Identity Theory to Functionalism
Transition from Identity Theory to Functionalism
Context of Identity Theory: The lecture begins by revisiting the Identity Theory, famously advocated by JJC Smart.
Identity Theory was proposed as the simplest explanation for the widespread correlations observed between mentality and neurology.
It posits a complete reduction of mentality to brain activity, famously summarized by the placeholder identity: pain is identical to "C-fibers firing."
In this framework, certain types of mental states are considered strictly identical to certain types of physical states.
The Challenge of Multiple Realizability: Multiple realizability is introduced as a physicalist challenge to the Identity Theory, emerging from a different brand of physicalism.
The Concept of Multiple Realizability
Biological Metaphor for Multiple Realizability:
Life is viewed as a physical property that does not require a nonphysical "life force" or substance to animate creatures. Biological processes—movement, reproduction, and cellular activity—can be explained physically.
Diversity of Composition: There is a vast diversity in the composition of living things. For example, the cellular structures of plants, warm-blooded animals, cold-blooded animals, invertebrates, and insects with exoskeletons are markedly different.
High-Level vs. Low-Level Structure: Despite these different low-level cellular structures, we apply the same high-level concepts such as "reproduction" or "energy exchange."
An oak tree and a kangaroo produce progeny through radically different molecular and physiological processes, yet both are functionally classified as "reproduction."
General Principle: The same high-level structure can be realized through many different low-level structures. This does not imply the existence of a new entity at the high level, but rather that the high-level phenomenon is independent of the specific low-level material.
Economic Metaphor: The properties of currency (inflationary pressures, changes in value) remain the same regardless of whether the low-level material is gold, seashells, or Bitcoin.
Non-Carbon Based Life: In principle, we can conceive of living beings composed of different elements on the Periodic Table, such as silicon-based or tin-based life forms. Life is defined by its structure rather than its specific elementary mixture.
Type Identity vs. Token Identity Physicalism
Rejection of Type Identity: Accepting multiple realizability requires the rejection of Type Identity Physicalism.
Type Identity claims that categories of mental states (e.g., thinking about cabbage soup) are identical to specific categories or patterns of brain activity.
Critique: It is unlikely that every time an individual thinks of a "ham sandwich," the exact same neurons fire in the exact same sequence. For a philosopher to claim this before neurological evidence is finalized is a bold and potentially problematic commitment.
Adoption of Token Identity: Functionalists typically accept Token Identity Physicalism.
This view holds that each particular instance (token) of a mental event is identical to a particular brain event occurring at that specific moment.
However, there is no sweeping identity across all instances of that type of mental event. The brain is a complex system where the same mental thought could be realized by different patterns of activity at different times.
Functional Definitions of Objects and Mentality
Functionalism Defined: Mind is viewed as something that matter does rather than what it is made of.
Examples of Functional Definitions:
Chair: Can be made of glass, plastic, or metal; it is defined by its use or function.
Computer: Historically uses vacuum tubes or punch cards; future computers may be quantum-based. It is defined by its computational role.
Other Examples: Knives (cut), refrigerators (cool), leaders (role in group), art (social interaction), and appendages (wings vs. arms).
Defining Mind Functionally: Since minds appear multiply realizable, they should be defined by the role the mental state plays in the organism rather than material parts.
Pain: Could be defined as a "tissue damage detector" or a state that leads an organism to avoid certain stimuli and communicate that avoidance to others.
Beliefs/Desires: These are information states and dispositions that combine to lead to action.
Inter-definition: Specific mental concepts are often defined by their relationships to each other within the whole of mental life, which supports the functionalist perspective.
Machine Functionalism and the Turing Machine
Machine Functionalism (Computationalism): This specific branch of functionalism views the mind's role as information processing. The mind is the activity of a computer processing information.
The Turing Machine Model:
Physical Components: An infinite tape divided into squares, a scanner/printer that moves and modifies symbols on the tape.
Internal Configurations: The machine operates based on specific internal states or configurations.
Rules and Procedures: The machine follows mechanical rules: "If in state and scanning symbol , then perform action (erase, print, or move)."
Probability: Machines can be modified to be probabilistic rather than strictly deterministic (e.g., a chance of one output versus a or chance of others).
The Brain as a Machine: The brain stores and recalls information using neurons that vibrate at different frequencies or carry electrical charges. This mirrors magnetic storage (ons and offs, ones and zeros). A machine description can explain the brain in terms of inputs, internal states, and outputs.
Critiques and Limitations of Machine Functionalism
Problem 1: Intentionality and Meaning:
Computers process symbols according to rules but do not "mean" or "understand" the symbols. Meaning requires a conscious, intending agent in the backdrop to interpret the outputs.
A computer might display "Click OK to confirm," but the words only have meaning because of the human user, not the computer itself. Information processing is not equivalent to understanding or believing.
Problem 2: The Turing Test vs. Real Understanding:
The Turing Test suggests that if two systems are equivalent in terms of inputs and outputs, they have the same psychological/intelligent state.
The lecturer notes a recent attempt to fool people into believing a computer was a Ukrainian boy, using the child’s age and non-native English as a way to excuse grammatical errors.
John Searle’s Chinese Room (1980):
Strong AI: The claim that an appropriately programmed computer literally is a mind and possesses cognitive states.
The Experiment: Searle imagines himself in a room with a rulebook (dictionary) for Chinese symbols. He receives inputs and, following the rules, provides correct outputs in Chinese. To an observer outside, he appears to be a fluent Chinese speaker (passing the Turing Test).
Conclusion: In reality, Searle does not understand Chinese; he is merely manipulating symbols. Therefore, a computer operating on a set of rules does not possess understanding or consciousness, no matter how sophisticated the output.
Conclusion and Future Directions
Ongoing Relevance: Despite objections like Searle's, functionalism remains a powerful idea because it offers a physicalistically acceptable alternative to Identity Theory while accommodating multiple realizability.
Next Steps: The transition continues into Chapter 6, which will explore "Causal-Theoretical Functionalism" as a potential improvement over Machine Functionalism.