Final Prep

Let's tackle the exam prep questions systematically. I’ll start by addressing the Small Questions. These answers are brief yet precise, and tied to the readings and notes you provided.

Small Questions

1. What is phenomenal consciousness? Give three examples of phenomenally conscious mental states.

   • Phenomenal consciousness refers to the subjective, qualitative experience of mental states—what it feels like to experience them.

   • Examples:

     • Feeling pain from a stubbed toe.

     • Seeing the color red.

     • Tasting the sweetness of sugar.

2. What is intentionality? Give three examples of intentional mental states.

   • Intentionality is the "aboutness" of mental states, their capacity to represent or be about things in the world.

   • Examples:

     • Believing that the Earth is round.

     • Hoping it will snow tomorrow.

     • Perceiving a tree in a forest.

3. What is identity theory?

   • Identity theory asserts that mental states are identical to brain states. For instance, feeling pain is the same as specific neural activity in the brain. It’s a materialist view supporting the idea that the mind is reducible to the physical brain.

4. What is dualism?

   • Dualism is the view that the mind and body are distinct substances. Descartes famously argued that the mind is a non-physical, thinking substance, while the body is a physical, spatial substance.

5. What is behaviorism?

   • Behaviorism holds that mental states are reducible to behavioral dispositions or observable behaviors. For example, "pain" might be defined as tendencies to cry or wince.

6. What is functionalism?

   • Functionalism posits that mental states are defined by their functional roles—what they do—rather than their physical makeup. For instance, pain could be any state that causes avoidance behavior and distress, regardless of its physical basis.

7. What’s the hard problem of consciousness?

   • Coined by Chalmers, the hard problem asks: why and how do brain processes give rise to conscious experiences, such as the feeling of pain, rather than simply explaining their functions?

8. What is the distinction between sense and reference?

   • Sense is the way an expression picks out its reference, while reference is the object the expression denotes. Example: "The morning star" and "The evening star" both refer to Venus but have different senses.

9. What is a category mistake? Give three examples of category mistakes.

   • A category mistake involves ascribing properties to something that doesn’t belong in that category. Examples:

     • Asking to see "the university" after being shown its buildings.

     • Assuming numbers have color.

     • Saying a headache weighs five pounds.

10. What is multiple realizability?

    • The idea that the same mental state (e.g., pain) can be realized by different physical systems, such as human brains, octopus nervous systems, or computer circuits.

11. What is the computational theory of mind?

    • The mind is like a Turing Machine that manipulates symbolic representations through computations. Thinking involves processing representations akin to computations.

12. What is the language of thought (LOT) hypothesis?

    • LOT posits that thought occurs in a mental language with syntax and semantics. Mental representations have a structure analogous to natural language sentences.

13. What is compositionality?

    • The meaning of a complex representation derives from its parts and how they are syntactically combined. For example, “red apple” combines meanings of “red” and “apple.”

14. What is strong AI?

    • The claim that appropriately programmed computers can possess genuine mental states akin to human minds, not just simulate them.

15. What is connectionism? What is the difference between implementational and radical connectionism?

    • Connectionism models cognition using artificial neural networks.

    • Implementational connectionism views neural networks as implementing classical cognitive theories.

    • Radical connectionism rejects classical structures, modeling cognition as emergent from distributed neural patterns.

16. What’s the distinction between local and distributed representations?

    • Local representations: each concept is represented by a specific unit.

    • Distributed representations: concepts are represented by patterns of activity across multiple units.

17. What is the distinction between atomic and complex representations?

    • Atomic representations lack internal structure (e.g., "Mary").

    • Complex representations have structured components (e.g., "John loves Mary").

18. What is belief-desire psychology?

    • The common-sense framework that explains behavior via beliefs and desires. For example, attending class is explained by a belief that class is happening and a desire not to miss it.

19. What is misrepresentation? Give three examples of misrepresentations.

    • Misrepresentation occurs when a mental state inaccurately represents the world. Examples:

      • Mistaking a stick in the water for a snake.

      • Misremembering a face.

      • A faulty thermostat activating heat when it’s warm.

20. What is the disjunction problem to causal theories of mental content?

    • If a mental state (e.g., "horse") is defined by what causes it, misapplications (e.g., mistaking a cow for a horse) would imply that the state represents both “horse” and “cow in dim light,” undermining the idea of misrepresentation.

Big Questions

1. Smart argues for identity theory on the basis of simplicity.

Reconstruct the argument

Smart’s argument for identity theory centers on simplicity. He posits that mental states are identical to brain states, such as "pain" being identical to C-fiber firings. Identity theory is simpler than dualism because it does not require positing non-physical substances or complex interactions between mind and body. By applying Occam’s Razor, which favors simpler explanations when all else is equal, Smart concludes that identity theory should be preferred.

Justify the premises

• The simplicity of identity theory stems from its reduction of mental phenomena to physical processes. Dualism, by contrast, complicates explanations by introducing a non-physical mind that interacts mysteriously with the brain.

• Advances in neuroscience, such as the correlation between brain activity and mental states, support the premise that the mind can be explained physically.

• Occam’s Razor, widely accepted in philosophy and science, justifies choosing simpler theories when they account for the same phenomena.

2. Lewis argues for identity theory on the basis of the causal role of mental states.

Reconstruct the argument

Lewis’s argument emphasizes the causal role of mental states in producing behavior. He argues:

1. Mental states cause behavior.

2. Brain states are the only physical entities capable of causing behavior.Conclusion: Therefore, mental states must be brain states.

Justify the premises

• Premise 1: Mental causation is evident in everyday experiences, such as thirst leading to drinking water. Rejecting this premise would undermine our understanding of human action.

• Premise 2: Empirical studies demonstrate that brain states directly correlate with and cause behavior, making non-physical explanations unnecessary. Physicalism supports this view.

• By combining these premises, Lewis concludes that mental states and brain states are identical. Critics challenge this by invoking phenomena like multiple realizability.

3. In the course of arguing against behaviorism, Putnam presents an analogy between mental states and diseases.

Explain the analogy

Putnam compares mental states to diseases like polio. Historically, diseases were identified by symptoms but are now understood as caused by underlying mechanisms (e.g., viruses). Similarly, while mental states may be associated with behavior, they are not reducible to it but rather are the causes of behavior.

Explain how this analogy is supposed to help in arguing against behaviorism

• Behaviorism defines mental states as dispositions to behave in certain ways. Putnam’s analogy undermines this by showing that causation, not mere correlation, is key.

• Counterexamples: Spartans suppress pain behavior but still feel pain. Paralyzed individuals can feel pain without exhibiting behavior. Both cases show that mental states are distinct from behavioral dispositions.

• The analogy shifts focus to underlying mechanisms, laying groundwork for functionalism and identity theory.

4. Block argues against functionalism on the basis of the Chinese nation thought experiment.

Reconstruct the argument

1. Functionalism claims that mental states are defined by causal roles and can be realized by any system that implements these roles.

2. A large nation of people connected via radios, simulating a brain’s causal roles, would fulfill functionalist criteria.

3. However, it is absurd to think that such a system would have mental states (e.g., feel pain).Conclusion: Functionalism is false.

Justify the premises

• Premise 1: Functionalism emphasizes causal roles rather than the physical substrate. This allows for non-biological realizations of mental states.

• Premise 2: The thought experiment illustrates functionalism’s implications. A nation communicating via radios can simulate neural activity.

• Premise 3: Intuitively, such a system lacks subjective experience (qualia). Functionalism’s inability to explain qualia highlights its limitations.

5. Block argues against passing the Turing test being a sufficient condition for intelligence on the basis of the Aunt Bubbles thought experiment.

Reconstruct the argument

1. The Turing Test evaluates intelligence based on behavioral outputs.

2. Aunt Bubbles, a massive look-up table of pre-written responses, could pass the Turing Test without genuine intelligence.

3. Therefore, passing the Turing Test is not sufficient to establish intelligence.

Justify the premises

• Premise 1: The Turing Test measures outputs without examining internal processes.

• Premise 2: Aunt Bubbles produces outputs that mimic intelligent responses but lacks generative capacity or understanding.

• The argument demonstrates that intelligence requires more than behavior—it involves generative, systematic processing, which Aunt Bubbles lacks.

6. Fodor argues for the language of thought (LOT) hypothesis in a number of contexts on the basis of the claim that the mind is productive and systematic.

Explain what Fodor means by saying that the mind is productive and systematic

• Productivity: Humans can generate and understand an infinite number of novel sentences and thoughts, implying a combinatorial system.

• Systematicity: If one can entertain “John loves Mary,” they can also entertain “Mary loves John.” This reflects an underlying structural organization in thought.

Explain why Fodor thinks that if the mind is productive and systematic, the LOT hypothesis is true

• LOT posits that thoughts are structured like sentences, with syntax and semantics.

• Productivity and systematicity arise naturally from LOT’s combinatorial structure, where simpler elements combine to form complex thoughts.

• Connectionist models lack this structure and struggle to explain productivity and systematicity, making LOT the best explanation.

7. Searle argues against Strong AI on the basis of the Chinese Room thought experiment.

Reconstruct the argument

1. Programs are purely syntactic, operating on symbols without understanding meaning.

2. Minds have semantics, involving understanding and intentional states.

3. Syntax is not sufficient for semantics.Conclusion: Therefore, programs cannot produce minds, and Strong AI is false.

Justify the premises

• Premise 1: Computers follow formal rules to manipulate symbols but do not comprehend them.

• Premise 2: Human minds exhibit understanding (e.g., interpreting language meaningfully).

• Premise 3: The Chinese Room experiment shows that manipulating symbols according to rules does not lead to understanding. The person in the room follows instructions but does not understand Chinese.

• The argument implies that computation alone is insufficient for mental states.

8. Block discusses the connection between intentionality and intelligence.

Can a system have intentionality without intelligence?

Yes, a system can have intentionality without intelligence. Examples include photographs or tree rings, which represent information (about objects or age) but do not process it intelligently. Intentionality here depends on representational content, not cognitive processes.

Can a system have intelligence without intentionality?

Yes, Block argues that intelligence without intentionality is possible. For example, a "swamp brain" (hypothetical duplicate of a brain formed randomly) could exhibit intelligent behavior without the historical connections to its environment that give rise to intentionality. This shows intelligence depends on internal mechanisms, whereas intentionality depends on external and historical factors.

9. Block and Fodor think that the mind is a “syntactic machine driving a semantic engine.” Rescorla disagrees.

Explain what Block and Fodor mean

Block and Fodor describe the mind as a system where syntactic processes (formal symbol manipulations) underpin semantic content (meaning). Syntax operates at a mechanical level, while semantics emerges from the interpretation of syntactic structures.

Why does Rescorla disagree?

Rescorla argues that this view reduces semantics to a byproduct of syntax, making it explanatorily redundant. He uses the "shadow of a car" analogy: semantic content, like a shadow, may correlate with syntactic processes but does not causally contribute to behavior. This challenges the claim that semantic content has an independent role in explaining actions.

10. Rescorla uses the example of the shadow of a car in order to argue against the claim that mental content plays a role in explaining our behavior if Fodor and Block are right that the mind is a “syntactic machine driving a semantic engine.”

Explain the shadow example

Rescorla compares semantic content to a car’s shadow. The shadow moves with the car but does not influence its movement. Similarly, he suggests semantic content might correlate with behavior but lacks causal efficacy if it is merely a byproduct of syntactic processes.

How does it support Rescorla’s argument?

The shadow example illustrates that semantic content could "follow" syntactic processes without influencing them. Rescorla argues that if mental content is epiphenomenal in this way, it cannot explain behavior. This undermines Block and Fodor’s claim that syntax drives a semantic engine that causally impacts actions.

11. Fodor thinks that if the LOT hypothesis was false, we couldn’t explain our linguistic competence in the way psycholinguists do.

How do psycholinguists explain linguistic competence?

Psycholinguists propose that understanding language involves constructing and manipulating mental representations with syntactic structure. For example, parsing a sentence requires applying rules that generate hierarchical representations.

Why is LOT necessary for these explanations?

• The LOT hypothesis provides a framework for structured mental representations akin to sentences, with syntax and semantics.

• Without LOT, it is unclear how the mind could generate, store, and manipulate the structured representations required for linguistic competence. LOT explains productivity and systematicity in language use.

12. Putnam uses the example of an ant walking in the sand to argue against the claim that similarity is sufficient for intentionality.

Explain the example

Putnam imagines an ant unintentionally tracing a pattern in the sand that resembles Winston Churchill’s face. Although the pattern looks like Churchill, the ant’s actions lack the intentionality to represent him—it is a coincidence.

How does this refute similarity as sufficient for intentionality?

The ant’s tracing is similar to Churchill’s face but does not represent him. This shows that mere resemblance is insufficient for intentionality; representation requires more, such as causal or historical connections to the represented object.

13. Putnam uses the Twin Earth example in order to argue against the claim that narrow psychological states determine the extension of our terms.

Present the Twin Earth thought experiment

Twin Earth is identical to Earth except that its "water" is composed of XYZ, not H2O. Despite identical mental states, an Earthling’s term "water" refers to H2O, while a Twin Earthling’s term refers to XYZ.

How does this argue against narrow psychological states determining extension?

Narrow psychological states (internal to the mind) cannot fully determine reference because external environmental factors (e.g., H2O vs. XYZ) influence the extension of terms. Reference depends on external relations, not just internal states.

14. Fodor presents the disjunction objection to the Crude Causal Theory (CCT) of mental content.

What does CCT say?

CCT claims that the content of a mental state is determined by the set of things that reliably cause it. For example, "horse" means horse because seeing horses reliably causes the representation "horse."

What is the objection?

The disjunction problem arises when non-horses (e.g., cows in poor lighting) also cause the representation "horse." According to CCT, this would mean "horse" refers to "horse or cow in poor lighting," undermining the concept of misrepresentation. Fodor counters by proposing asymmetric dependence: only the primary cause (horses) determines content, while secondary causes depend on this primary relationship.

15. Both Dretske and Neander appeal to historical functions in order to solve the problem of misrepresentation.

What is the problem of misrepresentation?

Misrepresentation occurs when a mental state inaccurately represents the world (e.g., mistaking a cow for a horse). A theory of mental content must account for how representations can be wrong.

How do historical functions solve this?

• Historical functions: A representation’s content is determined by what it was selected to represent through evolution or learning. For example, a frog’s vision system evolved to detect flies, so it represents flies even if it occasionally misrepresents other small objects as flies.

• Historical functions introduce a normative standard: representations are "supposed" to correspond to their selected targets. Misrepresentation occurs when they fail to meet this standard.

• Dretske emphasizes learning-based selection, while Neander focuses on evolutionary selection, highlighting different sources of historical functions.