No, that's a big help. Sort of what I thought. Rödl calls this kind of statement "a thorn in the side" of propositional logic, in part because of the linguistic ambiguities you note. And these in turn lead to conceptual puzzles, such as the one about whether to conceive of thinking as an experience or an action. Also, perhaps even more crucially, whether "a thought" is a World 2 or a World 3 item. Frege would of course not have that terminology to hand, but I think a lot of what he had to say about psychologism is an attempt to sort it out.

Yes, I remember that article, I'll reread, thanks.

I may not have asked what I meant to ask here. I was wanting to affirm that all 4 statements are well-formed propositions -- that is, they are candidates for having all the things done to them that we do with propositions, including asserting, questioning, etc. In that sense, "I see the oak tree" may not be a judgment, but can't it be asserted?

Or, leaving Frege out of it, I'm trying to home in on whether 1st-person statements that quote a proposition internally, as it were (such as C and D), are in any way logically problematic.

Still beavering away at this, not ready for prime-time yet. But you can help with a clarification:

A. "The oak tree is shedding its leaves."
B. "I see the oak tree."
C. "I think* 'The oak tree is shedding its leaves'."
D. "I think** 'The oak tree is shedding its leaves'."

*Popper's World 2 sense of "think" = a mental event
**World 3 sense of "think" = results in Frege's "proposition"

Am I right that all four of these sentences are propositions in good standing, according to Frege? The fact that A is from no particular point of view, whereas B - D are, doesn't matter, correct? Also, the fact that C & D quote another proposition is likewise kosher?

Oh, no worries, these are difficult topics. FWIW, my question about representation was strictly about its Kantian use, since it seems that, if Rödl fudges anywhere, it's there. You say:

representation is absolutely necessary for any and all Kantian speculative metaphysics, — Mww

And that's my understanding too, but until now I would have said that substituting "thought" for "representation" (again, within Kant-world) isn't a major misunderstanding. Could you say more about what hinges on this for him? Many thanks.

Thanks, that's it exactly.

I'm not sure if Kant had that in mind, but I'm pretty sure that's what Rödl means. Sorry to be a broken record, but once again it's the ambiguity of "think" that is critical. Rödl wants to take "think" and "thought" in the sense of "a mental event occurring to an actual thinker or subject." And not only does he not take it in the sense of "propositional content," he's trying to construct an argument to the effect that this sense isn't meaningful.

Kant, on my best reading, is asking for a structural interpretation of the term "accompany". For him, it isn't a matter of the self-consciousness of the thinker, a la Rödl. Rather, I read him as saying that for me to have a thought at all, there must be something he calls the I think that makes the thought possible for me -- the same way space and time structure perceptual experience. Once again, we're talking "thought" as mental event, not propositional content.

I mean, the dude himself said, “Kant said, more precisely…” at the expense of his own statement’s accuracy.

But he’s got letters after his name and I don’t, so….there ya go. — Mww

Heck, I don't care about anybody's letters -- if it's wrong, it's wrong. I think you know Kant very well; would you say that Rödl's qualification here ("more precisely") does affect the accuracy of the statement "The I think accompanies all my thoughts"? And I'm still struggling to understand what's at stake in contrasting "representation" with "thought" in this context. Any help with that?

Kant's failure to draw and maintain the distinction between thought and thinking about thought. — creativesoul

Truly, I wasn't aware there was a problem here. Can you say more about Kant's failure? I know you feel you've said what needs to be said already, but I'd be grateful for a little more explication.

Yes, I read Sartre that way too.

"The oak tree is shedding its leaves" is a valid proposition but not a thought. "Think the oak tree is shedding it leaves" is not a valid proposition, as it doesn't indicate who is having the thought. "I think the oak tree is shedding its leaves", "they think the oak tree is shedding its leaves" and "he thinks the oak tree is shedding its leaves" are valid propositions expressing thoughts.

A thought cannot be had without someone having that thought. — RussellA

Yes, these reflections are in the spirit of the problem Rödl is considering. You're right to connect it with the ambiguity of "thought" as "event of thinking," which requires a thinker, and "thought" as "proposition" (this is how Frege used it), which is supposed to be available to us objectively, without needing to indicate a thinker.

Anyway….not that big a deal. — Mww

I'm sure it isn't, but I hate to get anything wrong. Let me fill out what Rödl says. His footnote for the claim "Kant said: The I think accompanies all my thoughts" reads: "CPR, B 131. More precisely, he [Kant] says that the I think must be able to accompany all my representations, for all my representations must be capable of being thought. This presupposes (what is the starting point of Kant's philosophy and not the kind of thing for which he would undertake to give an argument) that the I think accompanies all my thoughts."

Hmm. Does this respond to your point? I'm not completely clear what's at stake with the distinction between "thought" and "representation" here, but Rödl does seem to be interpreting, rather than innocently paraphrasing. Granted, the difference is blurry when it comes to reading Kant.

The semantics of “uncertainty” and “doubt” being an utterly different issue to that of the thread, granted, but I do find interest in it. — javra

Me too, thanks for clarifying. On your interpretation, "uncertainty" is definitely the better term for what I was trying to get across.

Several folks (@banno, @creativesoul, & others, sorry if I've forgotten) have responded to the OP by affirming that Pat is “right” in her report of her experience. I want to look at that more closely, starting with what exactly it is that she is right about (hence the scare-quotes).

Basically, Pat is telling us that she hasn’t had the experience that she believes she would have to have if it were true that the “I think” accompanies all our thoughts. She believes that experience would be something like an additional thought that says, “I think p,” and that occurs simultaneously or in close proximity to her thought of p. She tells us that this does not characterize her usual experience of thinking.

So, is Pat right about her experience of thinking? There’s certainly no reason to doubt it. Full disclosure: It was quite easy to write Pat’s lines for her because I pretty much share that experience. So I think we ought to say that Pat is right about this.

But I don’t think that’s what some people mean when they say that Pat is right. I think they mean that, because Pat has disconfirmed a particular version of what the “I think” would entail, therefore she is right that the “I think” does not accompany all our thoughts. That was response #4, back in the OP. But note that Pat never actually says this. She starts with a particular interpretation of what the “I think” would be, and (rightly, we’re saying) reports that she hasn’t experienced it. But unless she’s unusually dogmatic (or possibly unversed in philosophy), she wouldn’t go on to say that no other understandings of the “I think” are possible.

Here’s where I think this leaves us. We can accept that Pat’s (putatively accurate*) report rules out the possibility that the ubiquity of the “I think” consists in its being some kind of affirmative, conscious thought that accompanies every one of our mental representations – or even our propositions. But that leaves quite a bit still to explore. Response #4 states, “If your report is accurate, then the thesis that ‛the “I think” accompanies all our thoughts’ has been proven wrong.” But we see that isn’t so. What has been proven wrong is the notion that the “I think” is a subject of experience.

Now let’s compare this to the Kantian perspective. Imagine that a different experimental subject I’ve created :wink: tells us, “Kant claims that time and space are constitutive concepts of the understanding, and form the basis of any possible experience. Well, sorry, but when I experience something, I don’t also have an experience of ‛time’ and an experience of ‛space’. I just experience whatever it is that happens.”

We could reply, “Quite right, time and space are not themselves experiences, they are constitutive of experience. They are the without-which-nothing. When Kant says that they are ubiquitous throughout all possibilities of experience, he doesn’t mean you can discover them as some additional ur-experience.”

I hope the parallel with the “I think” question is clear. Referring back to my original four responses, it looks like #3, which argues basically what I just wrote, is the one we should choose if we want to explain to someone who believes that Pat’s experience justifies #4, why that isn't so.

But I don’t see the issue as settled yet. Are there good reasons for claiming that this transcendental “I think” has any reality at all? We’re not there yet. Maybe, if it isn’t an experience, it’s just a hoax. But I think we’ve made some progress by showing the alleged role of experience in all this a little more clearly.


*And response #2 is available as well, if we want to try to make a case that Pat is mistaken, or misguided, about her experience.

the stipulation that “I think” as a proposition always accompanies the proposition “I think (proposition) p” is, for my part, utterly absurd: — javra

Agreed. I resisted including a specifically propositional understanding of the "I think" as one of my suggested retorts to Pat because it seems like a non-starter. I don't believe anyone, from Descartes on, ever meant that.

A more plausible option is that the "I think" is in fact a thought of some kind, even if not a proposition. We're seeing some good reasons on this thread to question even that, though.

To answer “I did” and “I think I did” to some question is in no way and at no time equivalent: the first expresses a fact one is confident about regarding what one did, this while the second expresses something along the lines of a best presumption based on one’s best reasoning (i.e., thinking) regarding what one in fact did (presumably about a past deed one does not hold a clear recollection of). The second does not however require doubt of what one thinks is the case, but only allows for certain degrees of uncertainty. — javra

And notice what happens when we ask whether the doubt being expressed is about the thought or what the thought is about. I can be absolutely certain that, right this minute, I am having the thought "I think I did" concerning some previous action I'm not too sure about. Again, the ambiguity of "thought" as mental event (yep, definitely happening) and "thought" as that thought's intensional content (not too sure).

I see what you're getting at but I meant something a little different. It hinges on the ambiguity of the word "thought". We commonly use the word to mean two distinct things: a mental event occurring at a particular place and time, and the content or import of said event ("proposition," in Fregean terms). Traditionally, the latter is common to any particular instantiations of the "thought" in the first sense.

So we could be talking about the mental event "thought that Paris is crowded" or the proposition "Paris is crowded". We can assert things about a mental event that we couldn't assert about a proposition, including something like, "I had Thought X come into my mind but I don't understand the proposition it states." We couldn't say (and mean it), "I'm asserting X (the proposition) but I don't understand it."

Just as a preliminary, I'll lay out Rödl's very stark position:

What is thought cannot be isolated from the act of thinking it; it cannot be understood as the attachment of a force to a content. This may seem hard to accept. — Rödl

Indeed! He goes on to list the benefits that are generally agreed to accrue if we think in terms of force and content:

[The claim is that] We need to distinguish force from content in order to describe disagreement among different subjects. . . . We need to distinguish force from content if we are to represent the progress one makes from asking a question to answering it. . . . Further, the distinction is needed if we are to understand inferences that involve hypothetical judgments. . . . Thus it [would have] great explanatory power. Giving it up is costly. — Rödl

But:

As the force-content distinction makes no sense, it has no explanatory power. There is no cost to abandoning it. — Rödl

So if this is true, Rödl has got to show not merely that the force/content distinction is an incorrect analysis of how propositions work. He also has to make the case that the distinction is literally nonsensical, that there is a deep basic confusion on Frege's part in the way he divides up the conceptual territory involved here.

Why in the world would Rödl think this? He believes that Fregean logic can't make sense of self-conscious thought -- that we need a clearer way to describe what is actually happening when a person thinks.

There can be no Fregean account of first-person thought, no account that provides it with a Fregean thought as its object. — Rödl

And by "Fregean thought," he of course means a proposition. A real cliff-hanger, more to follow . . . I'll let you cool down ! :smile:

I agree that when you say, "Paris is crowded," you're stating your thought (belief, opinion, judgment) that Paris is crowded. We can make up unusual circumstances when that wouldn't be true, but let's grant that this is generally what's going on.

However, when you say "I think Paris is crowded," you can be saying either of two things. You can be saying, "I find myself thinking the thought, 'Paris is crowded'. Hmm, wonder if that's true." (In which case we'd be more likely to punctuate your statement as "I think, 'Paris is crowded'."). Or you can be saying, "I do in fact think (believe) Paris is crowded." The first instance foregrounds the thought, the fact of thinking; the second focuses on the content of the thought. Both are pretty common usages, I would say. We often report a thought qua thought, as an interesting mental event.

And see my attempt at a similar disambiguation, above.

I think you're saying that the use of "I think" here is my sense 1, not sense 2. If I already believed or asserted it, why ask? Right, that would be a good example of why the force/content distinction seems necessary. I'm still working my way toward articulating some of the possible Rodelian responses here, but I definitely appreciate the problem.

I agree that we cannot think without the I think at the very least subtly implied or lurking in the shadows of thought, but I do not think that reflects the ultimate reality. — ENOAH

Sure. I believe someone who agreed that the "I think" is ubiquitous could go either way on whether there was more to reality than this relation -- whether the "I think" constitutes reality or only reflects or discovers it in some way.

Yeah, the reflexivity gets confusing. How about if we do this:

p = "The oak tree is shedding its leaves."
q = "I think, 'The oak tree is shedding its leaves' = "I think p"

So Pat is saying, "I think p but I am not aware of thinking q."

Now, you're wanting to add a new relation -- a new reflexivity:

r = "I think 'I think p'" = "I think q"

That's the move I don't understand. Can you say why this next level of reflexivity is needed to make the situation clear? When does r arise for Pat?

Also, I'm not sure the first-person is all that important to the distinction being drawn. We talk about other people's mental events, just as we talk about other people's affirmations and claims and all that. "Judy thought you had gone home." "Judy thinks you should go home. — Srap Tasmaner

Right. The disambiguation between the senses of "I think" applies equally well to "you think" and "she thinks."

I guess the biggest question is how you intend to handle the mental events side. Space of reasons or space of causes?

"Judy thought you had left because she heard the front door" as causal: "If Judy had not heard the front door, she wouldn't have thought you had left"; or as not: "If Judy had not heard the front door, she would have had no reason to think you'd left." ― The trouble with the second is that it should really have "and so she didn't" at the end, but it's pretty hard to justify. People think all kinds of stuff, or fail to.

Does any of that matter for the theory? — Srap Tasmaner

Good question. My allegiance, apart from this particular theory, is usually with reasons rather than causes. I haven't worked it out here yet at all. Put a pin in it!

So, the 'cleavage' is not so much 'oppositional' in nature so much as comparative. — creativesoul

Good. It only becomes a cleavage if we find that some philosopher, in putting forward the theory that the "I think" is ubiquitous, is depending on one or the other of these construals of "think."

So, #4 is 'right' in some way/sense of being right.

Pat is right to deny that that is always the case. However, some of the other answers are also correct, depending upon the specific candidate of thought under consideration. — creativesoul

Say more about that? Can you give an example of a thought-candidate that would make one of the other answers also correct?

However, this whole thread just glosses over the underlying issue. Kant did not draw the distinction between thought/thinking and thinking about thought/thinking. Rödi just assumes and further reinforces that error. — creativesoul

No, it's the opposite. Here's what I wrote in the OP, with relevant passages bolded:

The “I think” accompanies all our thoughts, says Kant. Sebastian Rödl, in Self-Consciousness and Objectivity, agrees with this but points out that “this cannot be put by saying that, in every act of thinking, two things are thought: p and I think p.” He calls this a confusion arising from our notation, and suggests, not entirely seriously, that we could devise a more accurate notation “that makes I think internal to p: we may form the letter p by writing, in the shape of a p, the words I think.” He interprets Kant as saying the same thing: for Kant, “the I think is not something thought alongside the thought that it accompanies, but internal to what is thought as such.” — J

Both Rödl and Kant agree with you that these do not represent two thoughts.

'p', 'I think p', and "I think 'p'" all presuppose truth. — creativesoul

See my attempt to disambiguate these. To report that I have a thought p is not to claim truth for it, at least according to the "standard model."

There's a lot here. Let me start with this (I italicized p when it stands unquoted, for clarity):

Why isn’t the p/“p” dualism backwards? Objectivity is the thing given to sensibility, whatever it is, it is that thing, so should be denominated as p. What I think about is nothing more than the affect that thing has on my senses, the affect cannot possibly be identical to the (p) thing itself, so can justifiably be denominated “p”, which in turn is referred to as representation of p. Shouldn’t it be the case that objectivity is p, subjectivity being how I am affected by p, which would be thought by me, post hoc ergo propter hoc, as “p”. — Mww

Are you suggesting we call "the thing given to sensibility" -- that is, the object we encounter -- p? So p doesn't stand for a proposition any more, on this usage, but names an object? And then "p" would be the representation (or thought) of that object p? If I've got this right, it seems reasonable enough except that traditionally p is used to refer to a proposition, not an object. Also, if "p" is any sort of representation, don't we still need a 3rd term to use for actual propositions? When you say, "What I think about is nothing more than the effect that thing has on my senses," you close the door on the idea of propositional content, it seems to me. Which may be what you intended, but it's an unusual construal, unless you're limiting the discussion to objects of perception.

So with these recent posts, we’re going a bit deeper into the question of “I think p” and its relation to p.

Let’s start with an important disambiguation, which is going to affect everything from Rödl’s “self-consciousness” to the standard Fregean model of force and content.

“I think p” can be understood and used in two distinct ways:

1. “I think ‛p’.”
2. “I think that p is the case.”

Filling it in:

1. “I think, ‛You’ve left your book in my car’.”
2. “I think that you’ve left your book in my car.”

1 is about an act or occurrence of thinking. It’s a report about something that has happened “in my mind.” It carries no commitment to the truth or affirmation of the content of the thought in question. Similar constructions would be: “I had the thought that Cindy was tired,” “Right now I’m thinking, ‛The dog is on the log’.” This version of “I think” foregrounds the thought, the act of thinking. The person addressed isn’t being asked to agree with whether Cindy was tired, or the location of the dog. Rather, an appropriate response would be something like, “Oh, so that’s what you’re thinking.”

2 is an affirmation in the Fregean sense. We can go so far as to say that, taken in this sense, it’s synonymous with “I judge p” or “I affirm p”. By reporting my thought that “Cindy was tired,” I mean to additionally report that I believed this thought to be true. Similar constructions would be: “Q. Did Washington cross the Delaware? A. “I think so.”; “I think you lost this”; “I think that I see a wren.” We can note that, in ordinary usage, “I think so” can be a somewhat diluted form of affirmation, but it’s an affirmation nonetheless; it expresses the speaker’s agreement with p.

Now we can ask, when philosophers contrast “p” with “I think p”, which usage do they mean, 1 or 2? This will vary case by case. But for idealists/monists like Kimhi and Rödl, I believe that they mean to dissolve the distinction between 1 and 2. They want to say that the very act of thinking always affirms something (though neither likes to use that language). And this is a major reason why they question Frege: There is no such thing, for them, as a thought of p that is only a report of a mental event which in turn contains a “propositional content.”

This sounds outrageous and wrong. But before I go any further, I invite comment on the above. Is it reasonably clear?

"I think" necessitates a self that is conscious of thinking. — RussellA

This is one of Rödl's key points.

Possibility one = p is external to the self, internal to the self but not a part of the self or accompanies the self. If this were the case, the self would have no way of knowing about p. — RussellA

I don't follow this. Can you say more? Why couldn't the self have knowledge of something external to it?

I wasn't as clear as I should have been, thanks for your patience.

the issue is contingent on what one interprets the term “thought” to signify. — javra

I tend to agree, based on the interesting responses to the OP. The key cleavage seems to be whether thought is meant to be essentially sentential or propositional, as opposed to "representational". As usual with philosophical terms, there's no dictionary we can consult about this, and usage differs, so we have to make our best choices for clarity as we go along.

you asked about a statement made from a thesis concerning pure speculative reason, which couldn’t be anything other than metaphysical — Mww

Again, I'm not sure this is right. Is the thesis "The 'I think' accompanies all our thoughts"? I was trying to include, in my possible replies to Pat, the possibility that this is meant as a report about experience, not a metaphysical position. We may not find that very promising, but we should at least be able to say why. In other words, let's be completely clear that the thesis of the ubiquity of the "I think" is metaphysical, that no one's experience can discredit it.

Something here bothers you, but it remains unclear what. So I'm taking this thread as your articulating what it is you find troublesome. If you can see some error or lack of clarity in what I've said, it might help. — Banno

Good way of putting it. This is a quintessentially philosophical experience: something is bothering me, for all the evident clarity, and I'm trying to put my finger on it. Maybe I can construct some Rodelian replies to your laying out of the "standard model." Something to do during the massive snowstorm we're about to have . . .

As a thesis, can it be falsified by experience?
— J

As a thesis, speculative metaphysics can’t be falsified at all, without altering the parameters upon which it rests. ‘Course, neither can it be proved from experience. — Mww

OK, but we don't want to beg the question that it is speculative metaphysics. On one construal, we're supposed to be able to actually experience the "I think" as an accompaniment or component of our thoughts. In that case, I would say that's an empirical question that could be falsified. Especially if the construal claims that we must experience it.

Someone recently told me about Noesis and Noema — Patterner

I know these terms from Husserl. I'll read it and get back to you, thanks.

Pat is correct. I know this isn't what you're after, but... — creativesoul

Sure it is, or could be. If I thought this had a cut-and-dried answer, I wouldn't be bothering y'all with it. All opinions are welcome. So, same question to you as to @Banno, earlier: If Pat is correct, does that mean that my #4 is the right response?

Really appreciate everyone's participation here.

I don't get from the discussion where this "I think p" resides. — T Clark

Right, I'd say that was the very question up for discussion, or one of them, at any rate. I was trying to lay out some possibilities about the "I think" -- is it meant to be an experience? a thought in addition to whatever I'm thinking about p? an unexperienced structure? another way of naming or describing self-consciousness? etc. And one of the problems of conceiving the "I think" as a new thought is precisely the one of infinite regress.

See @fdrake's post.

there are thinking things that do not have a sense of self. — Philosophim

A cat is thinking about the leaves falling off the tree as it playfully leaps up to attack them as they're falling. But I do not believe a cat is capable of thinking about thinking about the leaves falling off the tree. That's a different level of thought, of which cats are not capable. — Patterner

some animals are definitely aware of themselves but don't have language. — fdrake

This is a question that I doubt would even have occurred to Kant, given his era's primitive understanding of animals. I'm not sure what we should say about non-human "thinking things." Certainly it could represent a limit case about self-consciousness. For now, I want to resist the temptation to divvy up "think" in terms of whether only humans can do it. If we need that discrimination later on, we can double back to it.

Even if I can't think the higher level thought without the lower level though, I can think the lower level thought without the higher. — Patterner

Yes, if "I think p" is indeed meant to be present to consciousness at all times that "p" is thought.

there's a bunch of sentential content bubbling up from/in the mind, some surveyor partitions it into A-OK and "dump it" - the latter of which is discarded somehow. The A-OK stuff gets labelled/willed as "I think", associated with the selfhood/subjectivity of that person, and that stuff can get asserted by that person. Call that account A.

Alternatively the "I think" is what takes mental/bodily gubbins and puts it in sentence form and filters it into A-OK and "dump it". Then the remainder of the first account holds of the sentence forms. Call that account B. — fdrake

Good, that's clear. These are different accounts of when thought becomes propositional, if I understand you. And yes, the respective qualia would be different.

"I think" has an implicit sentence placeholder — fdrake

This is one of the points that has come up quite quickly in this discussion. Is that what Kant meant? When Rödl considers "I think p", does he understand the thinking to begin with the sentential formation of p? How plausible is that?

I, personally, am just not aware of a cloud of sentences associated with environmental objects and my own thoughts. The majority of my meta-awareness is perceptual rather than sentential, and the parts of it which are linguistic are more broadly narrative than declarative. — fdrake

That's my experience as well, especially if you add in "memories". This could point us toward deciding that whole "'I think' is ubiquitous" thing is misguided. Or, we could accept B as the best account of when the "I think" emerges.

the expression of an idea rather than a particular sentence, — fdrake

Another way of forming the same question. Some philosophers will object that ideas don't occur separated from sentences. I think they can.

If I had no thoughts, "I" would not exist. "I" could not exist if I had no thoughts. — RussellA

Do the quotes around "I" mean that there is literally no self without thoughts, or only that the "I" of philosophy, so to speak -- the self-conscious cogito -- is constructed from our thoughts? And same issue here, of course: "thoughts" as sentential, or more broadly as images, perceptions, etc.?

the simple and in itself perfectly contentless representation “I” which cannot even be called a conception, but merely a consciousness which accompanies all conceptions. By this “I,” or “He,” or “It,” who or which thinks, nothing more is represented than a transcendental subject of thought = x, which is cognized only by means of the thoughts that are its predicates, and of which, apart from these, we cannot form the least conception…. — Mww

I bolded Kant's phrase, above, because it focuses on what we'd like to understand better -- just what the heck does it mean for consciousness to "accompany" something? Would we know it when it happened? As a thesis, can it be falsified by experience?

Meaning it's a regress, and therefore untenable?

...a new way of conceiving something...
— J
New?

It's just (illocutionary force(predicate(subject)), what is done with the proposition, or Frege's judgement stroke. — Banno

Well, the new way, if Kimhi and Rödl are on target, would deny the force/content distinction, as we know from that earlier thread about Thinking and Being. Another way of putting it: There is no way of stating p without stating p. Someone has to be doing the stating. Even reserving any judgment of p doesn't get you off the hook from having thought it, in some sense, to begin with.

In what sense, is the question. Are all propositions first-person propositions? Does it make a difference that a proposition can only occur as someone's thought? According to Kimhi, that makes a big difference in what "propositional content" means. Rödl looks to be going about the argument in a different way to get to the same place, but more of that later. For right now, let's just say that this possible "new way" is either genuinely new and useful, or a somewhat perverse misunderstanding about how logical language works. I am not sure, yet, which view is correct. Like you, like all of us, I'm super-trained in the "old way" -- which is enough to make me wonder if there's more to it.

Kant says “I think” must accompany all my representations (B133, in three separate translations), not my thoughts. Some representations are not thought but merely products of sensibility, re: phenomena. — Mww

This is a good clarification. Do I think a representation? German-to-English may be an issue here. Kant uses the "I think" to structure all mental representations; Rödl probably means only propositions -- I say "probably" because I'm still in the middle of reading the book so there may be further discussion about that. (Note: Although Rödl is German, he appears to have written S-C & O in English. Anyone know if that's true?)

I vote '1'. — Wayfarer

Yes, #1 has its attractions, but notice that, in its entirety, it commits us to the belief that the "I think" can be experienced. This may be hair-splitting, but the kind of self-awareness you're describing sounds less like an experience and more like an understanding, an enlightenment about what thought is. That would be closer to the spirit of #3, as I intended it. I wanted #3 to be the thoroughgoing transcendentalist position.

To be sure, there are no thoughts that could not be prefixed by "I think..."; but that is a very different point to the suggestion that all our thoughts are already prefixed by "I think...". — Banno

Right. And for Rödl (and I think Kant and Sartre) it isn't even a matter of "prefixed"; the "I think" is supposed to be structural or internal. That's why the question of whether this "I think" is experienced comes to the fore.

That there is a difference between a proposition and one's attitude towards that proposition - thinking it, believing it, asserting it, doubting it - is so ingrained that I have difficulty making sense of the alternatives. — Banno

Me too. It's almost like a phase shift, a new way of conceiving something that had always seemed obvious. Both Kimhi and Rödl are asking us to rethink what we thought we knew. The issue is difficult enough (and technical enough) to preclude just waving it away, though I wanted to when I started reading Kimhi.

What is the relationship of "p" and "I judge that [think/believe/propose etc.] p"? That's the most bare-bones way of posing the question.

Good start! Re Davidson, I'm not sure. But since I'm more or less thinking about three phil topics at once these days, I wouldn't be surprised

So, if Pat is right, #4 is a good response?