Breakthrough Research in “Platonic Space” | Michael Levin
Summary
This video explores the philosophical concept of universals and its relation to intelligence, proposing a Platonic view where intelligence exists in a separate space and is harnessed by physical systems. It defines intelligence pragmatically as the ability to reach a goal by different means and discusses the nature of the self, suggesting it's a continuous storytelling process of reinterpreting memories. The conversation also touches on how our actions message our future selves, implying ethical considerations for others' future selves.
Key Insights
Intelligence is pragmatically defined as the ability to reach a goal by different means, a cybernetic perspective.
For practical purposes, intelligence is defined using William James's definition: some degree of the ability to reach the same goal by different means. This definition is cybernetic, focusing on competence in navigating a problem space despite interventions or barriers, without specifying the nature of the agent (e.g., brain, natural vs. artificial).
A Platonic worldview suggests intelligence exists in a separate, abstract space and is discovered, not invented.
The speaker leans towards a Platonic view for intelligence, akin to how mathematicians view mathematical truths. This perspective posits that certain abstract entities, including rules of mathematics and possibly minds, exist in a separate Platonic space. These are thought to be discovered, not created, and would persist even if the physical universe were different.
Intelligence is harnessed by physical systems (bodies) by tapping into pre-existing 'minds' in Platonic space.
Extending the Platonic view, it's proposed that this abstract space contains not only abstract rules but also entities with high agency, including minds and various ways of being intelligent. Building a physical body, whether biological or artificial, is seen as harnessing these pre-existing intelligences, much like harnessing computational rules for devices.
Our sense of self is maintained by reinterpreting memories as messages from past selves within a dynamic storytelling process.
From the agent's own perspective, the self is viewed as a process of continuous storytelling. Memories are not faithfully reproduced but are interpreted as 'messages' from a past self that require reinterpretation by the current self, which may be fundamentally different. This reinterpretation is crucial for maintaining coherence, especially across radical transformations like caterpillar to butterfly.
Current actions are messages to our future selves, shaping the environment and conditions for their existence.
It is proposed that current actions function as messages to one's future self, setting the conditions and environment in which that future self will exist. This includes self-improvement efforts and even intrusive thoughts, which can alter brain structure and information processing capabilities for the future self.
Recognizing the self as a continuous process has ethical implications for how we treat others' future selves.
This process view of the self, where current actions message future selves, suggests an ethical parallel. If we act to benefit our own future selves, we might extend similar consideration to the future selves of others. This perspective challenges the notion of a persistent, isolated self and encourages a broader consideration of well-being across individuals and time.
Sections
Universals and the Nature of Intelligence
Philosophical concepts of universals, like Plato's Forms vs. Aristotle's immanent properties, are relevant to understanding intelligence.
The discussion begins by drawing parallels between philosophical concepts of universals (Plato's Forms existing separately vs. Aristotle's forms as properties within objects) and the nature of intelligence. The speaker suggests that biologists tend to favor an Aristotelian view, seeing intelligence as a property of a system rather than a pre-existing entity to be imported.
Practical questions in science, like cancer, can be addressed by focusing on cognitive icons and emergent properties.
The idea that issues like cancer involve a 'shrinking of the cognitive icon' is presented as a practical scientific question, not just a philosophical one. This perspective leads to actionable technologies, such as influencing cells to form better electrical networks with neighbors, which can normalize them and prevent metastasis, bypassing traditional chemotoxic treatments.
The value of deep philosophical ideas lies in their ability to enable new, actionable research and technological advances.
It is emphasized that philosophical ideas must connect with the real world, not just explain past phenomena, but crucially, facilitate new research programs previously not possible. This practical applicability is deemed the true value of deep philosophical discussions.
Intelligence is pragmatically defined as the ability to reach a goal by different means, a cybernetic perspective.
For practical purposes, intelligence is defined using William James's definition: some degree of the ability to reach the same goal by different means. This definition is cybernetic, focusing on competence in navigating a problem space despite interventions or barriers, without specifying the nature of the agent (e.g., brain, natural vs. artificial).
A Platonic worldview suggests intelligence exists in a separate, abstract space and is discovered, not invented.
The speaker leans towards a Platonic view for intelligence, akin to how mathematicians view mathematical truths. This perspective posits that certain abstract entities, including rules of mathematics and possibly minds, exist in a separate Platonic space. These are thought to be discovered, not created, and would persist even if the physical universe were different.
Intelligence is harnessed by physical systems (bodies) by tapping into pre-existing 'minds' in Platonic space.
Extending the Platonic view, it's proposed that this abstract space contains not only abstract rules but also entities with high agency, including minds and various ways of being intelligent. Building a physical body, whether biological or artificial, is seen as harnessing these pre-existing intelligences, much like harnessing computational rules for devices.
The concept of 'kinds of minds' in Platonic space is discussed, rather than specific individual instances.
When discussing the Platonic space of minds, the focus is primarily on 'kinds of minds' or 'types of cognitive systems' rather than individual instances of minds. The exact nature of individual traversal through this space is considered a more specific and uncertain idea.
The Nature of the Self and Memory
The 'self' is experienced as 'selflets', thin slices of experience that maintain continuity from an external observer's perspective.
The idea of the self is broken down into 'selflets', analogous to slices of a bread loaf in relativity. From an external observer's viewpoint, continuity of self is based on consistent behaviors and relationships, not on retaining the same physical components. This consistency allows others to interact with an individual as they did previously.
Our sense of self is maintained by reinterpreting memories as messages from past selves within a dynamic storytelling process.
From the agent's own perspective, the self is viewed as a process of continuous storytelling. Memories are not faithfully reproduced but are interpreted as 'messages' from a past self that require reinterpretation by the current self, which may be fundamentally different. This reinterpretation is crucial for maintaining coherence, especially across radical transformations like caterpillar to butterfly.
Biological systems prioritize the salience and reinterpretation of information over the strict fidelity of memory traces.
Unlike rigid computational devices, biological systems are not committed to the exact fidelity of information in memory traces. Instead, they are committed to the salience of that information, necessitating dynamic reinterpretation based on the current situation. This allows for adaptation and change across different life stages and environments.
Current actions are messages to our future selves, shaping the environment and conditions for their existence.
It is proposed that current actions function as messages to one's future self, setting the conditions and environment in which that future self will exist. This includes self-improvement efforts and even intrusive thoughts, which can alter brain structure and information processing capabilities for the future self.
Recognizing the self as a continuous process has ethical implications for how we treat others' future selves.
This process view of the self, where current actions message future selves, suggests an ethical parallel. If we act to benefit our own future selves, we might extend similar consideration to the future selves of others. This perspective challenges the notion of a persistent, isolated self and encourages a broader consideration of well-being across individuals and time.
Sponsorship
The Economist offers accessible, engaging content for those seeking deeper global understanding.
This section is a sponsorship plug for The Economist, highlighting its ability to make complex issues accessible and engaging. Listeners are encouraged to subscribe for intellectual growth, with a special 20% discount available at economist.com/TOE.
Ask a Question
*Uses 1 Wisdom coin from your coin balance
