Indirect realism is the view that what we see is the representation. The alternate is that what we see is the tree, and that we see the tree by constructing a representation of the tree. — Banno

Seeing a representation of a tree and seeing a tree are not mutually exclusive, exactly as feeling pain and feeling my hand burning are not mutually exclusive. “I experience X” doesn’t just mean one thing.

The meaning of “see” in “I see a representation of a tree” is the meaning of “see” in “I see a cat in my dream” and the meaning of “hear” in “I hear a representation of thunder” is the meaning of “hear” in “the schizophrenic hears voices” and the meaning of “feel” in “I feel a representation of a rock” is the meaning of “feel” in “I feel pain”.

Naive realists reject the existence of mental percepts, indirect realists accept the existence of mental percepts, and then so-called “non-naive non-indirect” realists accept indirect realism but call it something else and then invent some strawman (“percepts of percepts”) to stand in for indirect realism.

In accepting the existence of mental representations you’re arguing for indirect realism, but for reasons unknown are refusing to call it what it is.

Not that the label really matters. Call it “Banno realism” if you want. Either way it entails the epistemological problem of perception and the warrant for scepticism. That’s the philosophical concern that gave rise to the dispute between direct and indirect realism in the first place.

But how does the cognition "see" anything? It is the mental image, the representation of the distal object, which is the "seeing"; the sensory perception. The cognition does not have its own set of sensory organs with which to perceive the mental image. — Luke

I see things when I dream and the schizophrenic hears voices when hallucinating. Sensory organs are not involved. Seeing and hearing occurs when the visual and auditory cortexes are active (which is why the cortical blind can’t see even though they have functioning eyes).

When I look at something I can see its qualities: height, width, shape, colours, textures; I don't need to infer those properties. — Janus

Those are the qualities of the experience, not the properties of the distal object.

Experience exists within the brain. Distal objects exist outside the body. Therefore distal objects (and their properties) do not exist within experience. Everything that is present in experience (smells, tastes, colours, etc.) is a mental phenomenon.

You might want to argue that some of these mental phenomena (e.g. visual geometry) resemble the properties of the distal object, but it is nonetheless the case that the mental phenomena is the intermediary from which the properties of the distal object are inferred.

One of the conundrums with indirect realism is that it seems to start as direct realism, where the scientist assumes he sees the world exactly as it is, then concludes from what he's observed that he's not seeing the world exactly as it is. How do you deal with that problem? — frank

I addressed this in a previous comment.

Firstly, if direct realism is true then scientific realism is true, and if scientific realism is true then direct realism is false. Therefore direct realism is false given that it entails a contradiction.

Secondly, given that scientific realism entails the existence of objects that cannot be directly observed (e.g. electrons), it is not a contradiction – performative or otherwise – for an indirect realist to be a scientific realist. Presumably even the direct realist can trust in a Geiger counter despite not claiming to directly see radiation. Direct perception is not required to accept something as true.

The experience of pain does not have pain as an object because the experience of pain is identical with the pain. Similarly, if the experience of perceiving is an object of perceiving, then it becomes identical with the perceiving. Just as the pain is identical with the experience of pain, so the visual experience is identical with the experience of seeing.

The painting of some fictional landscape is identical with the painting – and identical to the paint itself. But the painting is still of something. So even if the experience of something is identical with the experience it is nonetheless the experience of something.

And what is the intentional object of perception if not whatever follows the word "of"? Perhaps Searle is being ambiguous with the phrase "object of perception".

Not that this really matters, as per my comment here, but I thought I should address it anyway.

I’m with you. As I’ve repeatedly argued, all these “non-naive direct” or “non-indirect” realisms are just describing indirect realism but refusing to call it what it is.

Both fall prey to the fallacy of ambiguity; there is some ambiguity with the verb "see", for example. In the case of hallucination there is no object of perception. If there was, it wouldn't be a hallucination. So we're confusing the object of perception with perception itself. — NOS4A2

Schizophrenics hear voices. I feel pain.

As I’ve said before, arguing over the grammar of “I see X” is a confusion, precisely because as you say the term is ambiguous.

The only relevant concern is the epistemological problem of perception. Conscious experience - percepts - do not extend beyond the body. Distal objects do not exist within conscious experience. Our knowledge of distal objects is indirect, inferred from the effects they have on our body (specifically from conscious experience and its qualities).

Any “intentionality” in experience is akin to the intentionality found in paintings and in books. This isn’t the kind of “directness” that indirect realists reject, that naive realists accept, and that would save us from scepticism. See Howard Robinson’s Semantic Direct Realism.

It is true that organisms perceive. It is untrue that brains do. — NOS4A2

If you define "perception" as the body responding to external stimulation then this is a truism, but this isn't at all relevant to the debate between direct and indirect realism.

We see things when we dream and hear things when we hallucinate. The things we see and hear when we dream and hallucinate are percepts – phenomena either reducible to or supervenient on brain activity – and these percepts exist when awake and not hallucinating, having been caused by the body responding to some appropriate external stimulation. Visual percepts are what the sighted have and the cortical blind lack, and this can be shown by comparing the activity in the primary visual cortex of the sighted and the cortical blind.

These percepts are the intermediary from which we infer the existence and nature of some external stimulus and/or some distal object (e.g. where the stimulus is light), given that conscious experience does not extend beyond the brain, let alone the body, and so these stimuli and distal objects do not exist within conscious experience.

This is what the science shows and this is quite clearly indirect realism.

Conscious experience, perception, or whatever other activity is impossible if one or the other is missing or deceased or uncoupled. That’s a brute fact we ought to consider, in my opinion. — NOS4A2

That's not at all relevant.

If my computer isn't plugged into the wall socket then it won't even start, but the operating system is to be found in the SSD, not in the wall socket.

That A depends on B isn't that B contains A.

The only thing a disembodied brain can do is rot. So brains do not think or experience or perceive. Only bodies do. And the body is, conveniently, the only thing standing between your perceiver and other objects in the world. — NOS4A2

Bodies are required to keep brains alive and functioning, but conscious experience is to be found in the brain activity. When there's no (higher) brain activity there's no conscious experience, e.g. those in a coma or in non-REM sleep, even if the bodies respond to stimulation.

The attempt to dismiss the rest of the body in the act of perception is clearly motivated by something other than scientific inquiry, and it would be interesting to find out what that motivation is. — NOS4A2

No, it's just what the science shows.

When we put a brain on a table it’s impossible to say the brain feels pain or experiences — NOS4A2

We can say anything we like.

therefor it is just untrue to say brains feel pain and experience. — NOS4A2

That's a non sequitur.

---

Of relevance is the anatomy of pain and suffering in the brain and its clinical implications:

A stimulus produces an effect on the different sensory receptors, which is being transmitted to the sensory cortex, inducing sensation (De Ridder et al., 2011). Further processing of this sensory stimulation by other brain networks such as the default mode, salience network and frontoparietal control network generates an internal representation of the outer and inner world called a percept (De Ridder et al., 2011). Perception can thus be defined as the act of interpreting and organizing a sensory stimulus to produce a meaningful experience of the world and of oneself (De Ridder et al., 2011).

...

Pain is processed by three separable but interacting networks, each encoding a different pain characteristic. The lateral pathway, with as main hub the somatosensory cortex is responsible predominantly for painfulness. The medial pathway, with as main hubs the rdACC and insula are involved in the suffering component, and the descending pain inhibitory pathway is possibly related to the percentage of the time that the pain is present.

As I see it the science is very clearly in support of indirect realism (e.g. the first paragraph of the quote above). Any armchair philosophy that tries to defend direct realism is either contradicted by the science or has redefined the meaning of "direct perception" into meaninglessness and so is not in conflict with the actual substance of indirect realism.

Repeating yourself doesn't answer my question. You're just throwing in the word "directly" without any meaning or justification.

perceptible properties of distal objects are directly observed — Janus

Which means what? What does it mean to be directly observed? Given that conscious experience exists within the brain and given that the properties of distal objects exist outside the brain, the properties of distal objects do not exist within conscious experience.

Which means that everything that exists within conscious experience is the intermediary of which we have direct knowledge and from which the physically distant properties of distal objects are inferred.

What could direct perceptual knowledge be but reliable knowledge of its objects, as opposed to (presumably) indirect (because subject to intermediate distortions) unreliable perceptual appearances? And I'm talking about the vast amount of observational data in botany, zoology, geology, chemistry and so on, not about inferred, unobservable entities and events like electrons and the Big Bang. — Janus

Perception of distal objects is inference. Light from the sun travels to Earth, reflects off some object's surface, stimulates the sense receptors in the eyes, triggering activity in the brain, giving rise to conscious experience (which is either reducible to or supervenient on this brain activity). There are a multitude of mental/neurological/physical processes that exist and occur between the conscious experience and the distal object.

Distal objects and their properties are inferred from the effect (conscious experience).

I may have been overly simplistic in my account but the point stands: I feel pain, pain is not a distal object/property but a mental/neurological phenomena, and so the thing that I feel is not a distal object/property but a mental/neurological phenomena. The same for smells and tastes and colours.

You can argue that this mental/neurological phenomena involves a variety of different mental/neurological processes rather than just some simple sui generis qualia, but it still admits that it is some mental/neurological phenomena that is experienced rather than some distal object/property.

I don’t doubt the brain is involved, but clearly the toe is as well. I’m just wondering the biology of “experience”, for instance how far from the brain it extends. — NOS4A2

The toe is the trigger. It's where the sense receptors are. But the sense receptors are not the pain. Pain occurs when the appropriate areas of the brain are active.

The heater grate to my right is not a mental representation. It is a distal object. It's made of metal. It has a certain shape. It consists of approximately 360 rectangle shaped spaces between 48 structural members. The spacing is equally distributed left to right as well as top to bottom. However, the left to right spacing is not the same as the top to bottom.

The 'mental representation', whatever that may refer to, cannot be anywhere beyond the body.

According to you, all we have direct access to and thus direct knowledge about is mental representations.

Where is the heater grate? — creativesoul

We have direct perceptual knowledge of our body's response to stimulation. We have indirect perceptual knowledge of the distal objects that play a causal role in that stimulation.

The grammar of "I experience X" is appropriate for both direct (I feel pain) and indirect (I feel the fire) perception.

The idea that we have scientific knowledge relies on the assumption that we have reliable knowledge of distal onjects. Attempting to use purportedly reliable scientific knowledge to support a claim that we have no reliable knowledge of distal objects is a performative contradiction. — Janus

I didn't say that we don't have reliable knowledge, only that we don't have direct perceptual knowledge. Even the direct realist must admit that many of the things we know about in science, e.g. electrons and the Big Bang, are not things that we have direct perceptual knowledge of. Is it a performative contradiction for a direct realist to use a Geiger counter?

Alternatively, we can argue like this:

If direct realism is true then scientific realism is true, and if scientific realism is true then direct realism is false. Therefore, direct realism is false.

The direct realist would have to argue that direct realism does not entail scientific realism (and reject scientific realism) or that scientific realism does not entail indirect realism.

Then wouldn't experience be limited to the prefrontal cortex — NOS4A2

From a common neurobiology for pain and pleasure:

If I stub my toe, injure my toe, and feel the pain in my toe, is it your position that I am feeling it in my prefrontal cortex? — NOS4A2

I think that there are pain receptors in the toe, that these send signals to the brain, and then there is pain when the relevant areas in the brain are active. The brain is clever and able to make it seem as if the pain is literally in the foot, but that cleverness also leaves us susceptible to phantom limb syndrome.

Is the "I" that feels pain the organism or the organism's cognition? — NOS4A2

The cognition. I believe it’s to be found in the prefrontal cortex, whereas pain is in the somatosensory cortex.

Dreams are not perceptions, and "hearing voices" is an abnormal case of perception. — Luke

It is nonetheless the case that I see and hear things when I dream and hallucinate and that the things I see and hear when I dream and hallucinate are mental phenomena. The Common Kind Claim says that waking veridical experiences are of the same kind as dreams and hallucinations (e.g. the activity of the sensory cortexes) – differing only in their cause – and so that the things I see (e.g. colours) and hear when having a waking veridical experience are also mental phenomena.

Your picture suggests otherwise. — Luke

Perhaps this will make it clearer:



The same principle holds for smelling and tasting and hearing and seeing.

The picture maintains what I consider to be the false assumption of indirect realism: that we require a second-order cognition/awareness/perception in order to perceive the first-order perceptions. In other words, cognition/awareness/perception of perceptions, which seems to imply an infinite regress. Perceptions (i.e. first-order perceptions) are here treated as not something already present to consciousness, or as if they were themselves external objects. — Luke

Do I see colours when I dream? Does the schizophrenic hear voices when hallucinating? I say "yes" to both.

This is where you're getting confused by grammar into thinking that indirect realists are saying something they're not.

What is useful is knowing that the apple is either ripe or rotten and the color of the apple informs us which is the case. — Harry Hindu

Okay. We have direct knowledge of colours, which are a mental phenomenon. Given that we have inferential – i.e. indirect – knowledge of the apple's ripeness, which is a mind-independent property.

Our perception of the apple's mind-independent property is indirect.

What is the "I" that is made indirectly aware via mental phenomenon? How is it separate from the colours, mental phenomenon and other objects to say that the mental phenomenon is an "intermediary through which I am made indirectly aware..." — Harry Hindu

They're different aspects of consciousness, resulting from different areas of brain activity. The blind man has a self but doesn't experience visual phenomena because his visual cortex doesn't function.

I take it that the position of indirect realism is that perception never provides us with direct knowledge of distal objects. And the position of naive realism is that perception always provides us with direct knowledge of distal objects? — Luke

Take the picture here. If indirect realism is true then if we remove the mental image then we have no knowledge of the distal object (or, to be more precise, any knowledge of the distal object has been gained by some means other than perception). And I believe that's correct. The mental image is the necessary intermediary.

Another picture that may prove helpful, with the lines representing some relevant causal connection.

I don't follow. In what sense is your knowledge indirect here? Is the wavelength of the light a property of the distal object? — Luke

I know that I see the colour red.
I know that in most humans seeing the colour red usually occurs when the eyes react to light with a wavelength of 700nm.
I infer from this that I am looking at an object that reflects light with a wavelength of 700nm.

Of course this is only true because I am somewhat educated in science. For many, e.g. young children, all they know is that they see the colour red. They don't know anything about electromagnetism and so don't know anything about the distal object's mind-independent properties.

For example, we see colours. As per the Standard Model, colours are not a property of distal objects. Distal objects are just a collection of wave-particles. Colours are a mental phenomenon often caused by the body responding to particular wavelengths of light. I have direct knowledge of the colour red and indirect knowledge of a distal object reflecting light with a wavelength of 700nm.

I don't particularly like my own formulation of (Shrimp) btw, as it bifurcates seeing as a perceptual act and classification as a linguistic one, whereas there's evidence that the two are reciprocally related - both predictively/inferentially/causally and phenomenologically (citation needed). — fdrake

Are you suggesting that deaf and illiterate mutes don't see colours (or see everything to be the same colour)?

You could end up with a statement like:

(Shrimp) Mantis Shrimp Human sees X as P(X) and calls it "P(X)" if and only if human sees X as Q(X) and calls it "Q(X)".

Predicating of the distal object X now makes sense because we've reintroduced the idea that properties of distal objects influence the kinds they are seen and labelled as.

Do you think you need a numerical identity between the state of being that Mantis Shrimp Human has when they count X as P(X) and the human's that counts X as Q(X) even when P and Q have the same extension? — fdrake

I'm not entirely sure what you're trying to ask here.

My argument is that:

1. There is some stimulus X
2. There is some organism A and some different organism B
3. Given their different physiologies, organisms A and B have different experiences when stimulated by stimulus X
4. Some of the words that organisms A and B use to presumptively describe X in fact describe some aspect of their individual experience (and that is not an aspect of the other organism's experience).
5. Colour words are one such example.

The mantis shrimp example is a nice way of illustrating the flexibility and potential for expansion in our color concepts, while still maintaining a realist commitment to colors as objective properties of objects. — Pierre-Normand

This is equivocation. There is "colour" as an object's surface disposition to reflect a certain wavelength of light and there is "colour" as the mental phenomenon that differs between those with 3 channel colour vision and those with 12 channel colour vision (and that occurs when we dream and hallucinate).

Despite sharing the same label these are distinct things – albeit causally covariant given causal determinism.

Those with 3 channel colour vision and those with 12 channel colour vision will agree that some object reflects light with a wavelength of 700nm, but they will see it to have a different colour appearance.

They'd still ascribe the colors within this richer color space to the external objects that they see. — Pierre-Normand

If they're direct realists, and they'd be mistaken. Naive colour realism is disproven by our scientific understanding of perception and the world.

It would be interesting to hear what a human with his eyes replaced with those of a mantis shrimp (with their 12 channel colour vision compared to our 3) would say.

Furthermore, people who disagree about the interpretations of the picture can communicate their disagreement by pointing at external paint color samples that are unambiguously blue, black, gold and white to communicate how it is that the pictured dress appears to be colored to them. Here again, their agreement on the color of the samples ought to give you pause. — Pierre-Normand

It doesn't give me pause. Given that our eyes and brains are mostly similar, and given causal determinism, it stands to reason that the same kind of stimulus will mostly cause the same kind of effect.

But it is still the case that the cause is not the effect and that colour terms like "red" and "blue" can be used to refer to both the cause and the effect, and so you need to take care not to conflate the two, but it seems that direct realists do conflate.

I'd now like to discuss an issue with you. Consider the definition expressed in the sentence: "For an apple to be red means that it has the dispositional property to visually appear red under normal lighting conditions to a standard perceiver." Might not a subjectivist like Michael complain that this is consistent with an indirect realist account that views redness as the (internal) subjective states that "red" apples are indirectly or inferentially believed to cause (but not seen to have)? Or else, Michael might also complain that the proposed definition/analysis is circular and amounts to saying that what makes red apples red is that they look red. Although, to be sure, our "in normal conditions" clause does some important work. I did borrow some ideas from Gareth Evans and David Wiggins to deal with this issue but I'd like to hear your thoughts first. — Pierre-Normand

I think you're overcomplicating it, being "bewitched by language" as Wittgenstein would put it.

An apple reflects light with a wavelength of 700nm. When our eyes respond to light with a wavelength of 700nm we see a particular colour. We name this colour "red". We then describe an object that reflects light with a wavelength of 700nm as "being red".

The indirect realist recognises that the colour I see in response to my eyes responding to a particular wavelength of light (and the colour I see when I dream and hallucinate) is distinct from an object's surface layer of atoms and its disposition to reflect a particular wavelength of light. The indirect realist recognises that this colour I see is a mental phenomenon and that this colour is the intermediary through which I am made indirectly aware of an object with a surface layer of atoms with a disposition to reflect light with a wavelength of 700nm (assuming that this is a "veridical" experience and not a dream or hallucination).

Perhaps this is clearer if we consider something like "the fire is painful" rather than "the apple is red".

This appears to be equivocation. We use the term "colour" to refer to both the disposition to reflect a certain wavelength of light and to the mental phenomenon that is caused by our eyes reacting to a particular wavelength of light, but these are two different things.

This is evidenced by the fact that we can make sense of different people seeing a different coloured dress when looking at this photo:



When I say that I see a white and gold dress and you say that you see a black and blue dress, the words "white", "gold", "black", and "blue" are not referring to some spectral reflectance property (which is the same for the both of us) but to some property of our mental phenomena (which is different for the both of us). The same principle holds for the colours we see when we dream and hallucinate.

Indirect realism accepts the existence of these mental colours and claims that they are the "intermediary" or "representation" of which we have direct knowledge and through which we have indirect knowledge of a distal object's spectral reflectance properties.

Whereas direct realism would entail the naive realist theory of colour.

As I see it, your account simply redefines the meaning of "direct perception", which I think is best understood as explained here and here.

See also Semantic Direct Realism where Robinson explains that the same kind of redefinition occurs for other so-called "direct" realisms like intentionalism.

The relevant issue is whether we have direct perceptions of real objects, not direct knowledge of perceptions. — Luke

The epistemological problem of perception concerns whether or not perception provides us with direct knowledge of distal objects.

One group claimed that it does, because perception is "direct". These people were called direct realists.

One group claimed that it doesn't, because perception is "indirect". These people were called indirect realists.

Therefore the meaning of "direct perception" is such that if perception is direct then perception provides us with direct knowledge of distal objects. Therefore if perception does not provide us with direct knowledge of distal objects then perception is not direct.

Given our scientific understanding of the world and perception it is clear that perception does not provide us with direct knowledge of distal objects. Therefore perception is not direct.

He may have refined his position since we began this discussion but he had long taken the stance that what I was focussing on as the content of perceptual experience wasn't how things really look but rather was inferred from raw appearances that, according to him, corresponded more closely to the stimulation of the sense organs. — Pierre-Normand

I am saying that appearances are mental phenomena, often caused by the stimulation of some sense organ (dreams and hallucinations are the notable exceptions), and that given causal determinism, the stimulation of a different kind of sense organ will cause a different kind of mental phenomenon/appearance.

The naïve view that projects these appearances onto some distal object (e.g. the naïve realist theory of colour), such that they have a "real look" is a confusion, much like any claim that distal objects have a "real feel" would be a confusion. There just is how things look to me and how things feel to you given our individual physiology.

It seems that many accept this at least in the case of smell and taste but treat sight as special, perhaps because visual phenomena are more complex than other mental phenomena and because depth is a quality in visual phenomena, creating the illusion of conscious experience extending beyond the body. But there's no reason to believe that photoreception is special, hence why I question the distinction between so-called "primary" qualities like visual geometry and so-called "secondary" qualities like smells and tastes (and colours).

Although even if I were to grant that some aspect of mental phenomena resembles some aspect of distal objects, it is nonetheless the case that it is only mental phenomena of which we have direct knowledge in perception, with any knowledge of distal objects being inferential, i.e. indirect, entailing the epistemological problem of perception and the viability of scepticism.

I don’t see how his account differs from indirect realism. Indirect realists simply claim that the thing we have direct knowledge of in perception is some sort of mental phenomenon, not some distal object, and so our knowledge of distal objects is indirect, entailing the epistemological problem of perception.

We can argue over what sort of mental phenomenon is the direct source of knowledge - sense data or qualia or representation or appearance or processed phenomenological content or other - but it all amounts to indirect realism in the end.

How do you conceptualise a distal object in the second construal of indirect realism? — fdrake

Depends on the indirect realist.

Some may believe that distal objects resemble our mental image, and so would replace the question mark with the Earth as shown in the person’s head.

Some may believe that distal objects resemble our mental image only with respect to so-called "primary" qualities, and so would replace the question mark with an uncoloured version of the Earth as shown in the person’s head.

Some may believe that distal objects do not resemble our mental image at all. A scientific realist would replace the question mark with the wave-particles of the Standard Model. A Kantian wouldn’t replace the question mark at all, simply using it to signify unknowable noumena.

What defines them as being indirect realists is in believing that we have direct knowledge only of a mental image. Direct realists believe that we have direct knowledge of the distal object because nothing like a mental image exists (the bottom drawing of direct realism).

I think the issue is that people misleadingly think of this as being the distinction between direct and indirect realism:



When in fact it is this:

On this view, phenomenology is concerned with describing and analyzing the appearances of those objects themselves, not the appearances of some internal "representations" of them (which would make them, strangely enough, appearances of appearances). — Pierre-Normand

The indirect realist doesn’t claim that there are “appearances of appearances”.

The indirect realist claims that a distal object's appearance is the intermediate representation.

We have direct knowledge of a distal object's appearance and through that indirect knowledge of a distal object.

You're describing indirect realism but calling it direct realism for some reason.

The direct realist rejects any distinction between a distal object's appearance and the distal object itself, entailing such things as the naive realist theory of colour.