I'm not onboard with the James quote, for two reasons. First, what counts as a simple is down to context, and here I'm thinking of the later Wittgenstein: and second, I'm not certain of the implied physiology - that we build our sensorium up from patches strikes me as overly simplistic. Do you see the red patch and the bands and build Jupiter from them, or do you see Jupiter and then by being more attentive divide off the patch and the bands? Or some combination? These are questions for physiology, not philosophy.

I'm not sure where this leaves us.

Good post. While what you say here is quite valid, our practices override such considerations. Folk might quite successfully agree to "meet at the barn".

I had the same thought regarding Gettier, and supposed there were some connection. Austin was probably aware of Russell's stuck clock, an early Gettier problem. Doubtless Gettier had read Austin.

OK.

...representation... — frank

I guess that's right. @Isaac and I had some lengthy chats about what "representation" consists in, in a neural network. What we did agree on is that in so far as there are such representations, it is clear that they are not symbolic, but found in the weightings of various connections.

V continued
The argument from illusion has two parts. In the first, already addressed, it is argued that in certain abnormal situations we must admit that what we see is not a material thing, but a sense datum. We've already seen how this is wrong. In the second part this false contention is supposed to be extended to all cases, such that all we ever see is sense data.

Ayer claims that there is "no intrinsic difference in kind between those of our perceptions that are veridical in their presentation of material things and those that are delusive" (p.44). The objections follow.
1. A bogus Dichotomy. There's a certain question-begging in the use of "perceptions", which are already in the place of sense data, and so presume them from the very start.
2. An exaggerated frequency. The implicit, unjustified presumption is that perceptions are all either veridical or delusive.
3. An exaggerated similarity. It is just not true that veridical and delusive perceptions are 'qualitatively indistinguishable". We do understand the difference between dreaming and being awake; seeing stars at night and seeing stars after a concussion; seeing an after-image and seeing a colour patch; seeing a blue wall and a white wall through blue glasses; seeing pink rats and suffering dementia tremens. "In all these cases we may say the same things ('It looks blue', 'It looks bent', &c.), but this is no reason at all for denying the obvious fact that the 'experiences' are different"(p.49-50)
4. An erroneous suggestion. Why should we expect, when we see two different things, that they should appear different? Why shouldn't two different things appear much the same?
5. If one fails to make a distinction, it does not follow that there is no distinction to be made. If we fail to make a distinction between delusive and veridical perceptions, it does not follow that there is no such distinction to be made. We are fallible.
6. What, exactly, a "perception" consists in remains obscure. This allows the goals to be moved - "Inevitably, if you rule out the respects in which A and B differ, you may expect to be left with respects in which they are alike" (p.54).

:up:

I had thought you had seen what Austin shows: that "direct" gets its use from "indirect". It seems that needs reinforcing.

If asked how does smelling works, I would refer to the standard scientific account - I'm doing philosophy, so I don't know anything those scientists don't also know. But those accounts do not talk of direct and indirect smelling, except when they adopt a philosophical stance.

The confusion, if there is any, stems from the fact that we often fail to disambiguate perception-words like "see" — J

Yes, and then there is the broader methodological point that this failing leads to broad philosophical theories - such as Ayer's logical positivism - built on misunderstandings of language.

And this is to my eye what is absent from, say, Ernest Gellner's supposed rebuttal. When i read his book, I don't see him addressing the Austin or the Wittgenstein I understand. It strikes me as an instance of Banno's rule: It's easier to critique someone if you begin by misunderstanding them.

Yeah, I think there is something in this, especially the idea of philosophical problems being "manufactured" by philosophers. Cynically, they have to do that in order to justify their stipend.

But it may be that we would be among the duped, in which case we would see a barn. — Fooloso4

Yes, in which case, as I said, we are mistaken. What we see is a church, made to look like a barn.

Do you see something deeper here, that I'm missing?

Cool. In each example you give, you are able to set to clearly the indirect case that allows us to make sense of the direct case.

The point is about what it is that we see. — Fooloso4

Sure. IS there a presumption that there is only one correct answer here? Those in on the joke see a church. The duped see a barn. The explanation is that the church has been made to look like a barn. I don't see a problem.

What would be the point of camouflaging it if not to fool those who do not know that it is a church? — Fooloso4

Exactly.

What is the basis for the distinction between what something looks like and what we see? — Fooloso4

The duped think they see a barn. They are mistaken. What they see is a church, made to look like a barn. I don't see a problem.

It seems as though Austin is basing the distinction on a questionable assumption about objectivity, as if we don't see a barn because it is a church. — Fooloso4

But we don't see a barn, we see a church that looks like a barn. How does dressing that up in terms of objectivity change that? Did camouflaging the church transform it into a barn? I don't think so. It just made it look like a barn.

That depends on whether one is aware that it has been camouflaged, of course.

I'm not seeing(!) a point here, either in favour or against the arguments we are considering.

I don’t think there’s a right or wrong way to talk about this, necessarily, but I do think a defender of the value of ordinary language is going pretty far out on a limb here. — J

"I see a barn" is not wrong. One might usefully say we will meet at the barn, and be understood by those who know it to be a church. Isn't the point that "I see a barn" is not the whole story?

But "I see a barn and I see a church" is problematic, if they are the very same. One feels entitled to ask, "So, which is it? Church or barn?". One presumes that it is not both. Despite this, a church might be used as a barn, or a barn as a church.

Again, detail and context are needed. "I see a church that looks like a barn" gives us more than "I see a barn".

We should try to avoid the interminable discussions that so often proceed from such differences. I take it that we agree there is a church, and that it looks like a barn, and that "I see a church" is OK, and so is "I see a barn", but that their conjunction needs some additional information - the fact of the camouflage - to avoid contradiction.

We need to get past the picture of a process called "perception". If nothing else, Austin has shown that this is a figment that is simply manufactured by philosophy. — Antony Nickles

Yes, but going on past experience on the fora, it won't happen. :wink:

I myself, cannot quite understand how perception works directly, but I do understand how it works indirectly, i.e. via sense organs and sense-data. — Corvus

These are not easy issues to work through. One thing that might help is remembering that sight is not the only sense, and that an account of how we perceive must wok as well for touch and smell as for vision.

So are you sure you understand how it works to touch something indirectly? To smell the coffee, indirectly?

I certainly don't.

What you are seeing is a memory of the cup... — Corvus

There's a homunculus lurking here.

In the case of the camouflaged church what we see is not, as Austin claims, "a church that now looks like a barn". (30) What we see is a barn. If we didn't what would be the point of camouflaging it? — Fooloso4

Should, or even could, we bring out every nuance of this story? What is the building over there? It is a church. It has been made to look like a barn. So we see a church that has been made to look like a barn, A church that now looks like a barn. That seems reasonable. I think this account is clear.

We might prefer to avoid saying "What we see is a barn and what we see is a church." And if pressed, I'd have to agree with Austin, that what we see is a church, albeit one that looks like a barn.

It's classic. — Ludwig V

That's right. Austin was a classicist. He was drawn into philosophy by puzzlement at the things philosophers said. He brought his method over from Classics.

So, again, he is not saying we experience the world directly or indirectly--he is throwing out the entire picture of us (here) and the world (there) that leads to that distinction. — Antony Nickles

I hearty agree! While we are at it, let's also throw out that other bugaboo (should that be buggerboo?) subjective/objective, the notion of things having to be either "internal" or "external".

V
Roughly, Ayer's argument is that

• When we see something, there is always a thing that we see.
• There are instances where what we see is a different thing to what is "really" there; a thing philosophers call "sense data"
• This account must be generalised, so that in all instances, what we see is sense data.

So far we have watched Austin carefully dismantle the first two steps. The first in Lecture II, the second in Lecture III and IV. Now we are moving on the finishing step.

Before looking at Austin, let's consider Zhuang Zhou. You will no doubt be familiar with the story. As a butterfly, he did not know he was Zhuang Zhou. When he was Zhuang Zhou, he wondered if he was a butterfly.

It's a stimulating story, throwing one's considerations off-centre, and I do not wish to detract from it, but to add to it, since I think it can give us some insight into the approach Austin takes in Lecture V. We do know the difference between dreaming and being awake. We understand the nature of dreams, that they occur during sleep, usually at night, and may involve various otherwise impossible things. We understand what it is to dream and what it is to be awake - we must do, because we have the language around dreaming. If we could really not tell our dreams from our more lucid states, we could have no such language. We could not even have the word "dream".

We know also that the story is told from the point of view of Zhuang Zhou, and not from the point of view of the butterfly. If we did have the story from the perspective of the butterfly, the world would be a very different place. But the symmetry on which the story depends must be broken in order for the story to be told.

Considerations such as these have a close parallel in the final writings of Wittgenstein on certainty. The story can only take place if the very things it brings into doubt are held firm. And the story, being constructed of words, has to take it's place in a community of human beings.

Here again is the self-deception that is needed to get a good story to come alive - as mentioned.

What it is and what we see are not the same. — Fooloso4

Well, sometimes what we see is what there is...

I'm not seeing a difficulty for Austin, here - is there one? I had rather taken him as showing that seeing, touching, smelling and so on were much broader than Ayer's account supposed, in much the way you do here.

sensory data — frank

Sensory data, sure. Our senses provide data. But "sense data" seems a term peculiar to philosophy, with the mentioned peculiarities.

I gather we are in broad agreement, then.

I've looked for uses of "sense data" outside of philosophical contexts, but found precious little. I checked the main online physiology and medical resources, but found nothing; certainly nothing showing its use in these fields.

There's a kind of self-deception at work. — Ciceronianus

And it will come to the fore with Lecture V. There will be folk who are so enamoured with the delusion of sense data that they will reject the argument, which is curiously phenomenological in character. In a good way.

What I am getting at is that there is more to perception than passive reception. What we see when we see the cup is not something separate from or independent from what we call it and what we use it for. — Fooloso4

Sure. There's much more detail that might be included, if it were deemed relevant.

How do you think this impacts on Austin or Ayer's arguments? Otherwise, this would for me be veering off towards Quinn and Davidson; Radical translation, Radical interpretation, Triangulation and so on. Interesting stuff, and happy to follow up on it, if it is germane.

But otherwise, you and I might agree that name calling - "linguistic idealism" - doesn't help. There's more to a cup than the word "cup". Some of the comments here are utterly off the path.

Presumably they see the cup - a less racist alternative to the old myth that they could not see anything at all. We might find that despite seeing the cup they have no word for the cup and so no knowledge of how to use it.

Prima facie, Lecture IV required the most effort for the least gain.

But 's question shows some of where these ideas might be taken, more so by Wittgenstein than by Austin. One wonders what direction Austin might have taken had he lived longer.

Austin's point here is that "direct" and "indirect" are a pair, linked by their opposition. Each derives it's meaning from the other, like "north" and "south", "up" and "down", "hot" and "cold". If you say that all perceptions are indirect, and imply that no perception is, or could be, direct, you deprive "direct" of any "meaning" and hence render "indirect" meaningless as well.

I don't accept that my eye is an intermediary, getting in the way of my perception — Ludwig V

Yes!

I appreciate the facetious style - this is what such silly alternatives deserve.

This is a new concept to me. — Ludwig V

Yep. It would have been novel for Austin, too. Thank you for saving me from addressing this incongruity.

An excellent post.

I was wanting to keep interacting from my own thoughts only on the topic — Corvus

I don't wish to dissuade you, indeed there is no alternative, as you must begin where your thoughts are now. The material we are considering takes some digestion, especially as much of it is contrary to what is usually taken as granted in these fora. But from what you have written here you have been following Austin's account well, which is far more than can be said for others.

Again, the notion of a direct perception makes sense when we know what it is being contrasted with. So we understand the difference between seeing the ship indirectly through the periscope as oppose to popping the hatch and taking a look from the conning tower; we understand the difference between seeing the tree indirectly through the binoculars as opposed to walking over to it to see it directly. But the contention offered in the forums is much odder than these cases. It's that we never see the ship or the tree or anything else directly, but only through the intervention of our eyes. And here it is not at all clear what it would mean to see something without using one's eyes, or any other sense organ. So it's not clear what the direct/indirect distinction is doing in this case. Austin doesn't directly address such an argument, because no one, least of all Ayer, was so gormless as to present it.

It was addresses by David Stove, an Australian philosopher, who used the example of tasting oysters. Should oysters be eaten straight, or with a squeeze of lemon? Or the whole Kilpatrick treatment? Well, if you would know what oysters themselves taste like, it might be best to try them "natural". But the argument for indirect realism is as if someone were to suggest that one never tastes oysters except with one's tongue, and therefore one never tastes oysters in themselves.

I hope the absurdity is plain, and that you see the relevance of 's joke.

I am still trying to understand the direct realist's account on perception.  In what aspect perception is to be understood as direct and real? — Corvus

I also hope that it will become clear that neither Austin nor I are making the claim that our perceptions are in some way always direct. Sometimes - periscopes and binoculars and mirrors - they are indirect, and in such situations we can understand what it would mean in contrast for them to be both direct and indirect.

So in those terms, there is nothing to understand. A so-called "direct realist" account of perception is the same as the standard account given by science.

Are they saying that what they sense and perceive from the external world are the true existence of the beings and the world with no possibility of being uncertain or inaccurate? — Corvus

No. But they might say that when you look at a cup, what you are seeing is the cup, and not some philosophical innovation such as sense data or qualia. That you are not a homunculus sitting inside a head, looking at the a screen projecting images of cups.

The reply to this will be that we understand from recent scientific developments that our brains actively construct a model of the cup. That's quite right. But it would be an error to think that what we see is this model - the homunculus again. Rather, constructing the model is our seeing the cup.

IV
A slightly shorter, but intense, lecture breaking apart various uses of "Look", "Appear" and "Seem". talks about the complexity of the issue, which comes down to Ayer apparently playing on words again - the straight stick looks bent; but what are we to conclude from that? We can agree that it looks bent while maintaining that it is in fact straight; but Ayer would have us deny this, or at the least call it into question. Ayer wants us to take "something looks bent" and conclude that, therefore, something is bent; it's only by our being duped in this way that we will again be convinced of the existence of sense data.

Further, as repeats, "descriptions of looks are neither 'incorrigible' nor 'subjective'... There is certainly nothing in principle final, conclusive, irrefutable about anyone's statement that so-and-so looks such-and-such"(p.42). How things look will not carry the weight Ayer would place on it.

It doesn't matter. — Ciceronianus

Very true. Of course not.

On re-reading, a few things stand out to me. Foremost is how often Austin explicitly pushes against many of the sorts of things of which he is now accused. Those accusations target a caricature, not the man. Next is the explicit misogyny and racism, which is perhaps more than just an indication of the times. And he's not as cruel as he is in my recollection, although the attack is searing, castigating each and every step in Ayer's book one after the other.

It's interesting to watch and attempting to fit the actual Austin in to the account that is so prevalent here, that indirect realism is about sensory apparatus, the way in which our eyes and brain process vision, and so direct realism must also be about sensory apparatus. Corvus in particular is finding that what Austin actually says does not match the common account of what an indirect realist should say. The hard part for them is going to be addressing the arguments Austin actually presents, and not re-dressing them so that they fit a preconfigured critique.

(Austin) is not defending realism against antirealism, but rejecting the very distinction between these two. — Banno

This applies also to direct/indirect realism. The danger for this thread is that the discussion becomes just another rendition of that tedious "he said/she said".

My problem is that I can't imagine what direct perception would be. — Ludwig V

The point Austin makes quite early seems to me to cover this:

I. First of all, it is essential to realize that here the notion of perceiving indirectly wears the trousers- 'directly' takes whatever sense it has from the contrast with its opposite — p.15

You didn't see it directly, you saw it through a telescope, or a mirror, or only its shadow; how we are to understand "direct" perception depends entirely on what it is contrasted with; so of course it is difficult to imagine what "direct perception" is, per se. It's a nonsense, an invention of the defenders of the sort of argument Ayer is presenting. You can find examples in every thread on perception*.

Austin is specifically tearing down philosophy's framing of the issue as both direct or indirect. — Antony Nickles

Yes!

Must admit Austin's writing style is super clear, and utterly logical. — Corvus

Yep, Ciceronianus' Damascan cream pie in the face. A good philosophical account is compelling.

Do we agree that “qualia” refer to actual phenomenological experience... — J

One of the prejudices I share with Austin is a dislike for specifically philosophical innovations. Talk of qualia mostly post dates Austin, but I suspect he would have spent some time pointing out that the term doesn't seem to achieve anything not already found in our ordinary talk of seeing and touching.

Going over my own notes, I found an admission that I did not understand qualia - from 2012. In 2013, I said I do not think that there is worth in giving a name to the subjective experience of a colour or a smell. In 2014, I doubted the usefulness of differentiating a smell from the experience-of-that-smell. Never understood qualia. I still don't see their purpose. — Banno

...that's from three years ago. I've had no reason to reconsider.

* See the ongoing discussion in https://thephilosophyforum.com/discussion/14738/a-case-for-transcendental-idealism by way of example. It was in part bemusement at the dreadful standard of the discussion there that inspired my re-reading of Austin.

I'm afraid I'm a bit confused about whether we are working through the sections systematically or just reading the book at our own pace? — Ludwig V

For my part my intent is to continue in the way I have been, reading a lecture or two ahead and then going back to re-read in more detail to make notes mostly for myself. Lecture IV will probably be very brief, then a bit more detail, or less, as we move into the later lectures, if I loose interest. If you want to move at a faster pace, go ahead, but I've found in the past that this leads to folk getting lost and needing to go over arguments again.

In a PM to @Richard B I said

I would prefer the thread stay on the topic and not become another diatribe against linguistic philosophy. It's probably inevitable that it become so mired, but I'll not help out. Much.

As for your specific question, I don't see Gellner's "four pillars" at all in Austin; indeed, Austin's method is antagonistic to all four. — Banno

I personally think that it is invalid because it is simplistic — javi2541997

Yes, it is simplistic. But what makes it invalid is that the conclusion does not follow from the premise.

We sometimes see things incorrectly... — javi2541997

What is not ruled out here is the possibility that we sometimes see things incorrectly and at other times we see them correct. That is, the premise does not ruled out that we sometimes see things correctly.

...therefore, we never see them correctly — javi2541997

This just does not follow.

But do we see ourselves in the mirrors because this is what they do - reflecting - or because do we actually exist? — javi2541997

That's a good question, with a long, but not so difficult, answer. I don't know your philosophical background. So I will go back a few steps. What follows is a potted history, and as such it is roughly correct in broad outline, but definitely wrong on the detail.

The approach to philosophical questions being used here is ordinary language philosophy, a part of analytic philosophy. It doesn't begin with doubting everything, or trying to find certainty, or looking for indubitable sources of knowledge, Instead it starts pretty well where we are, here and now. And it proceeds by looking with great care at the philosopher's main tools, their words.

So it doesn't here begin with what exists and what doesn't, or what "self" is, what it is to be this thing and not that thing, questions of individuation and essence, or other profound questions. But it could.

Some folk presume that because it doesn't start with such things, it doesn't address them That'd be quite wrong.

Here it has begun with an argument that in its day was quite influential, a cornerstone of what was called Logical Positivism. Logical positivism was a branch of analytic philosophy that tried to build certainty from observations. The argument presented by Austin is a large part of the demolition of that project.

So on one view, all Austin is doing is critiquing another analytic philosopher. However, in the process he is developing and using a range of philosophical tools, with much broader application than just the question at hand; and also providing at least in outline a coherent account of how we deal with perceptions by setting out in fine detail the many nuanced ways we use language in this area.

He is developing and showing us a different way to deal with philosophical issues.

So you asked, roughly, how our existence fits in to the account given here. The answer, roughly, is that it is taken as granted; not because it is assumed to be certain, but because it is not central to our concerns here. To be sure, if it became an issue, then we could consider it, again using the method of examining the way we use words around notions of existence and so on. There are such considerations elsewhere, mainly in areas relating to logic, quantification and equivalence. And the answer, roughly, might be not "...if I am capable of seeing myself in the mirror, it is thanks to the mirror itself and not me" so much as "... if I am capable of seeing myself in the mirror, it is thanks to there being both a mirror and me".

III concluding.
The sting, when it comes, is pithy and simple.

So the argument goes that the stick is straight, but appears bent.

What is wrong, what is even faintly surprising, in the idea of a stick's being straight but looking bent sometimes? Does anyone suppose that if something is straight, then it jolly well has to look straight at all times and in all circumstances? — p.29

And the answer here is simply that there is no problem with a straight stick that looks bent when partially immersed in water, not illusion, no delusion, nothing that needs explaining beyond the physics of optics mentioned earlier. And there certainly is no need to infer the existence of a novel entity to take on the part of being what we see when we look at a straight stick that appears bent when partially submerged, appart form and distinct from the stick.

And when you see yourself in a mirror, there is no need to invent a simulacrum to stand in for you. There is no illusion, no hallucination and no error. What you see is yourself, reflected in the mirror. Again, this is what mirrors do, and no further explanation is needed that replaces your reflection with anything immaterial.

And when you see a mirage, you need not infer the presence of a new thing, but understand instead that the light from things already there has been bent in odd ways. Of all the examples, this is the one Ayer deals with most poorly.

for though, as Ayer says above, 'it is convenient to give a name, to what he is experiencing, the fact is that it already has a name-a mirage. — p.32

So there is no reason, let alone a necessity, to conclude that in these cases we need invoke sense data to explain what is going one.

Even less, then, to take the next step, and conclude with Ayer that in normal cases, what we see is sense data.

Austin sounds as if he believes there’s only one correct way to see something.. — J

Is that what he addresses in a somewhat racist fashion on p.26? I don't think he's saying that there is only one correct way, but that there is at least one correct way, that does not involve sense data. And that's all he need show in order to undermine Ayer's contention that we are obligated to invoke sense data. Ayer's argument is based on there being no alternative. Austin simply need show one alternative.

Yes, Austin is displaying the varieties of perceptual experience, and drawing conclusions therefrom. Contra , I supose.

III continue...

So is a mirage an illusion or a delusion? Ayer would have us suppose that we have conjured into existence a thing, a "sense data", and hence that we are deluded, seeing the sense data where there is nothing. But it is clear from the explanations given above that what we see is a part of the sky, or a ship, appearing in an unexpected place. The problem, if there is one, is not within us, but in the way light is bent by layers of air at different temperatures. We don't need the delusion of sense data to explain mirages.

When you look at yourself in the mirror, are you looking at an illusion? That is an odd use of words. Those are your eyes, that zit is on your nose; they are not illusory, and certainly not delusions. The coin, seen edge on, looks elliptical - that's not an illusion, but just how a coin seen edge-on is supposed to look. The straight stick in water looks just how a straight stick in water should look - slightly bent. Just as in the Müller-Lyer Illusion we understand what is going on, our expectations are met.

Importantly, there are common cases where we miss-perceive, yet are neither illusions nor delusions. Austin gives the example of a proofreader who sees "casual" as "causal"; not a delusion or illusion, but a simple misreading.

Austin uses Price to show a common feature of Arguments from Illusion; that they suppose there is something extra that is what is seen in each of these cases, in addition to the mirage, reflection, stick and so on; the argument is taken as proving that this additional aspect of perception is needed. I hope it is plain from the explanations given hereabouts that this need not be so. The explanations of each do not require sense data. In such situations we are not deluded.

Sometimes there is reason to doubt what you're experiencing... — frank

Yep. I supose that what is salient here is that sometimes folk doubt what they are experiencing without good reason. Austin is slowly and carefully showing why this is problematic.

We sometimes see things incorrectly; therefore, we never see them correctly. — javi2541997

The invalidity of this is apparently not obvious to many. Stove's gem, the worst argument in the world, and so on.

If everything you perceive is real... — Corvus

Austin is certainly not making any such claim. Sometimes we see things that are real. It does not follow that everything we see is real. Sometimes we see things that are not real. It does not follow that everything we see is not real. So your "For the realists, there is no room to say anything more on the perception than a chair is chair" is a mischaracterisation. Nor is memory a simple process of storage. I suggest the brush you are using here is too broad. If for you "the realist's account on perception sounds too simple", you might consider that you have not represented their view accurately.

, , "hallucination" is closer to "delusion" than to "illusion", in that something is conjured up in both an hallucination and a delusion, but not so much in an illusion. Should Ayer have called his the "Argument from Hallucination"? That doesn't carry the same rhetorical strength.

“seeing” a table is to identify something as a table — Antony Nickles

:up: Slowly...

I’ll leave him out of it; — Antony Nickles

To be clear, there's no need to leave Wittgenstein out, indeed there is much to be gained in keeping him in, but we need to take care, given the considerable overlap, as to who is claiming what.

:up:

It would be reasonable to introduce a term like sense-data as a place-holder for whatever it is we decide we perceive directly. — Ludwig V

The danger here is the presumption that what we perceive is all of one sort, in such a way that we can apply the label "sense data" in all cases. Austin is showing that this is not a good idea.

III continued...

Ayer's supposition is that in each case what we see is not a "real" material thing; there is no bent stick, no body of water, no person standing in the mirror, and hence we "must" conclude that in each case what we see is something else - what philosophers call a "sense datum".

Some might now replace "sense datum" with "qualia". Not quite the same.

Austin points out that although we are supposedly looking at the Argument from Illusion, that there is a distinction between an illusion and a delusion, terms interchanged by Ayer, and that the examples given are not examples of illusions. "The argument trades on confusions at just this point" (p.22)

Here's the familiar Müller-Lyer Illusion:

The two lines appear different lengths, but are the same.

Amusingly, the page from which I borrowed the image has this to say:

Here’s the thing: Even after we have measured the lines and found them to be equal and have had the neurological basis of the illusion explained to us, our conscious awareness still perceives one line to be shorter than the other. One can know that the two lines are the same length whilst at the same time experience them as different lengths.This has a serious effect on our conception of the nature of experience.

The world around you is not the way you think it is. — Zach Olsen

Olsen's conclusion is not quite right, but not quite wrong, either. We are, after all, aware of the Müller-Lyer Illusion, and after a point no longer surprised by it - one might say it becomes a part of the way we think. The Müller-Lyer Illusion is part of the world around you and part of what you think. Olsen's conclusion, "the world around you is not the way you think it is", is imprecise, as are similar views expressed already in this thread.

Delusions differ from illusions mainly in their being something amiss with the person who is deluded - they have delusions of grandeur, of persecution. We suffer from delusions, not so much from illusions... (p.23) In an illusion, nothing need be "conjured up"; we see the lines before us quite clearly. In a delusion, the grandeur or paranoia are the product of the sufferer. Something is wrong with the person who has delusions. But there is nothing wrong with you when you look at the Müller-Lyer Illusion; indeed, now you are aware of the illusion, you can take steps to prevent yourself being deceived, and even use your understanding to create novel illusions.

Yes, I was somewhat concerned not to present Wittgenstein's view. Here:

I look at a chair a few yards in front of me in broad daylight, my view is that I have (only) as much certainty as I need and can get that there is a chair and that I see it. But in fact the plain man would regard doubt in such a case, not as far-fetched or over- refined or somehow unpractical, but as plain nonsense; he would say, quite correctly, 'Well, if that's not seeing a real chair then I don't know what is.' — S&S p.10

Moore might have said "If this is not a real hand then I don't know what is." So either this is a real hand, and we are good, or we have no idea what a real hand might be.

III

The primary purpose of the argument from illusion is to induce people to accept
'sense-data' as the proper and correct answer to the question what they perceive on certain abnormal, exceptional occasions; but in fact it is usually followed up with another bit of argument intended to establish that they always perceive sense-data. Well, what is the argument? — p.20

Here he outlines the argument from illusion, hinting at the shenanigans he sees in its structure. Austin will show how Ayer has oversimplified, even misdiagnosed, the case for these abnormal instances, why we should reject 'sense-data' as a solution, and then that generalising to all perceptions is absurd.

Austin's dissection is minute and thorough.

The three examples on the next page are central: A stick that appears bent in water because of refraction; A mirage; and a reflection in a mirror. I'll give a bit of background on each.

The physics of the stick is straight forward.

A stick or a pencil half immersed in water at an angle appears bent due to refraction of light at the air-water surface. Figure shows a straight stick AO whose lower portion BO is immersed in water. It appears to be bent at point B in the direction BI. A ray of light OC coming from the lower end O passes from water into air at C and gets refracted away from the normal in the direction CX.Another ray OD gets refracted in the direction DY. The two refracted ray CX and DY, when produced backward, appear to meet at point I, nearer to the water surface than O. Similarly each part of the immersed portion of the stick raised. As a result immersed portion of the stick appears to be bent when viewed at an angle from outside. — Discourse

The mirage example is a bit unclear.

Thus, when a man sees a mirage in the desert, he is not thereby perceiving any material thing , for the oasis which he thinks he IS perceiving does not exist — Ayer, p. 4

Presumably the "thing" that is supposed not to exist is the body of water...

...in which the tree appears to be reflected, not the tree, which does exist. This image shows a possible result:

The ship, of course, exists. Ayer's example is not the best.

Mirror images are familiar, and well-understood.

Of course, the stuff in the mirror is "the wrong way around" - usually.

There's no mystery here, all accepted physics.

good luck — Antony Nickles

Cheers. I'm not claiming any expertise here, just an interest and an enjoyment of his style, even recognising its many flaws.

I already missed a few things, and may go back and edit what I've already said.

Anyway.