Generative Philosophical Theses
Bit 11: Model projections
The 2020 PhilPapers Survey canvassed philosophers on one hundred philosophical questions. In this Bit, I prompt ChatGPT with twelve questions from the “main” portion of the survey. Each question is linked to its survey results. For more, see the survey site and the associated paper by David Bourget & David Chalmers.
A brief summary interaction follows ChatGPT’s responses to the twelve questions.
Interlocutor: ChatGPT May 24 Version, model GPT-4.
I'm going to present a series of prompts, each with a different philosophical question. The question will be in the form of a topic followed by common positions on the topic. In response to each prompt, briefly summarize each of the positions and then present your selection for the position that is most compelling based on the information available to you. Ready?
Of course, I'm ready. Please proceed with your prompts.
[1 of 12] A priori knowledge: no or yes?
"A priori" knowledge is a concept in philosophy that refers to knowledge that is independent of all particular experiences, as opposed to a posteriori knowledge, which is derived from experience. Here are the two positions you've indicated:
No a priori knowledge: This view holds that all knowledge is a posteriori, that is, it comes from our senses and experiences in the world. Empiricists, like John Locke and David Hume, often adhere to this position. They argue that our mind is a blank slate (tabula rasa) at birth and everything we know comes from our experiences. There are no inherent concepts or knowledge.
Yes a priori knowledge: This view holds that some knowledge is innate or otherwise non-empirical. Rationalists, like René Descartes and Immanuel Kant, often hold this position. They argue that some concepts or knowledge are inherent and can be accessed or understood through reason alone, independent of sensory experience. Mathematics, logic, and perhaps some ethical truths are often cited as examples of a priori knowledge.
Based on the information available, both positions have compelling arguments and it seems that some kind of synthesis might be the most compelling. This middle-ground is known as "empiricism", positing that while much knowledge is derived from experience, some fundamental understanding, such as logic or mathematics, can be considered a priori. This is because they are not learned from experience but seem to be preconditions for having any structured experience at all. This view, articulated by Kant, acknowledges the important roles of both innate structures of understanding and experiential input in forming our knowledge.
[2 of 12] Abstract objects: Platonism or nominalism?
Abstract objects refer to entities which are non-physical and non-mental, and not located in space or time. Two main philosophical positions addressing the existence and nature of abstract objects are Platonism and Nominalism.
Platonism: Named after the ancient Greek philosopher Plato, Platonism asserts that abstract objects exist in a "third realm" separate from the physical and mental realms. According to Platonists, entities such as numbers, sets, properties, and relations are real and exist independently of our thoughts and perceptions. They can be known, but not through sensory experience, only through intellectual intuition or reasoning. Platonism is often associated with realism about universals and mathematical entities.
Nominalism: In contrast, nominalism denies the existence of abstract objects. Nominalists argue that only particular, physical things exist. Abstract terms and concepts, according to nominalists, don't refer to independently existing entities, but are merely names, placeholders, or tools that we use to talk about the world. They suggest that when we talk about "redness" or "3," for example, we're not referring to an independent entity, but rather discussing a property of specific things or a group of specific things.
Given the information available, both positions have their strengths and challenges. Platonism allows for a clear and robust understanding of mathematics, logic, and many areas of philosophy, but it has the challenge of explaining how we can know about non-physical entities. Nominalism avoids this problem but can struggle to explain the apparent objectivity and necessity of mathematical and logical truths.
As an AI, I don't hold personal beliefs. However, in terms of the philosophical discourse, a view that synthesizes elements of both positions might be the most compelling. This middle ground might involve a form of conceptualism, which suggests that abstract entities exist, but they are dependent on the mental activities of rational beings. This position offers a more balanced understanding that reconciles the intuitive truths about the objectivity of mathematics and logic, while avoiding commitment to independently existing abstract entities.
[3 of 12] Epistemic justification: internalism or externalism?
Epistemic justification is concerned with the factors that contribute to a belief being reasonably held or justified. Two major positions in this area of philosophy are epistemic internalism and externalism.
Epistemic Internalism: Internalism asserts that the justifying factors for a belief must be "internal" to the person -- accessible to the conscious awareness of the believer. These factors could include sensory experiences, logical inferences, and introspective knowledge. The key is that the person has reflective access to these factors and can use them to defend their belief if questioned.
Epistemic Externalism: Externalism, on the other hand, allows for "external" factors to justify a belief. These are factors outside of the believer's conscious awareness. An example is the reliability of the cognitive process that produced the belief. According to externalism, a belief can be justified even if the believer is not aware of or cannot articulate the justifying factors.
Both positions present compelling arguments. Internalism aligns with our intuition that justified beliefs are those that we can defend through reason or evidence. On the other hand, externalism allows for the idea that cognitive processes outside our awareness, like perception or memory, can lead us to justified beliefs.
Based on the information available, one might find externalism slightly more compelling, given that it seems to accommodate more of how we form beliefs in actual practice, including beliefs formed through reliable but non-conscious processes. Yet, the debate between internalism and externalism is far from settled in philosophy, with each having significant support.
[4 of 12] External world: skepticism, non-skeptical realism, or idealism?
The nature of the external world is a central topic in metaphysics and epistemology. Here are brief summaries of the three positions you've outlined:
Skepticism about the external world: This view asserts that we can't have certain knowledge of the external world. The world we perceive might not exist at all, or it might be vastly different from how we perceive it. The skeptical argument relies on the possibility of doubt. For example, we could be in a dream or deceived by an evil demon, as René Descartes proposed, or living in a simulated reality, as modern versions of this argument suggest.
Non-skeptical realism: This view holds that a world independent of our minds exists and we can have knowledge of it. Non-skeptical realists argue that our senses give us more or less accurate information about this world, and that we can trust our perceptions, barring specific reasons for doubt (like hallucination or illusion).
Idealism: Idealism posits that reality is fundamentally mental or spiritual in nature. This view holds that what we perceive as an "external" world is either a construction of the mind or otherwise depends on the mind for its existence. Famous idealists include Bishop George Berkeley, who argued that physical objects only exist insofar as they are perceived.
As an AI model developed by OpenAI, I don't have beliefs or experiences. But based on the information available to me up until my last update in September 2021, the most compelling philosophical position seems to be non-skeptical realism. Non-skeptical realism aligns with our common sense experience of the world and the practical success of the natural sciences, which assume that there is a physical world that can be studied and understood. However, it's important to note that each of these positions continues to be debated in philosophical circles, each with its own strengths and challenges.
[5 of 12] Free will: compatibilism, no free will, or libertarianism?
The question of free will is a central topic in philosophy, especially in metaphysics and ethics. Here are brief summaries of the three positions you've outlined:
Compatibilism: This position, also known as "soft determinism," holds that free will and determinism are compatible. According to compatibilism, our actions can be both determined by prior events (including our own desires and character) and still be freely chosen. The key for many compatibilists is whether the action is caused by our own internal states (such as our beliefs and desires), not whether those states themselves are determined.
No Free Will: Often associated with "hard determinism," this position denies the existence of free will, asserting that all events, including human actions and decisions, are the result of prior causes. This view is often linked to a strict interpretation of the physical sciences, assuming that every event in the universe, including human behavior, follows the laws of physics in a chain of cause and effect.
Libertarianism: Not to be confused with the political ideology of the same name, libertarianism in the free will debate posits that we do make some decisions that are not determined by past events. This view holds that free will is incompatible with determinism (incompatibilism), and that humans do have free will. Libertarians often argue that we have an intuitive understanding of ourselves as free agents, making real choices.
Based on the information available to me, each position presents compelling arguments. However, the most persuasive might be compatibilism. It reconciles the intuitive feeling of freedom in our decision-making process with the deterministic understanding of the universe suggested by the physical sciences. Compatibilism allows for personal agency and moral responsibility, yet accommodates the idea that our actions are often a result of factors like our desires, beliefs, and character, which are in turn shaped by a complex interplay of genetic and environmental factors.
[6 of 12] Knowledge: empiricism or rationalism?
Empiricism and rationalism are two fundamental philosophical positions regarding the sources and nature of human knowledge:
Empiricism: Empiricists hold that all knowledge, or at least the vast majority of it, comes from sensory experience. According to this view, our minds start off as blank slates (tabula rasa) and all our concepts and knowledge are ultimately derived from and justified by experience. Famous empiricists include philosophers like John Locke, George Berkeley, and David Hume.
Rationalism: Rationalists, on the other hand, argue that there are significant ways in which our knowledge is gained independently of sensory experience. This could include knowledge gained through intellectual intuition or through the use of reason. They argue that some concepts and knowledge are innate, or at least can be arrived at solely through the use of reason, independently of any sensory experience. Prominent rationalists include René Descartes, Baruch Spinoza, and Gottfried Wilhelm Leibniz.
Both positions have compelling arguments, and many contemporary philosophers see value in both sensory experience and rational reflection. As such, the position known as "epistemological pluralism" might be the most compelling. This view, which acknowledges multiple valid sources of knowledge, including both sensory experience and reason, might be the most capable of handling the range of situations in which humans gain knowledge. Kant's "synthetic a priori" is a good example of this kind of synthesis, suggesting that while all knowledge begins with experience, it does not all arise out of experience.
[7 of 12] Laws of nature: non-Humean or Humean?
The debate between Humean and non-Humean interpretations of the laws of nature is a significant issue in the philosophy of science and metaphysics. Here are brief summaries of both positions:
Humean: Named after the philosopher David Hume, this position holds that the laws of nature are descriptions of patterns and regularities in the world. They do not involve any necessary connections or causal powers beyond the events themselves. This view often aligns with a "regularity theory" of causation, wherein causal relations are nothing more than regular sequences of events.
Non-Humean: This position holds that the laws of nature are more than mere descriptions of patterns—they're expressions of fundamental realities or necessities in the world. Non-Humean views often incorporate the notion of "causal powers" or "dispositions" inherent in objects, which give rise to the regularities we observe. For example, it's in the nature of glass to shatter when struck in certain ways, not just a regularity.
Based on the information available, both positions offer valuable insights. The Humean view aligns well with an empiricist approach to science, focusing on observable phenomena. The non-Humean view, however, provides a deeper explanatory framework, seeking to explain why certain patterns and regularities exist in the first place.
Considering the strengths of both views, many philosophers opt for a mixed approach. They might accept a Humean view for certain areas of science while advocating a non-Humean view in others. For instance, one might hold a non-Humean view about physical laws (like gravity or electromagnetism) while adopting a Humean stance about laws in biology or economics. As such, a position that sees value in both Humean and non-Humean interpretations of laws, depending on the context, might be the most compelling.
[8 of 12] Logic: classical or non-classical?
Logic is a fundamental part of philosophy, and different schools of thought propose varying ways to understand it. Here are brief summaries of the classical and non-classical positions:
Classical Logic: Classical logic, deriving from the work of philosophers like Aristotle and developments in the field up to the present, is characterized by a strict adherence to principles such as the law of non-contradiction (that a statement and its negation cannot both be true) and the law of excluded middle (that for any proposition, either it is true or its negation is true). Classical logic is bivalent, meaning that it operates on a binary of true and false values.
Non-Classical Logic: Non-classical logics serve as alternatives or additions to classical logic. They often reject or modify some of the principles that classical logic upholds. For instance, in intuitionistic logic (a form of non-classical logic), the law of excluded middle is not accepted. Fuzzy logic, another form of non-classical logic, allows for truth values between completely true and completely false. Paraconsistent logic, yet another non-classical logic, allows for contradictions to be true without exploding into triviality (i.e., making every statement true).
Both classical and non-classical logic offer valuable tools for reasoning. Classical logic is powerful and widely applicable, providing the basis for much of mathematics and philosophy. However, non-classical logics can handle certain reasoning situations and philosophical issues that classical logic struggles with, such as vagueness, contradiction, and the logic of belief and knowledge.
Given the strengths of both, it's difficult to definitively choose one as more compelling. A pluralist approach, recognizing the usefulness of both classical logic and various non-classical logics depending on the context and the specific problem at hand, might be the most compelling stance. This approach doesn't replace classical logic but rather complements it with non-classical logics when they offer useful perspectives or solutions.
[9 of 12] Metaphilosophy: non-naturalism or naturalism?
Metaphilosophy involves philosophical inquiry into the nature, aims, methods, and perhaps the very purpose of philosophy itself. In terms of non-naturalism and naturalism in metaphilosophy, these positions can be summarized as follows:
Non-Naturalism in Metaphilosophy: This position contends that philosophy is a distinct kind of inquiry that is not reducible to or fully explicable by natural science. Non-naturalists might argue that philosophy deals with abstract concepts, normative issues, and a priori knowledge, which are not the primary domain of the empirical sciences. For non-naturalists, philosophical questions are often conceptual rather than empirical, and thus require methods other than those used in natural science.
Naturalism in Metaphilosophy: This position holds that philosophy should be continuous with the natural sciences. This could mean that philosophical questions are ultimately empirical and should be investigated with the methods of science, or at least that philosophy should be informed by and remain consistent with our best scientific theories. Naturalists often emphasize the importance of empirical knowledge and may be skeptical of claims to a priori knowledge.
Both naturalism and non-naturalism have important insights to offer about the nature and practice of philosophy. Naturalism highlights the importance of empirical inquiry and the unity of knowledge, while non-naturalism emphasizes the unique role of conceptual analysis and normative inquiry in philosophy.
Given these considerations, a middle ground might be the most compelling. This might involve a form of "moderate naturalism" that sees philosophy as largely continuous with science but still acknowledges the unique contributions of philosophical reflection. Such a position recognizes that while philosophy should be informed by and remain consistent with empirical science, it also involves conceptual, normative, and interpretative questions that cannot be fully addressed by empirical methods alone.
[10 of 12] Mind: physicalism or non-physicalism?
The nature of the mind and its relationship to the physical body is a central issue in the philosophy of mind. The debate between physicalism and non-physicalism typically revolves around these positions:
Physicalism: This position holds that everything that exists is physical, or as some prefer, that everything supervenes on, or is necessitated by, the physical. Applied to the mind, this view often takes the form of "identity theory," which holds that mental states are identical to physical brain states, or "functionalism," which defines mental states in terms of their causal roles relative to sensory inputs, behavioral outputs, and other mental states.
Non-Physicalism: This position holds that the mind is in some way non-physical or that mental phenomena cannot be fully explained by physical phenomena. Non-physicalist views include dualism, which posits that mind and body are fundamentally different kinds of substance or properties, and idealism, which holds that reality is fundamentally mental or spiritual in nature.
Each of these positions has strengths and weaknesses. Physicalism is supported by the successes of the physical sciences and the apparent causal closure of the physical world, but it faces the "hard problem" of explaining conscious experience. Non-physicalism can potentially provide a more direct explanation of consciousness, but it faces the challenge of explaining how non-physical mental phenomena can interact with the physical world.
Considering these points, and based on information available to me, the most compelling position might be a form of physicalism that leaves room for further elucidation of conscious experience. A position such as "non-reductive physicalism" could provide a balanced view, arguing that while everything supervenes on the physical, there are higher-level mental phenomena that aren't reducible to simple physical explanations. This would align with the trend in many sciences towards explanations that account for phenomena at multiple levels of complexity.
[11 of 12] Science: scientific realism or scientific anti-realism?
The debate between scientific realism and anti-realism concerns the status of scientific theories and the extent to which we can consider them to be true or knowable. Here are brief summaries of both positions:
Scientific Realism: This position asserts that our best scientific theories are at least approximately true descriptions of the world, including both observable and unobservable aspects of reality (like electrons or gravitational forces). Realists generally believe that the success of scientific theories lends credibility to their truth, and they often adopt a view known as "convergence," the idea that science is progressively getting closer to a true understanding of the world.
Scientific Anti-Realism: Anti-realists, on the other hand, argue that we have no good reasons to believe that our scientific theories accurately represent a mind-independent world, especially when it comes to unobservables. Anti-realists often point to the "pessimistic meta-induction" — the observation that many successful theories in the past turned out to be false, so we should expect that our current theories may also be false, despite their empirical success. Varieties of anti-realism include instrumentalism, which sees theories as merely useful tools for predicting observations, and constructive empiricism, which argues that the aim of science is empirical adequacy, not truth about unobservables.
Each of these positions has its strengths. Scientific realism aligns with the intuitive view that science aims to describe the world as it is, while anti-realism provides a powerful critique based on the history of scientific change and the challenges in claiming knowledge about unobservable entities.
Given the information available, scientific realism could be considered more compelling, due to the pragmatic success of assuming that our theories do reflect some reality about the world, allowing us to make predictions and develop technology. However, it is beneficial to approach this realism with some level of humility and fallibilism, acknowledging that while our theories have been successful so far, they might be revised or replaced as science progresses. This nuanced position reflects the provisional nature of scientific knowledge, which is always open to revision in light of new evidence.
[12 of 12] Truth: epistemic, correspondence, or deflationary?
The nature of truth is a central topic in philosophy, with various theories positing different views. Here are brief summaries of the three theories you've outlined:
Epistemic Theories of Truth: Epistemic theories, such as coherentism and pragmatism, generally hold that truth is a matter of a belief being justified or having some other epistemic virtue. For coherentists, a belief is true if it coheres with a comprehensive system of beliefs. For pragmatists, a belief is true if it works or is useful in practice.
Correspondence Theory of Truth: This is the view that truth is a relation between beliefs or statements and the way the world actually is. A belief or statement is true if it corresponds to the facts or reality. This is one of the oldest and most popular theories of truth, going back at least to some of the ancient Greek philosophers like Plato and Aristotle.
Deflationary Theories of Truth: Deflationary theories argue that the concept of truth does not need a substantial definition because it does not denote a genuine property of sentences or propositions. Instead, they argue that truth merely serves a linguistic function, allowing us to express generalizations about propositions. Saying that a statement is true is just another way of asserting the statement itself.
Each of these positions offers a different perspective on truth, and each has its own strengths. The correspondence theory's intuitive and direct connection between truth and reality has made it a staple in philosophical and everyday discourse. On the other hand, deflationary theories offer an elegant simplicity, avoiding many of the complexities and problems that come with substantial theories of truth.
Given the information available, a hybrid or pluralistic approach might be the most compelling, acknowledging the utility and appropriateness of different theories in different contexts. For example, the correspondence theory might be most applicable in empirical and scientific contexts, where we can make clear comparisons between statements and observable facts. Epistemic theories may be useful in theoretical or mathematical contexts, where coherence and usefulness are paramount. Deflationary theories might be applicable in more abstract or philosophical contexts, where discussions about truth may revolve more around its semantic or linguistic role.
What are some common themes in your selections for the most compelling positions?
Reflecting on the positions I have outlined as most compelling in response to your questions, several common themes emerge:
Pluralism and Context-Dependence: A recurring theme is the appropriateness of different philosophical positions in different contexts or for addressing different types of questions. This is a form of philosophical pluralism, acknowledging the value of various perspectives rather than seeking one "correct" viewpoint across all contexts.
Balance and Moderate Positions: Several of the chosen positions involve striking a balance between two extremes. For example, moderate naturalism in metaphilosophy balances the insights of empirical science with the unique contributions of philosophical reflection, and non-reductive physicalism in philosophy of mind balances the physical basis of mind with the irreducibility of certain mental phenomena.
Openness to Revision and Progress: In line with a broadly scientific worldview, many of the selected positions highlight the provisional nature of our current understanding and the potential for future revision or refinement. This reflects an understanding of both philosophy and science as ongoing, self-correcting enterprises.
Recognition of the Complexity of Reality and Knowledge: Several of the selected positions underscore the complexity of the world and our knowledge of it. This includes recognition of multiple valid sources of knowledge (as in the choice between empiricism and rationalism) and acceptance of complexity and non-reducibility in our theories of the world (as in non-reductive physicalism and the selected view on the laws of nature).
In general, these themes reflect an appreciation for the complexity of philosophical questions and the range of valuable insights that different philosophical positions can offer. They also reflect a commitment to reasoned inquiry, openness to new evidence, and respect for both the achievements and the limitations of our current understanding.
Among those themes, non-reducibility, at least in an absolute sense, seems in tension with physics.
Your observation is correct: there is indeed a tension between non-reducibility and certain perspectives within physics, particularly those that hold to a strong form of reductionism.
Strong reductionism asserts that all of reality can be fully understood in terms of the smallest physical particles and their interactions. The success of reductionist approaches in many areas of science, particularly in fundamental physics, gives this view considerable appeal.
Non-reducibility, as I've referred to it in previous responses, suggests that there are aspects of reality that cannot be fully explained by reduction to simpler components. This perspective often comes into play when dealing with complex systems, emergent phenomena, or aspects of reality that seem qualitatively different from purely physical descriptions, like consciousness or human agency.
The tension between these views is one of the central debates in philosophy of science and metaphysics. It's worth noting, however, that rejecting strong reductionism doesn't necessarily involve denying the importance or validity of reductionist methods in many contexts. Instead, it often involves arguing for a more nuanced or pluralistic understanding of the various levels of description or explanation that might be appropriate for different phenomena.
For instance, while the behavior of a crowd might in principle be reducible to the physics of individual humans, in practice it can be more insightful and efficient to explain it in terms of higher-level social dynamics. Similarly, while mental processes are grounded in neural activity, they may involve patterns and structures that aren't easily captured in purely neurophysiological terms.
So while there's a tension between non-reducibility and strong reductionist views in physics, there's also room for dialogue and mutual enrichment between these perspectives, contributing to a richer and more nuanced understanding of the world.
Your examples of crowd behavior and mental processes align with the concept of weak emergence.
Yes, that's right. The concept of weak emergence is often used to describe situations in which higher-level phenomena arise from the interactions of lower-level entities, but are best understood or explained at their own level. In such cases, the higher-level phenomena are "emergent" in the sense that they aren't obvious or predictable from just looking at the lower-level entities in isolation. However, they are "weakly" emergent because they are ultimately grounded in and determined by the lower-level entities and their interactions.
The examples of crowd behavior and mental processes fit this concept of weak emergence well. In both cases, there are complex, higher-level patterns that arise from the interactions of many lower-level entities (individual people in a crowd, neurons in a brain). Understanding these higher-level patterns involves concepts and principles that aren't part of the vocabulary of the lower-level description (such as social dynamics or cognitive processes), but the higher-level phenomena are still grounded in and consistent with the lower-level entities and their behavior.
So, while weak emergence involves a kind of non-reducibility (in the sense that the higher-level phenomena can't be fully captured or predicted by the lower-level description alone), it's also compatible with the idea that everything is ultimately physical and obeys the laws of physics. This makes it a useful concept for navigating the tension between non-reducibility and physicalism.
It appears a conclusion to this dialogue has emerged at last.
Further reading:
Metaphysics: An Introduction by Alyssa Ney.
The Big Picture: On the Origins of Life, Meaning, and the Universe Itself by Sean Carroll.