Representationalism in Perception and Belief
A chat about a terminological ambiguity
For a breezy introduction, listen to a generated Deep Dive podcast here.
August 13, 2023
Prompt:
Sometimes representationalism, such as in perception, is the position that the mind perceives only mental representations. Other times, it is the broader position that mental states have propositional content. Does the representational theory of perception entail the representational theory of mind?
ChatGPT:
The relationship between the representational theory of perception (often just called “representationalism”) and the representational theory of mind is complex, but they are deeply intertwined. Let’s unpack this a bit:
Representational Theory of Perception (Representationalism about Perception): This is the view that conscious perceptual experience is a matter of having mental representations with a certain content, and the qualitative character of these experiences (i.e., what it’s like to have them) is determined by the representational content of these states. For instance, when you see a red apple, your experience of redness is due to your mind’s representation of the apple’s color.
Representational Theory of Mind (or Cognitive Representationalism): This is a broader view, and it posits that mental states in general (not just perceptual experiences) are representational in nature. For example, beliefs and desires, according to this view, involve mental representations with propositional content. If I believe it’s going to rain, I have a mental representation with the content “It’s going to rain.”
Now, to your question: Does the representational theory of perception necessarily entail the representational theory of mind?
While the two theories are conceptually distinct (one could, in principle, hold the former without committing to the latter), in practice they often go hand in hand. Here’s why:
Shared Theoretical Foundations: Both theories draw on the idea that the mind operates in terms of representations. If you’re inclined to think that perception operates via representations, you might be naturally predisposed to think that other mental processes (like belief formation or problem-solving) also involve representations.
Explanatory Consistency: If one adopts a representational account for perception, it provides a consistent framework to explain other mental phenomena in terms of representations. This makes the broader representational theory of mind an attractive position.
Empirical Considerations: Cognitive science and neuroscience research have provided considerable evidence suggesting that many mental processes involve the manipulation of representations. If one accepts that the brain processes perceptual information in terms of representations, it seems plausible that it would also process other types of information (like beliefs or desires) in a similar manner.
However, it’s worth noting that while many philosophers and cognitive scientists accept some version of representationalism about both perception and cognition, there are disagreements about the specifics. For instance, some might argue that while perception involves non-propositional representations (like maps or images), cognitive states like beliefs involve propositional representations.
In conclusion, while the representational theory of perception doesn’t necessarily entail the representational theory of mind, there's a strong theoretical and empirical motivation for many to adopt both positions for the sake of explanatory consistency and coherence.
Prompt:
It’s not obvious exactly how they go together. Consider the belief that it is going to rain. Are the intentional objects of this belief the very constituents of the representation?
ChatGPT:
You’re right in noting that the relationship between the representational content of a belief and its intentional objects can be complex. Let's delve into this a bit.
In philosophy of mind, the term “intentional object” refers to the object or content towards which a mental state (like a belief, desire, or thought) is directed. For instance, when one has a belief about the Eiffel Tower, the Eiffel Tower is the intentional object of that belief.
Consider your example: the belief that it is going to rain. The intentional object of this belief is the state of affairs of it raining. But how is this state of affairs represented in the mind?
There are different ways to understand this:
Direct Realism: Some might argue that the intentional objects of beliefs are directly present in the mind. On this view, when one believes that it’s going to rain, the state of affairs of it raining is somehow directly “grasped” or “apprehended” by the mind. This position faces challenges, especially when we consider beliefs about non-existent objects (e.g., unicorns).
Representation via Propositions: A more common view in the philosophy of mind and cognitive science is that beliefs have propositional content. On this view, the belief that it's going to rain involves a mental representation with the content "It's going to rain." This content can be seen as a proposition—a statement that has a truth value. The belief is directed towards its intentional object (the state of affairs of it raining) by virtue of representing that proposition. The constituents of this proposition (like the concept of "rain") play a role in determining the intentional object of the belief.
Structured Representations: Some might argue that mental representations are structured, much like sentences. On this view, the belief that it’s going to rain involves a mental representation with constituents that stand in certain relations to one another, similar to the subject and predicate in a sentence. The intentional object is determined by the way these constituents are related within the representation.
To answer your question directly: In many representationalist views, the constituents of the representation play a crucial role in determining the intentional object of a belief. However, the intentional object itself is not identical to these constituents; rather, it is the state of affairs in the world that the representation is about.
It’s worth noting that the nature of representation and intentionality is a deeply debated topic in philosophy of mind, and the above are just a few prominent views on the matter.
Prompt:
If the intentional object of belief is not itself a constituent of the representation, then why is it said in the representational theory of perception that what is perceived is the representation and not the represented objects?
ChatGPT:
The confusion you’re highlighting stems from a subtle distinction in the philosophy of perception and mind. When we discuss representationalism in the context of perception, it’s essential to differentiate between the phenomenological aspect of perception (i.e., what it’s like to have a particular perceptual experience) and the relational aspect (i.e., how our perceptual experience relates to the external world).
In the representational theory of perception:
Phenomenological Aspect: The idea is that the qualitative character of our perceptual experiences (what it's like to see red, hear a bird, etc.) is determined by the representational content of our perceptual states. Thus, in a certain sense, we “directly perceive” representations, meaning that our immediate conscious experience is of the representational content.
Relational Aspect: Representationalism does not deny that our perceptual states are about the external world or that they can accurately or inaccurately represent objects in the world. So, while we might say that our immediate perceptual experience is of the representational content, this content itself is about (or represents) objects in the external world.
Now, when considering beliefs, the situation is slightly different:
Phenomenological Aspect: Beliefs don’t typically have the same kind of qualitative character that perceptual experiences do. When I believe that it’s going to rain, there’s not a particular “what it’s like” associated with that belief in the same way that there is a “what it’s like” to see the color red.
Relational Aspect: Like perceptual states, beliefs have representational content that is about the world. The belief’s content can accurately or inaccurately represent the world, and this content directs the belief towards its intentional object.
When philosophers say that in perception “we perceive representations,” they often mean that the qualitative character of our perceptual experience is determined by the representational content. They are not suggesting that we don't also perceive external objects, but rather that our conscious, phenomenological access to these objects is mediated by representations. This is why representationalism about perception is sometimes presented in opposition to direct realism, which posits that we directly and immediately perceive external objects without any representational mediation.
In conclusion, while both beliefs and perceptual experiences involve representations, the ways in which those representations factor into our conscious experience and our relation to the world are different in each case.
Prompt:
I recognize that there is iconic non-propositional representation, so my ontology of mental states with propositional content is insufficient. The trouble, however, is that representation cannot mean only having aboutness or intentionality. It also means re-presenting, or presenting again; a representation is re-presentation, a repeated presence. As an entity with a referent, this is fine. But as an intentional object of content, we have a problem. If the perception representationalist says that we do not see the objects or presentations themselves directly, because what we do see directly are the representations, then I cannot agree. Even the perception direct realist admits that there are mental states with referential content. The divisive question is whether the referents of the perception are the representations or the presentations.
How about this for compromise? I accept the representationalist theory of mind, meaning that there are mental states with referents, but deny that the referents of the mental states, including the phenomenal qualities of perception, are these representations. For this reason, I’d reject the representationalist theory of perception. I still consider these names unfortunately misleading, but it is understandable given a history in which from Locke to the present, most philosophers accept both theories.
As for whether the phenomenal properties are in the object or the representation, I’d say, with Shoemaker, that they are in the object. But of course the representation importantly determines which of the phenomenal properties of the object we are referring to. If the flower looks like this shade of green to me, but maybe otherwise to a bee, then what determines its appearance to me is conditioned by both the flower and my iconic representation, but the appearance itself remains a part of the flower. When a girl cares about her appearance, she cares about an aspect of herself, not about the representations of her appearance in those who look at her, though of course both are determinants of what she looks like to those who see.
ChatGPT:
Your position is nuanced and offers a thoughtful middle ground between representationalist and direct realist views on perception. Let’s unpack your stance and highlight some of its features:
Representational Theory of Mind but not Perception: You accept that there are mental states with referents (i.e., aboutness or intentionality). This allows for cognitive states, like beliefs and desires, to represent aspects of the world without requiring that phenomenal perceptual experiences be understood purely in terms of representations. It’s a distinction that provides a robust foundation for many cognitive processes while acknowledging the unique nature of perception.
Phenomenal Properties and Objects: You adopt a view reminiscent of some direct realist positions, suggesting that the phenomenal properties we perceive (like the shade of a flower) are properties of the external objects themselves. This challenges a strict representationalist view where the phenomenal properties we experience would be properties of our internal mental states or representations.
Role of Representation in Perception: You don’t deny the role of representation in perception. Instead, you view representations as mediating or conditioning our perceptual access to the phenomenal properties of objects. This is a way to acknowledge the influence of our perceptual apparatus and cognitive processing on our experience of the world while still holding that the phenomenal qualities we perceive are real properties of external objects.
Appearance and Representation: The example of the girl caring about her appearance serves as a poignant illustration. On your view, the girl's appearance is an aspect of her, even though the way it is perceived by others may be conditioned by their representations. The appearance is a real-world property, while the representations determine or influence how that property is perceived.
Your position strikes a balance between acknowledging the role of representation in mediating our perceptual experience and insisting that the qualities we perceive are genuine properties of external objects. In philosophical parlance, you’re advocating for a kind of “relational” view of perception, where both the object and the perceiver (and their representations) play roles in determining perceptual experience.
This sort of position is quite attractive to many in the philosophy of perception because it captures the intuition that we're genuinely "in touch" with the world around us while still acknowledging the vast body of empirical evidence showing that our perceptual experience is shaped, conditioned, and sometimes even distorted by our cognitive and neural apparatus.
As you pointed out, the terminology and labels can sometimes be misleading or inadequate to capture the richness and subtleties of these views. Philosophical discussions about perception have evolved over centuries, and the positions have become more refined as our understanding of the mind and brain has grown. Your stance is a testament to this nuanced evolution.