our confidence that we do have such knowledge.

• We cannot reasonably continue to hold all these conflicting beliefs.

• One reaction is to discard all beliefs and start over (Descartes’ approach), but this fails because the skeptic also questions the cognitive processes used to form beliefs.

• We cannot discard both beliefs and cognitive processes, or we have nothing to start again with.

• Otto Neurath’s metaphor: “We are like sailors who must rebuild their ship upon the open sea,” meaning we must repair beliefs and cognitive processes from within.

• G. E. Moore argued for beginning with what we already have, using common-sense proofs (e.g., holding up hands to prove they exist).

• Although Moore’s complete rejection of skepticism has issues, his point stands that some beliefs are held with greater confidence than others.

• It is reasonable to rely more on beliefs in which we have greater confidence and, when conflicts arise, to reject the least certain beliefs.

• Skeptical arguments typically have the form:

  • P1, P2, …, Pn (premises)

  • Therefore, Q (conclusion)

• Normally, if we believe all premises, we accept the conclusion even if it denies some other belief.

• However, validity only shows we cannot reasonably hold all of P1,…,Pn and ~Q (denial of Q).

• The argument does not tell us which belief to reject because we can construct equally valid arguments denying any one premise instead of Q.

• For example:

  • P2, …, Pn, ~Q

  • Therefore, ~P1

• When faced with skepticism, we usually believe the premises and also the denial of the conclusion (~Q).

• The argument forces rejection of at least one belief, but does not specify which.

• We should reject the belief of which we are least certain, often one of the premises rather than the knowledge denied by the skeptic.

• Typically, we are more certain of the knowledge claimed than of some premises in skeptical arguments.

• Therefore, it is unreasonable to accept the skeptical conclusion that the knowledge does not exist.

• Instead, the rational response is to deny one or more premises of the skeptical argument.

• A typical skeptical argument is better understood as a reductio ad absurdum of its premises rather than a proof of its conclusion.

• Inductive reasoning draws general conclusions from a finite number of observed instances, e.g., observing many white swans and inferring all swans are white.

• Induction is common in science for testing theories by observing particular instances.

• The truth of premises in induction does not logically guarantee the truth of the general conclusion (e.g., presence of black swans disproves “all swans are white”).

• Hume’s skeptical argument against induction:

  1. Premises of inductive arguments do not logically entail the conclusion.

  2. If premises do not logically entail the conclusion, it is unreasonable to believe the conclusion based on those premises.

  3. Therefore, inductive reasoning is illegitimate for acquiring knowledge of general truths.

• Though the conclusion seems initially absurd (we rely on induction), the premises seem convincing.

• The second premise was once widely accepted: if an argument’s conclusion is not logically entailed by premises, we shouldn’t believe it.

• The skeptical argument is typical: premises seem compelling, conclusion seems absurd, so it functions as a reductio ad absurdum, meaning one premise must be false.

• Some philosophers (Russell, Lewis) tried to deny the first premise by adding the “uniformity of nature” principle (future resembles past), aiming to make inductive arguments deductively valid.

• Difficulty: formulating a precise, strong principle of uniformity is hard because the future is not always like the past—things change.

• More serious problem: if the argument is deductively valid, then since the conclusion can be false, some premise must be false. All premises but uniformity are true, so uniformity must be false.

• Therefore, the uniformity of nature premise is false, making it useless for inductive reasoning, leading back to Hume’s skeptical conclusion by another route.

• To avoid Hume’s conclusion, we must deny the second premise instead: it must be possible for premises to provide a reason to believe a conclusion without logically entailing it.

• Such reasons are called defeasible reasons: they justify belief provisionally, but can be defeated by new evidence.

• Example: observing all A’s are B’s may justify believing all A’s are B’s until a counterexample appears.

• Recognition of defeasible reasoning is a major epistemological advance in the 20th century.

• Most human reasoning proceeds defeasibly, not deductively.

• Another example: perception usually justifies belief about surroundings unless contradictory information appears (senses can mislead).

• Defeasible reasoning will be discussed further in the next chapter.

• Skeptical arguments are important not because their conclusions might be true, but because they reveal insights about knowledge and justification.

• The epistemologist’s task is not to refute skepticism outright but to explain why the skeptic is wrong by examining which premises of skeptical arguments fail.

• This approach assumes that we have various kinds of knowledge (~Q) and explores what that implies.

• For example, given perceptual knowledge, we investigate how perceptual beliefs relate to sensory experience.

• This reasoning takes the form: “We have such knowledge; if X were true, we could not have that knowledge; therefore, X is false.”

• Such reasoning results from contraposing the premises and conclusion of skeptical arguments and is common in contemporary epistemology.