And that question is another category error. It's a dream; it doesn't take place at all. It happens in the magical land of unicorns. — unenlightened

But you have an experience of seeing a tree in your dream. That experience is like the experience of perceiving a tree. If the first is a mental image, why isn't the second experience?

But you have an experience of seeing a tree in your dream. That experience is like the experience of perceiving a tree. — Marchesk

No. bite my bullet first. Dreams do not happen anywhere; they are not real. If I dream of a tree I have not seen or experienced a tree; there is no tree, it was just a dream. Start being strict with your language, and everything indirect will disappear, because it is all a series of category errors, and literalised metaphors.

Start being strict with your language, and everything indirect will disappear, because it is all a series of category errors, and literalised metaphors. — unenlightened

It's not about being 'strict' with language, it's about using it in a particular way. You're trying to enforce a use of 'see' where it is not normally so restricted and you've not yet provided any argument as to why. I can guarantee that most people, when describing a dream, would quite happily say "I saw a tree", and everyone to whom they're speaking would understand them. If you want to claim that usage is confused or leads to problems you'll have to show what those problems are. And you'll have to be able to defend them pretty robustly. Trying to restrict normal language use (for whatever reason) is no trivial undertaking.

It's not about being 'strict' with language, it's about using it in a particular way. You're trying to enforce a use of 'see' where it is not normally so restricted — Isaac

Yes it is very common. One talks of seeing 'with the minds eye', and you can feel free to talk like that, but do try to remember that the mind does not have a literal eye, and there are not images in heads. You can talk however you like, but if you make category errors you will fall into folly and indirect realism is a very venerable folly, that has deluded philosophers for a long time, so I recommend folks to pay close attention to it so as to see where their thoughts are going astray. If you want to carry on though, then carry on.

Earlier i was accused of focussing on the word 'experience', and now it is the word' seeing'. People are heavily invested and don't want to 'see' things differently. "See" what I did there?

but if you make category errors you will fall into folly and indirect realism is a very venerable folly, that has deluded philosophers for a long time — unenlightened

I couldn't agree more. What I'm enquiring about is this list of follies. What problems arise from speaking this way? I've listed a few problems I think arise from not speaking this way, I was hoping you could provide a few problems which arise from speaking the way you recommend against.

Talk of thoughts being "about" things, in a sense needing subsequent unpacking, can too easily become talk of the thoughts "representing" the things, in a sense more suited (category error?) to words and pictures. Thoughts-in-the-head become pictures-in-the-head. But such a progression is unnecessary. Thoughts are "about" things in that they are the brain so shivering its neurons as to adjust its readiness to act on those things. Conscious thoughts, in particular, adjust its readiness to select among symbols for pointing at those things. This kind of thought is thus (whether waking or dreaming) thought "in" symbols, and consequently prone to making us think (mistakenly, though often harmlessly) that the symbols are in our heads.

I was hoping you could provide a few problems which arise from speaking the way you recommend against. — Isaac

This whole thread does that. And a deal of it happens in question form - "where is experience?" and so on. But I don't want to go all through it again. It might be instructive to make a list, but I think I've annoyed myself and everyone else enough already.

This whole thread does that. And a deal of it happens in question form - "where is experience?" and so on. — unenlightened

I'm not seeing the problem with such a question, apart from you (and others of your opinion) declaring it to be. By 'problem' here I'm meaning something like a failure to grasp some otherwise useful concept, a sense of confusion or distress resulting from the language use, an actual failure in prediction or potential for such failures. Something like that. But obviously if the discussion is annoying you there's no good cause to continue.

Let's look at two claims.

(A) Perception is an active relationship between a body and its environment.
(B) Perception results from an active relationship between a body and its environment.

In a context outside this debate, we would maybe be able to agree that (A) and (B) are saying much the same thing. If people are really disagreeing over whether humans can see tables, that is extremely silly, and direct realists shouldn't be throwing "are you saying you can't see the table?" gotchas at indirect realists.

(B) admits of a minor modification that substantially changes the metaphysical intiutions associated with it.

(B1) A perception results from an instance of an active relationship between a body and its environment.

(A) and (B1) look compatible to me; (A) states that perception is a dynamic process between a body and its environment, (B1) splits the process into process components; minimally, distinct events of perception.

I think a direct realist and an indirect realist individuate perceptions very differently when arguing in this context. For a direct realist, it seems to me a perceptual event is an instance of the relationship between one's body and one's environment, for an indirect realist a perceptual event results from an instance of the relationship between one's body and one's environment.

Let's just grant that perception is model based, so what we're seeing is a subset of all the possible aspects of the environment, and we're seeing it in terms of our affordances - or in terms of proposed interventions for goals in other vocabulary (@Isaac Friston even approves of Gibson's theory of perception, which is a form of direct realism, so it's no so clear cut that indirect realism is the only way to be consistent with neuroscience). Our perceptions are samples from a model.

But that model itself is a direct relationship between our body and our environment; it is the medium of perception, and perception itself is a mediation of body and world. The samples from it are instances of a (filtered; yes, partial, yes; incomplete, yes; flawed, yes; predictive/inferential, yes) direct relation. (that model doesn't really care whether a state is internal or external to the body, it deals with both because predictive models of the effects of our actions are concerned with that which is external to our body but are still part of our perceptual events and even phenomenal content!)

Discussions like this on the forum rarely get off the ground because we individuate instances of perception differently, and people with intuitions that perception is model based have a habit of concluding that the representational varieties of indirect realism are the only way forward; even when the representation/modelling that constitutes perception itself is a direct relationship between body and environment

Discussions like this on the forum rarely get off the ground because we individuate instances of perception differently, and people with intuitions that perception is model based have a habit of concluding that the representational varieties of indirect realism are the only way forward; even when the representation/modelling that constitutes perception itself is a direct relationship between body and environment — fdrake

What would constitute indirect for a direct realist? Going back to the neural implant, let's say when you close your eyes the implant receives radio signals from a camera mounted on a robot moving about some environment. The implant translates that to electrical signals the brain can interpret as images, and the result is a visual perception of what the robot camera is recording.

The reason for brining that up is to ask whether any possible process of perception could be indirect for a direct realist. Because if the answer is none, then the direct realist is playing a word game.

I wish to be annoyed more.

The reason for the "venerable folly" of indirect realism is because illusions and hallucinations raise the possibility that perception isn't what we naively take it to be. Of course you can say that "seeing" used properly means only veridical perception. But that doesn't address the issue.

The possibility that perception is something other than direct awareness needs to be dealt with. Insisting on using language correctly won't make the issue go away because, as Michael pointed out, the language is based on a naive realist understanding, which could be mistaken.

The reason for the "venerable folly" of indirect realism is because illusions and hallucinations raise the possibility that perception isn't what we naively take it to be. — Marchesk

But this is not true. Humans have known about these experiences since the earliest times, and we know about them individually from an early age.

Indirect realism is much more historically specific, and has its roots in specific ways of thinking about what it means to perceive, what it means to be a person at all.

It doesn't follow from illusions and hallucinations.

Great post. I guess my angle is to ask why exactly some people have the indirect realist intuitions. I mean, it's not just like ice cream. It's cultural.

What would constitute indirect for a direct realist? Going back to the neural implant, let's say when you close your eyes the implant receives radio signals from a camera mounted on a robot moving about some environment. The implant translates that to electrical signals the brain can interpret as images, and the result is a visual perception of what the robot camera is recording. — Marchesk

I can't answer for direct realism generally. But I would say that an indirect instance of active perception would have its percept as an output of the process of active perception; as if the process of perception produces phenomenal and mental content associated with perceptions; in a diagram, perceptual relation$\rightarrow$phenomenal and mental content of perception. The associated intuition is a sequential ordering of perception to perceptual content (related to post-hoc thematisation/schematisation as @jamalrob channeled photographer with in another thread)

Conversely, I would say that a direct instance of active perception would have its percept as a component of the process of active perception; as if the phenomenal and mental content associated with perception is a part of the perceptual modelling relation between body and environment; in a diagram, phenomenal and mental content of perception $\subset$ perceptual relation. The associated intuition is that perceptual content (the phenomenal/mental stuff) occurs within a relational event of perception.

Seeing an object is not an object, rather it's an occurrence.
That's one possible category mistake.

Indirect realism is much more historically specific, and has its roots in specific ways of thinking about what it means to perceive, what it means to be a person at all. — jamalrob

The problem of perception in general goes back to ancient philosophy, and it's not limited to the Greeks.

They Cyreneacs used the skeptical arguments around perceptual relativity to say that we can only know the sensory impression and not the external cause. Therefore for them, the proper linguistic use was "I am sweetened" instead of "The honey is sweet". Or "I am whitened" instead of "The table is white".

Indirect versus direct realism may be historically specific, like the current debate over consciousness, but the wider problem of perception is not. As soon as people started asking philosophical questions, perception became an issue. Or maybe because of issues with perception people started asking those kinds of questions.

I can't answer for direct realism generally. But I would say that an indirect instance of active perception would have its percept as an output of the process of active perception; as if the process of perception produces phenomenal and mental content associated with perceptions; in a diagram, perceptual relation->phenomenal and mental content of perception. I would say that a direct instance of active perception would have its percept as a component of the process of active perception; as if the phenomenal and mental content associated with perception is a part of the perceptual modelling relation between body and environment; in a diagram, phenomenal and mental content of perception ⊂⊂ perceptual relation. — fdrake

I don't think this distinction properly addresses what it is that direct and indirect realists are trying to say. They are trying to argue that we either can or can't trust that things are (independently) as we see them to be (that's the epistemological problem of perception). Whatever it is that the direct realist means by seeing something directly, it follows that if perception is direct then if we see something to be red then it follows that it has the (independent) property of being red. And so if something seen to be red doesn't have the (independent) property of being red then perception isn't direct.

That's why I think that your approach (and unenlightened's approach, and jamalrob's approach) seem to sidestep the substance of the disagreement. We want to know if the properties present in experience (a red colour, a sweet taste, a round shape) are (independent) properties of external world objects or if they're properties only of the experience (whatever it is that experience is). Does experience show us the "true" (read: independent) nature of the world.

That's why I think that your approach (and unenlightened's approach, and jamalrob's approach) seem to sidestep the substance of the disagreement. — Michael

There seems to be a strong temptation on this forum to think that if only the terms can be used the proper way, the philosophical issue goes away. I take it that's Wittgenstein's shadow cast large over these sorts of disagreements.

However, the nature of perception has persisted as an in issue in philosophy across many time periods, cultures and languages, so it probably can't be resolved by just figuring out proper linguistic use, since it's a problem of perception, not language.

We want to know if the properties present in experience (a red colour, a sweet taste, a round shape) are (independent) properties of external world objects or if they're properties only of the experience (whatever it is that experience is). — Michael

What's wrong with the relational approach, that you and Marchesk might both be familiar with from other posts of mine, about colour realism and other things? Fire engines are red because they have properties that produce the experience of red in human beings, i.e., in perceivers that sense those properties in particular ways. Again, I think this shows how odd the question you're asking actually is.

Perceivers always have a perspective, in a general sense. That's what perceiving is.

Don't give in to the thought: in that case we can't say that fire engines really are red. Reject it. Banish it forever.

That's why I think that your approach (and unenlightened's approach, and jamalrob's approach) seem to sidestep the substance of the disagreement. We want to know if the properties present in experience (a red colour, a sweet taste, a round shape) are (independent) properties of external world objects or if they're properties only of the experience (whatever it is that experience is). — Michael

We want to know if the properties present in experience (a red colour, a sweet taste, a round shape) are (independent) properties of external world objects or if they're properties only of the experience (whatever it is that experience is).

The red I see is not by itself a property of the apple. Granted.

Nevertheless; the red I see is not independent of the properties of the apple.

The properties of the apple inform my perceptions of it. But they do not fully determine my perceptions of it. The perceptual content I have is part of a modelling relationship between myself and the apple, so the perceptual content is independent of neither; even though the apple properties are not existentially dependent upon my perception.

My expectations of what the apple properties are are dependent upon previous experience and the apple, though! If my perception of the apple is accurate, my perception of the apple is informative of its properties, and so its properties are not (statistically) independent of my perceptual content and actions regarding it. How my perception works in general is not dependent upon the apple's properties.

I meant to ask you when you made this distinction before; what do you see as the relationship between the apple properties and the red I see?

Don't give in to the thought: well then we can't say that fire engines really are red. Reject it. Banish it forever. — jamalrob

You can if you care about which properties are real (mind-independent), and which ones are created by the perceiver. Isn't that what science tries to do? How can we say the fire engine is really red if we know visible light is only a small part of the electromagnetic spectrum, and if we had eyes that could see other parts, it might not look red?

Part of the ancient skeptical argument was noting that animal senses differ from our own. So no, we can't just say the world is how it appears to us, since it can appear differently to other animals.

Anyway, I care about what's real, to the extent we can know.

Huh?

The relational approach answers all this. Red things are red, but only to certain perceivers. I don't think you understand my mockery of the question about whether or not the things really are red.

So there are no red things. I agree, it's a relational property of perceivers. Color isn't a property of objects.

Therefore, when we perceive a red fire engine, the redness is not a direct awareness of the object, since objects have no color.

No.

I dunno, maybe fdrake can explain things better.

I dunno, maybe fdrake can explain things better. — jamalrob

Red things are red, but only to certain perceivers. — jamalrob

Red things being red only to certain perceivers is the same thing as what the ancient skeptics were saying. Honey isn't sweet, it's only sweet to tasters. Sweetness isn't a property of the honey, it's a property of tasting. "I am sweetened".

I meant to ask you when you made this distinction before; what do you see as the relationship between the apple properties and the red I see? — fdrake

The surface of the apple reflects light at a certain wavelength, that light stimulates the eyes, the eyes send a signal to the brain, the visual cortex of the brain is activated, and we have an experience that we describe as "seeing a red apple". So I suppose I would say that the relationship is simply causal (a term I've seen elsewhere on the topic is "causal covariance").

What would constitute indirect for a direct realist? Going back to the neural implant, let's say when you close your eyes the implant receives radio signals from a camera mounted on a robot moving about some environment. The implant translates that to electrical signals the brain can interpret as images, and the result is a visual perception of what the robot camera is recording.

The reason for brining that up is to ask whether any possible process of perception could be indirect for a direct realist. Because if the answer is none, then the direct realist is playing a word game. — Marchesk

What is the difference between your example of an implant and hearing someone speaking on the phone? Hearing them in person still requires air molecules as the medium for their voice to travel. We still hear their voice, and understand what they say. So again, what is being lost to say that indirect vs direct are somehow different? If hearing them in person is more direct than hearing them on the phone and hearing them on the phone is more indirect than hearing them in person, then we are simply talking about degrees of indirectness/directness. It would be the amount of causal steps it took to get to your awareness of it that we would talk about how direct and how indirect our knowledge of something is. How many steps qualify the process to be direct vs indirect? Does the process take more than one step? Are the steps themselves products of our mind?

(A) Perception is an active relationship between a body and its environment.
(B) Perception results from an active relationship between a body and its environment. — fdrake

Glad to see someone taking my point that the words we are discussing need to be defined, seriously. Everyone is talking past each other because we haven't defined "perception", "experience" "awareness", "consciousness", "indirect vs. direct", etc. No wonder the thread has become what it has - total confusion.

If the goal here is to find some level of agreement, then we need to define what it is that we are talking about, or else we will always disagree, as disagreeing is the result of talking past each other.

It's probably a bit of both. The state of the environment is a priori to our perceptions of it. So while our perceptions can be about the relationship between our body and the environment, I would say that the part of our perception that is about the environment is delayed as it takes time for someone's voice to reach our ears and time for our minds to interpret that a voice is being heard and what the voice is saying. But the actual process of the mind is in real-time - what it is about isn't.

Yeah, but we were not wrong because we trusted our senses... but because we inferred things from them, that we had no real justification to infer.

There's no need for example to assume flat earth from the surface we see being mostly flat... because a circle with a big radius also looks flat from the perspective of a smaller being. Both flat earth and spherical earth fit that observational data, but we just assumed that it had to be flat for a time (for understandable reasons, but that is not the fault of the senses).

There is no way to verify what we perceive, with some other real world data... like I said earlier in the thread, we only started to make scientific progress when we started to take observations seriously. — ChatteringMonkey

Exactly! Our senses don't lie. Our interpretations of what we observe are the problem. A bent straw in water is exactly what you are suppose to see given that we see light, not objects. We infer objects from the information in light. The problem occurs if you think that you see objects.

Seeing the Earth as flat vs round is seeing the Earth from different perspectives. It seems to me that when you place yourself apart from the thing you are talking about that you attain some real sense of the real shape of the world. It is only when you go out in space - separating yourself from the Earth that you see it's true shape. This is the distinction between subjective views and objective ones.

What's wrong with the relational approach, that you and Marchesk might both be familiar with from other posts of mine, about colour realism and other things? Fire engines are red because they have properties that produce the experience of red in human beings, i.e., in perceivers that sense those properties in particular ways. Again, I think this shows how odd the question you're asking actually is.

Perceivers always have a perspective, in a general sense. That's what perceiving is.

Don't give in to the thought: in that case we can't say that fire engines really are red. Reject it. Banish it forever. — jamalrob

I think this is equivocation. The fire engine "being red" in the sense of "having properties that produce the experience of red in human beings when seen" isn't the same thing as being red in the context of a visual experience, just as "having properties that produce the experience of sweetness in human beings when eaten" isn't the same thing as being sweet in the context of a taste experience, and just as "having properties that produce pleasure in human beings when pressed against one's genitals" isn't the same thing as pleasure in the context of an orgasm.

The inverted spectrum hypothesis is proof enough of this. Two different people with different bodies can have a different kind of experience when responding to the same stimulus. One person sees it as red, the other as orange. Given the coherency of this, what is meant by "red" and "orange" isn't what is meant by "having a surface that reflects light at a wavelength of X nm". We can agree on the nature of the light reflected by the apple but still disagree on its colour. And it's not just a hypothesis. Tetrachromacy, or the infamous black-blue/white-gold dress, are practical examples of that.

Exactly! Our senses don't lie. Our interpretations of what we observe are the problem. A bent straw in water is exactly what you are suppose to see given that we see light, not objects. — Harry Hindu

Our brains could have evolved to correct for that, if it had been advantageous enough. Our brains do corrections for lighting conditions, and of course sometimes our brains get angles, lighting or motion wrong. Thus the various visual illusions.

The image on the retina is 2D, so the brain has to be making some inferences about depth as it produces the perception.

Sure, but I don't see how that goes against my point. Fire engines are red to most people, if you like. It doesn't matter. The point is not that red is some transcendent fact of the fire engine, but that a perceiver is in an active relation with its environment, in which perception depends on both.

There is probably a spectrum of terms that vary gradually in how much we can conventionally say, "this looks/sounds/tastes X to me" as opposed to "this is X"

Our brains could have evolved to correct for that, if it had been advantageous enough. Our brains do corrections for lighting conditions, and of course sometimes our brains get angles, and lighting and motion wrong. Thus the various visual illusions.

The image on the retina is upside down and 2D, so the brain has to be making some inferences as it produces the perception. — Marchesk

It seems to me that in order to say that the brain gets stuff "wrong" is implying that you know what is "right". How did you know what is right or wrong if not using your brain - directly or indirectly?

I would be willing to bet that this thread was started because you didn't understand Jamalrob's usage of certain terms - particularly the distinction being made between "direct" and "indirect". Others have been displaying a similar reaction to his article.

It is because the words haven't been clearly defined. Jamalrob's usage isn't the same as other's usage, or understanding. Jamalrob is not getting at how they think of things and how they think of things isn't getting at Jamalrob's usage. When words are used ambiguously nothing can every been found to be agreed on because the symbols themselves haven't been provided a concrete meaning to them for the purpose of this discussion.