I've been rereading your OP, and I think I get the argument now. — Srap Tasmaner

Yes! You’ve got it precisely, and have expressed it better than I did. Many thanks. The key difference is affirming a claim – that is, a statement -- rather than making your own statement about how the world is.

This seems slightly at odds with the descriptions involving a repeated identical 'p': there are no repeated complete symbols here. — Srap Tasmaner

I’m puzzled too. Can a more expert logician weigh in and help?

I think the general thrust of the whole modern Frege-Tarski-model-theoretic approach is to presuppose the existence of the objects within the universe of discourse — Srap Tasmaner

Sure. But what I was trying to point out (or what I think Roberts means, anyway) is that “the universe of discourse” isn’t neutral or discoverable or God-given or whatever. We have to determine it, which requires quantification. What we presuppose, I think, is that once we do this, there’s no problem with saying that objects exist.

That said, I know some Frege and Tarski but would flunk a test on "the whole modern Frege-Tarski-model-theoretic approach," so feel free to school me.

[In Frege] the duality between assertoric force and predicate may well be equally expressed as the duality between sense and reference. — RussellA

Good, that fits the Fregean picture.

The unity of thinking and being is the cornerstone of Wittgenstein's Tractatus. — RussellA

Yes, to the best of my understanding (with help from Kimhi).

In the Tractatus, it is not the case that a proposition has a sense prior to anything that is being referred to, in that sense may be disassociated from reference, in that the sense of "this grass is green" may be disassociated from its referent in reality, that this grass is green. But rather, the sense of the proposition is what is being referred to, in that there is a unity between the sense of "this grass is green" and its referent, that this grass is green. — RussellA

Cf. this from Notes on Logic:

When we say A judges that, etc., then we have to mention a whole proposition which A judges. It will not do to mention only its constituents, or its constituents and form but not in the proper order. This shows that a proposition itself must occur in the statement to the effect that it is judged. — LW

And cf. the elucidation @Srap Tasmaner provided between affirming a claim and making a judgment about the world.

- Great posts.

(1) Assertion is (a person, an agent) claiming that the possible state of affairs, let's say, described by a statement does in fact hold.
(2) Assertion is (a person, an agent) affirming the claim about the world made by a statement. — Srap Tasmaner

I am struggling to see the difference here, but maybe that is just me. I was understanding @J to be saying that propositions can have assertoric force independent of persons/agents who would speak them. Therefore I would prefer a distinction between a possessor of assertoric force which requires a speaker/asserter and one that does not. I thought J was saying, "This thing has assertoric force even before you pick it up and assert it."

This seems slightly at odds with the descriptions involving a repeated identical 'p': there are no repeated complete symbols here. — Srap Tasmaner

Interesting point. Substitution of individuals/particulars occurs beginning with (2), and substitution is a different kind of logical move.

But I would say the middle term (the recurrence of 'p') is found in the substitutability itself. (1) quantifies over all integers, and because '5' and '2' are integers they are substitutable into the formula of (1). Does the OP's point about the repeatability of p break down in cases of substitution? I shouldn't think so, but perhaps that argument needs to be refined.

I suppose it is worth noting:

Frege isn't remembered for the propositional calculus, which predates him, but for quantifiers and their use in tidying up the predicate calculus, to make it safe for mathematics. — Srap Tasmaner

According to Anthony Kenny's history of philosophy Frege and Peirce simultaneously and independently developed the propositional calculus (which therefore did not predate them, at least in this robust form). But you are right that Frege is remembered for his predicate calculus. The point, though, is that a critique of propositional calculus (and the repeatability of propositions) is a Fregian critique just as much as a critique of predicate calculus would be.

I noticed that too. Absolutely. I think the general thrust of the whole modern Frege-Tarski-model-theoretic approach is to presuppose the existence of the objects within the universe of discourse, and then the questions addressed are which objects satisfy which predicates, and that's all. — Srap Tasmaner

Yes, that's right. Jumping ahead a bit, I am curious about the Fregian presupposition which cleanly distinguishes predicate bearers from predicates, because apparently it associates existence with the former but not the latter. That is, apparently we can talk about non-existent predicates but not non-existent predicate bearers. This seems to reveal an odd lack of parity. For Aristotle a substance and an accident both play by the same general rules, even though an accident has a different kind of being than a substance does.

Excellent citations from Frege. My claim was twofold: 1) that predicate logic restricts what we can say about existence; and 2) we have to start with a logically grammatical proposition that fills the argument slot with a term, thus creating what Frege called a “name,” before we can say whether it exists or not. I’m not sure what “wider than existence” means exactly, but your citation clearly shows that Frege believed we have to presuppose that “sentences [can?] express judgments” and that there is a world out there, about which we are trying to say things. No disagreements here, and sorry if I seemed to say otherwise. — J

I am disputing your (2). For Frege we don't quantify things and then go on to decide whether they actually exist (and this is very much related to your QV thread).

One point about something Frege also says here. He asks: “Can you produce an example where a sentence of the form 'A is B' is meaningful and true, A being a name of an individual, and yet 'There are B’s' is false?” To me, this shows why quantification comes first in his method. — J

His claim seems to be counterfactual, not temporal. "If there are no B's then 'A is B' is neither meaningful nor true."

He requires, correctly, that “A is B” be “meaningful and true” before... — J

You seem to be reading this word "before" into the text, contrary to the text.

If we changed Frege’s question to read: “Can you produce an example where a sentence of the form 'A is B' is unasserted, A being a name of an individual, and yet 'There are B’s' is false?”, the answer would be, Of course we can. — J

I don't know if what you are saying here makes sense, as Frege's whole point is that if there are no B's then 'A is B' cannot be meaningful or true. Were you able to download Lukáš Novák's paper? I think it would be of great benefit.

The irony here is that Frege would presumably not say, "Of course we can." If there are no B's then 'A is B' is not merely unasserted, it is not meaningful. The second quote I gave has Frege literally denying that we can meaningfully deny that something exists (tout court).

But what I was trying to point out (or what I think Roberts means, anyway) is that “the universe of discourse” isn’t neutral or discoverable or God-given or whatever. We have to determine it, which requires quantification. — J

For Frege there is no non-existent universe of discourse. Existence is not an afterthought to quantification.

The charge is more radical than that. The Kimhi-inspired challenge says that the mandatory dissociation of force from sense in logic is wrong. Kimhi: “[Frege and Geach] want to dissociate assertoric force from anything in the composition or form of that which is primarily true or false in a propositional sign.” And yes, I hope Srap keeps pressing his points; we need to interrogate this challenge sharply. — J

Okay. I can see how Frege mandates a dissociation between sense and assertion. Is that the same as mandating a dissociation between sense and force? Or sense and assertoric force? Kimhi seems to believe that something can have assertoric force without being asserted. It seems like Frege wants to make one big distinction (between propositions and their truth values), and Kimhi wants to make lots of small distinctions (between different kinds of force, or different levels of assertoric force).

I find this all fascinating but, as I say, I don't want us to digress. — J

Agreed. :up:

Let me first say that I think the first half of your first post was excellent and deeply relevant. As to the second half, about Moore's paradox, I continue to vacillate on whether it is really relevant. It sheds a bit of light but also raises a lot of issues that seem to be tangential. With that said...

Just for the specific sentence "It is raining but I believe it is not raining", taken as a stand alone. When you read that, you can understand it. Even though you don't know who "I" refers to. You just know it's the person in the sentence. — fdrake

I can't understand it. The received view seems to be that it is absurd. I don't know who "I" refers to. And I don't know who "the person in the sentence" is supposed to refer to. The problem is that, taken at face value, the locution is schizophrenic, and therefore talking about a single speaker is not intuitive.

This is why talking about Moore's paradox seems to require a great deal of explanation and verbiage, in the first place as to how it is being interpreted.

I'd make the same conjured into existence analysis for "I" or "me" in the sentences:
A) It's an egg, I know it's an egg.
B) Ask not for whom the egg tolls, it tolls for me.
C) I have to block out thoughts of eggs so I don't lose my egg.

when they are presented without further context.

Because, as internet brainrot would have it, the who "I" is is ghosted, for real. — fdrake

I worry that we're on a tangent, but the difference is that any statement has a kind of implicit, "I say..." "(I say) It's an egg." Moore's sentence is absurd (and contradictory) because the speaker disagrees with himself (or else has a very idiosyncratic notion of belief).

It seems that originally Moore was looking at two propositions, both of which are said to be true:

1. It is raining
2. I believe it is not raining

He supposes that any two true things can be conjoined and spoken, hence, "<It is raining> and <I believe it is not raining>". I think this reflects the confusion in modern thought where it is presupposed that there can be statements without implicit speakers. This is all somewhat interesting, and there are many ways we could go with it, but you may first have to convince me that it is on topic for this thread.

This is all somewhat interesting, and there are many ways we could go with it, but you may first have to convince me that it is on topic for this thread. — Leontiskos

One way it seems relevant is that understanding the sentence as weird and contradictory on a gut level...

I think this reflects the confusion in modern thought where it is presupposed that there can be statements without implicit speakers. — Leontiskos

pumps the intuition that it must function as an assertion. It would never be uttered in normal circumstances since part of its mechanism asserts something and then undermines the act of assertion.

If it is indeed contradictory in some sense, it is contradictory by virtue of a property it has as an asserted statement, and not its propositional content (what makes it true or false). Since extensionally it's fine. It can be raining while its speaker believes it is not raining, so I can assert "It's raining and I believe it is not raining" and it could true.

The reason that would be relevant is that it highlights a kind of contradiction which can occur by virtue of the acts and attitudes contained in the statement relating to the statement's propositional content, which suggests that there is a type of contradiction which is not governed by the phrase's propositional content. In the sense that a stairway implies distinct floors.

Maybe analysing it in terms of illocutionary forces is good. And it would probably be better to look at it in the form "Sally said "It is raining but I believe it is not raining"", since that dodges all the weird crap involving "I", since we know who is saying it.

"It is raining but I believe it is not raining" is asserted by Sally. It has assertoric force.
"It is raining", the first clause, is something which could be true or false. Sentences that begin with "It is" parse as assertions. EG:

Sally said "It is an egg".
Sally said "It is a nice day today".
Sally said "It is going to be 3 degrees Celsius this evening".

So reading the sentence, the first clause invites us to interpret that Sally has asserted that it is raining.

Then there's "but", which registers an opposition or contrast between what came before and what came after. The sentence is still odd with "and" instead, so I shan't make too much of the "but" in it.

The second clause is "I believe it is not raining", which invites us to interpret it as an assertion. Sentences which begin with "I believe" parse as assertions of belief on the part of their speaker. EG:

Sally said "I believe it is an egg".
Sally said "I believe it is a nice day today".
Sally said "I believe it is going to be 3 degrees Celsius this evening".

When someone asserts something, they are often taken to believe it. I think that's a good default assumption when someone makes a simple claim, and you've no reason to otherwise doubt them. Though it isn't necessary that when someone asserts something, they believe it - they could be lying, they could misspeak, they could be confused, they could be deluded, they could have very unstable beliefs in the moment etc. The expectation is that when someone says "I believe (blah)", they count as asserting (blah) truthfully.

So when Sally says the second clause, "I believe it is not raining", a reading of the phrase in which Sally's assumed to be truthful and sincere associates the "I believe" in the sentence with asserting the claim "It is not raining". So the first clause appears to assert "It is raining", the second clause appears to assert "It is not raining", and those things clash together in our heads.

Nevertheless, Sally is not in a state of contradiction. For Sally only appears to assert that it is raining, and only appears to assert that it is not raining. With logical heads on, that feels like she has just asserted that it is raining and that it is not raining, which is a contradiction. But I think a better explanation of the weirdness in the sentence is that appearing to assert X is both logically and behaviourally consistent with appearing to assert not-X in some contexts. I'd bet this conflict of appearances and a scramble for context is something you sever if you remove the attempt to contextualise the statement (radical interpretation eh?). And moreover, I'd bet that this conflict of appearances is commonplace and essential out in the wild.

Eg, I've said "I don't believe it's raining!" while wincing up at a sky thick with summer rain. And I wasn't insane at the time, I was just complaining. Making sense of "belief" in that statement didn't require too much work on the part of my friend who was with me, since they'd known that was the only day with a good forecast that week.

A similar logic lets you provide a model for Sally's odd phrase. I'm sitting here now, I believe it's not raining since it wasn't forecast to rain this evening last time I checked. But my curtains are closed. I just said "It is raining and I believe it's not raining" aloud... and it turned out it wasn't raining after all, when I opened the curtain.

What I think makes Moore's paradox a good gateway in this discussion is that there's a whole context of cooperative use and interpretation, which contains a myriad of exploitable oppositions and contradictions, that just don't show up when you analyse the phrase as an instance of asserts(p & believes not-p) & asserts(p)=>believes(p). Particularly how you can make sense of it, and the kind of doubt you might have regarding Sally's faculties and situation. Maybe those are the kind of things @J was looking to incorporate into a logic.

Though there remains the question of whether this can be incorporated into normal flavours of logic, whether it's something that can be formalised, whether it should be formalised... and so on.

As an aside, when I said the phrase aloud I felt a powerful compulsion to immediately open the curtain to check... Surely something we expect Sally to have done in my shoes!

I'll give my understanding -- and that without going back to the source -- but you are encouraged to check my work.

the Fregian presupposition which cleanly distinguishes predicate bearers from predicates, because apparently it associates existence with the former but not the latter. — Leontiskos

The distinction is total and fundamental. Frege goes so far as to say you cannot talk about functions (i.e., predicates) at all, because to talk about them is to treat them as objects. We do, nevertheless, talk about them, because it's handy, but he considered this a shortcoming of natural languages. In his system, it is simply not possible: functions cannot be values of variables. ((That's first-order, of course, and it's well known that even to define arithmetic you have to pass on to second-order. I don't recall what he says about this, and whether a switch to classes as stand-ins for functions is good enough. Anyway, there's a gap in my account here.))

He goes further, and says that he cannot even tell you what a function is -- that is, what belongs to the type <function> -- for related reasons, but, and this is a key point, though he cannot tell you what the difference is between an object and a function, he can show you. This is the whole point of the Begriffschrift, to show this difference clearly, perspicaciously. Perforce that means logical form is not really something to be defined (though I don't recall him saying this) but shown.

((This distinction -- that there are some things that can only be shown -- I think had a tremendous influence on Wittgenstein, that was still percolating after the Tractatus, or so I believe.))

That is, apparently we can talk about non-existent predicates but not non-existent predicate bearers. — Leontiskos

Kinda, but I'd be more inclined to say that predicates neither exist nor fail to exist. No more than red is tall or short. It just doesn't apply. Objects are the sorts of things that exist (or fail to), and functions aren't objects.

I don't remember how Frege deals with non-existent objects, or if it even comes up, but in the world he left us, empty classes serve. I can name "the smallest positive rational number" but it will turn out I have defined an empty singleton class. (Extensionally equivalent to any other empty class, but not intensionally, if that matters.)

According to Anthony Kenny's history of philosophy Frege and Peirce simultaneously and independently developed the propositional calculus (which therefore did not predate them, at least in this robust form). — Leontiskos

Peirce had quantifiers too, I hear, but I've never studied his logic. I certainly defer to Kenny -- I just think of the likes of Boole and De Morgan being quite nearly there already.

###

That's all the housekeeping. I'm tagging the quotes below, because this is the meat of it, of course, but I'm going to hold off posting and think a bit more.

I am struggling to see the difference here — Leontiskos

I think we're all on the same page, I'm just using the word "claim" instead of "assert", and also drafting the word "say", all three of which have considerable overlap in everyday speech.

Therefore I would prefer a distinction between a possessor of assertoric force which requires a speaker/asserter and one that does not. I thought J was saying, "This thing has assertoric force even before you pick it up and assert it." — Leontiskos

The key difference is affirming a claim – that is, a statement -- rather than making your own statement about how the world is. — J

Some points I'm mulling:
(1) We have to decide something about locutions like "This sentence claims ..." or "This sign says ..." and so on. I consider it a live option to take them at face-value. It is more common to treat this as a manner of speaking, perhaps glossing "The sign says we have stop" as "If a person were to speak the word printed on the sign, she would be saying that we have to stop," on the smarty-pants grounds that signs don't talk and to say otherwise is anthropomorphizing them. You can also say that they are said to "speak" by courtesy, or argue directly that either an artifact or an abstract object like a proposition, as it were, "borrows" our ability to mean things, that we, as it were, "lend" them our ability to mean things --- as if to say a stop sign is a sort of ghostly police officer, and he has imbued the sign with his spirit.
(2) There's a little bit of a puzzle about the "affirming" language, because it makes asserting sound like it has an extra step, so that it strongly resembles indirect discourse. As if a person making an assertion were "channeling" a spirit guide: there's an internalized claim presented, which you speak on Ephraim's behalf, and by so speaking endorse it.
((3) And here I'll note that this pattern is reminiscent of the prosentential theory of truth, as well as other deflationary theories of truth such as Ramsey and Wittgenstein appeared to hold, such that the use of "... is true" is primarily to endorse what someone else has said.)
(4) @Leontiskos seems almost to suggest that statements have a sort of hole in them, like Frege's functions, waiting for an agent to be inserted and complete the assertion. But we need more than an agent, we need an actual utterance (even if internal), and then we're faced with the problem of intention as well --- some of that context will take care (I'm acting in a play), but some it won't (I was just saying what he wanted to hear).
(5) We have to decide --- (4) mentions some of this --- what we want to count as an assertion. Is it fully disambiguated? Are indexicals all resolved? Is the assertion the statement itself, or the claim in the context and at the time it was made? (Is an assertion an event?) ---- Several of these issues do not arise for the language of mathematics, which is entirely tenseless, to start with, requires no speakers or audience, has no sensitivity to context, etc.
(6) And, finally I guess, what about the social aspect of assertion? We generally think of making an assertion as incurring an obligation to stand by it, perhaps to provide justification, to license others to rely upon it, and so forth. There are pragmatic maxims such as "Do not say what you do not have good evidence for" (Grice) or "Do not say what you do not know" (Williamson). It's easy to talk about all this if assertion is entirely external to the "content" of the statement asserted, but goes wobbly if you want to push some of that into the sentence itself.

One way it seems relevant is that understanding the sentence as weird and contradictory on a gut level... pumps the intuition that it must function as an assertion. It would never be uttered in normal circumstances since part of its mechanism asserts something and then undermines the act of assertion. — fdrake

Good, this is precisely the way that it sheds light on the OP.

And it would probably be better to look at it in the form "Sally said "It is raining but I believe it is not raining"", since that dodges all the weird crap involving "I", since we know who is saying it. — fdrake

Yep, this helps a great deal to clear away the tangential issues.

So when Sally says the second clause, "I believe it is not raining", a reading of the phrase in which Sally's assumed to be truthful and sincere associates the "I believe" in the sentence with asserting the claim "It is not raining". So the first clause appears to assert "It is raining", the second clause appears to assert "It is not raining", and those things clash together in our heads. — fdrake

Right.

Nevertheless, Sally is not in a state of contradiction. — fdrake

Isn't she, though?

For Sally only appears to assert that it is raining, and only appears to assert that it is not raining. — fdrake

This is a different matter as far as I'm concerned:

1. Sally said, "It is raining but I believe it is not raining."
2. Sally appeared to say, "It is raining but I believe it is not raining."

Did she appear to say it or did she say it? First we were dealing with (1), but now we have suddenly switched to talking about (2), which is quite different. If she only appeared to say something, then of course she could not contradict herself. The conclusion would not be, "Sally contradicted herself," but rather, "Sally appeared to contradict herself."

Eg, I've said "I don't believe it's raining!" while wincing up at a sky thick with summer rain. — fdrake

But note that this is no longer an assertion. The assertion would be, drawn out, "It is false that it is the case that it is raining." Or, "It is not the case that it is raining."

A similar logic lets you provide a model for Sally's odd phrase. I'm sitting here now, I believe it's not raining since it wasn't forecast to rain this evening last time are checked. But my curtains are closed. I just said "It is raining and I believe it's not raining" aloud... and it turned out it wasn't raining after all, when I opened the curtain. — fdrake

This doesn't strike me as intelligible. Why did you say it was raining? Were you having a stroke, with random words exiting your mouth? There is no question here that people can say nonsensical things and contradict themselves. The question is whether some utterance is contradictory.

I would say that to assert is to believe. Therefore if Sally asserts that it is raining then she believes that it is raining. This is all that is needed to recognize her contradiction, and this premise seems very secure. What you have done is given some possibilities where she doesn't actually assert, but that strikes me as beside the point.

What I think makes Moore's paradox a good gateway in this discussion is that there's a whole context of cooperative use and interpretation, which contains a myriad of exploitable oppositions and contradictions, that just don't show up when you analyse the phrase as an instance of asserts(p & believes not-p) & asserts(p)=>believes(p). Particularly how you can make sense of it, and the kind of doubt you might have regarding Sally's faculties and situation. Maybe those are the kind of things J was looking to incorporate into a logic. — fdrake

That's fair. You've sufficiently established your thesis about the relevance of Moore's paradox.

Now, in my opinion, the sort of ways that you are defending the coherence of Sally's statement are not going to be plausible ways to critique Frege. But with that said, I have seen folks who are devoted to Fregian logic who have a tendency to oversimplify locutions, so there is that. That's a hard thing to critique.

...there's a whole context of cooperative use and interpretation, which contains a myriad of exploitable oppositions and contradictions, that just don't show up when you analyse the phrase as an instance of asserts(p & believes not-p) & asserts(p)=>believes(p)... — fdrake

I would say that the context-independent interpretation is clearly contradictory, and that it doesn't make much sense to present it as context-independent and expect the hearer to place it in some idiosyncratic context. The additional context could be as simple as, "Sally, a deeply intelligent woman, said..."

Though there remains the question of whether this can be incorporated into normal flavours of logic, whether it's something that can be formalised, whether it should be formalised... and so on. — fdrake

Right.

As an aside, when I said the phrase aloud I felt a powerful compulsion to immediately open the curtain to check... Surely something we expect Sally to have done in my shoes! — fdrake

:grin:

This is a different matter as far as I'm concerned: — Leontiskos

I suppose appears was inopportune. I meant to say that the reader is invited to interpret Sally to be asserting that it is raining, and invited to interpret Sally as asserting that it is not raining in virtue of her statement of belief. Those two things oppose each other somehow.

So with "appears to say", I meant something like "appears to us", "can be interpreted as", and so on. And I mean those expansively, provisionally and contingently. I think part of what makes Moore's Paradox interesting is because it invites us to bracket a normal functioning of language and thus throws it into relief.

I would say that to assert is to believe. Therefore if Sally asserts that it is raining then she believes that it is raining. This is all that is needed to recognize her contradiction, and this premise seems very secure. What you have done is given some possibilities where she doesn't actually assert, but that strikes me as beside the point. — Leontiskos

I don't agree with that. On the basis that I interpret Sally's utterances truthfully and sincerely, I believe it's appropriate to infer that Sally would be in unusual scenario that makes sense of the bizarre composite of asserting that she believes not-p and asserts p in the same breath. In the wild I'd be inclined to read "believe" somewhat figuratively, like an exasperation, or an alternatively that Sally is experiencing a disconnect between whatever engenders her to assert statements and whatever engenders her to assert her own belief in statements. Basically I want to trust Sally rather than calling her out.

My points might not be well targeted at Frege though, so point taken. I hope it is at least relevant to whether assertoric force should be severed from the logic (whatever that is) of an asserted statement.

So we have two different things, sense and reference on the one hand, and illocutionary force on the other. The distinction between them is not, I think, explicit in Frege. It seems instead that the idea of illocutionary force was developed in Oxford and Cambridge in the thirties.

I want to take a look at two more things: the use of ⊢, and the notion of extension.

As I understand it, extensionality enters into Frege's system with Basic Law 5:

$\stackrel{,}{\varepsilon}\!\!f(\varepsilon) =\, \stackrel{,}{\alpha}\!g(\alpha) \equiv \forall x(f(x) = g(x))$

(From SEP). This may be read as "the course-of-values of epsilon is the same as the course of values of alpha if and only if for all x, if x is f then x is g". That is, f and g are the same predicate if and only if every member of f is also a member of g.

The import of this fairly simple point might be clearer if we use a more recent nomenclature and example. I'll refer to the Open Logic Project:

Definition 1.1 (Extensionality). If A and B are sets, then A = B iff every element of A is also an element of B, and vice versa. — Open Logic Project

I'll ask the reader to note that this is the very first formula in this rather extended treatment of logic. This might give an indication of how foundational extensionality is in logic. It is worth I think lingering on what is being said here. Consider the groupings {a,a,b}, {a,b} and {b,a}. Extensionality says that for the purposes of doing the logic that follows, all of these can be treated as {a,b}. What we have here is a tool for simplifying whole groups of expressions down to a single form.

So why the fuss? This all seems straight forward enough. The Open Logic text goes on to give a further example, which I will modify slightly. Consider S, such that S={Ruth}. As it turns out, Ruth is Richard's sibling. So we also have the set S' such that S'={Richard's sibling}. Since Ruth is Richard's sibling, we have S=S'. We say that S and S' differ in sense but not in reference, they differ in intension but not in extension.

Treating groups of things in this way is one of the several great contributions Frege made to logic.

So when Frege 'wrote that his most important contribution to philosophy was “dissociating the assertoric force from the predicate”', he could not have been talking simply or explicitly about illocutionary force, but had in mind at least partly something of the sort given here, were the assertive force of "S is Richard's sister" is simplified by treating it extensionally as S={Ruth}. The "assertoric force" being removed here is at least in part the sense of our statements, so that we might set them aside and deal with the reference.

Again, this is by way of setting out what is at stake here, of what Frege did and how it has developed since. The device on which you are reading this might well not be available if it were not for the developments that took place from considering Basic Law 5. It is central to the logic used, albeit in sorting out its inherent inconsistency as much as in making direct use of it.

The distinction is total and fundamental. Frege goes so far as to say you cannot talk about functions (i.e., predicates) at all, because to talk about them is to treat them as objects. We do, nevertheless, talk about them, because it's handy, but he considered this a shortcoming of natural languages. In his system, it is simply not possible: functions cannot be values of variables. ((That's first-order, of course, and it's well known that even to define arithmetic you have to pass on to second-order. I don't recall what he says about this, and whether a switch to classes as stand-ins for functions is good enough. Anyway, there's a gap in my account here.))

He goes further, and says that he cannot even tell you what a function is -- that is, what belongs to the type <function> -- for related reasons, but, and this is a key point, though he cannot tell you what the difference is between an object and a function, he can show you. This is the whole point of the Begriffschrift, to show this difference clearly, perspicaciously. Perforce that means logical form is not really something to be defined (though I don't recall him saying this) but shown.

((This distinction -- that there are some things that can only be shown -- I think had a tremendous influence on Wittgenstein, that was still percolating after the Tractatus, or so I believe.)) — Srap Tasmaner

Very interesting. That all makes sense, and fills out my understanding a bit.

Kinda, but I'd be more inclined to say that predicates neither exist nor fail to exist. No more than red is tall or short. It just doesn't apply. Objects are the sorts of things that exist (or fail to), and functions aren't objects.

I don't remember how Frege deals with non-existent objects, or if it even comes up, but in the world he left us, empty classes serve. I can name "the smallest positive rational number" but it will turn out I have defined an empty singleton class. (Extensionally equivalent to any other empty class, but not intensionally, if that matters.) — Srap Tasmaner

Good, and this is more accurate than the way I was stating it.

My idea was basically that it is curious that Frege is comfortable saying that Fido is not a cat, but is unwilling to say that Fido does not exist. As you point out, this makes sense for Frege given that the former statement is just a matter of class exclusion, and given that classes—empty or otherwise—are not said to exist. ...But I think @J's confusion about Frege may stem from a similar place. J may be thinking, "Kimhi criticizes Frege for divorcing the sense of a proposition from its assertoric force; 'Fido exists' is a proposition; therefore Frege divorces this proposition's sense from its assertoric force; therefore Frege thinks we can quantify over Fido before predicating existence of Fido." At the same time, J knows that Frege does not accept the idea that existence is a predicate, and so there is a tension.

I don't remember how Frege deals with non-existent objects, or if it even comes up, but in the world he left us, empty classes serve. — Srap Tasmaner

According to the paper I have been citing:

Frege poses the rhetorical question as to whether it is possible to produce an example of a meaningful true statement of the form “A is B”, where A is a proper name, but there are no B’s. The challenge is not met by Frege’s opponent in the dialogue as preserved, but in case we attempted to suggest e.g. the sentence “Smaug is a dragon” as an obvious counterexample (assuming that, of course, there are no dragons), according to his principles Frege would be bound to saying that this is “not a real (i.e., meaningful) sentence”. This counterintuitive result was to be mitigated in a later phase of the development of Frege’s thought, as the distinction drawn between Sinn and Bedeutung enabled, under certain circumstances, expressions (including sentences) to have a Sinn despite lacking a Bedeutung. Nevertheless, the fundamental principle of Frege’s theory, viz. that objects capable of being judged about are, trivially, exclusively existing objects, was never abandoned by Frege, nor did he ever consider it in the least controversial. Quite the opposite – Frege regarded it as empty and tautological, since the term “existing”, insofar as it is applied to individuals, is devoid of any content and as such it does not impose any extensional narrowing: ... — Lukáš Novák, Can We Speak About That Which Is Not?, 158-9

Peirce had quantifiers too, I hear, but I've never studied his logic. I certainly defer to Kenny -- I just think of the likes of Boole and De Morgan being quite nearly there already. — Srap Tasmaner

Yes - my point is only that a critique of Frege could be a critique of his propositional or predicate calculus. I assume that bit about the repeatability of 'p' pertains to the propositional calculus.

I think we're all on the same page, I'm just using the word "claim" instead of "assert", and also drafting the word "say", all three of which have considerable overlap in everyday speech. — Srap Tasmaner

:up:

(4) Leontiskos seems almost to suggest that statements have a sort of hole in them, like Frege's functions, waiting for an agent to be inserted and complete the assertion. But we need more than an agent, we need an actual utterance (even if internal), and then we're faced with the problem of intention as well --- some of that context will take care (I'm acting in a play), but some it won't (I was just saying what he wanted to hear). — Srap Tasmaner

I have often given an impromptu and half-baked account, and this will be no exception. The idea is not so much that they have a hole, but rather that in order to be understood even qua proposition they must have an intentional sense, and they cannot have an intentional sense without an implicit speaker. And there is no neutral intentional sense, or non-intentional sense.

So to take @fdrake's example of Sally asserting Moore's paradox, there is no interpretation of Sally's linguistic utterance which is entirely divorced from an intentional sense and an implicit speaker (in fact fdrake cleared up the implicit speaker problem by making the speaker explicit). According to @J, and as I supposed, Kimhi is after something more substantial and controversial than this. But my point is that there is a kind of default or prima facie intentional sense of every proposition, given the fact that there is no way of interpreting or even apprehending a proposition without assuming some intentional context or another. It's not so much that the proposition has a hole or an intrinsic intentional force by its very nature, but rather that it can never be handled as a proposition, or as a linguistic utterance, without some intentional sense (and implicit speaker) being supplied. There is no possibility of fully prescinding from the intentional sense of a proposition, and the intentional sense would seem to involve a "force" dynamic. In your language we might say that affirming a pre-existing saying will involve one in different intentions and assertoric force, depending on the content of the saying.

But what is a "fictional assertion"? Isn't an assertion supposed to "judge p true"? Kimhi calls this case "assertion by convention" but I don't think that helps either.

This would be a fairly minor point were it not that this thread is trying to understand the exact connection between assertion and truth values. — J

I must be missing something. I can see no more of a problem with fictional assertions than I can with fictional imaginings, fictional events, fictional places, fictional characters and so on.

I'm also not clear on what an "exact connection between assertion and truth values" could be. If I claim something is the case I am either right or wrong depending on whether what I've claimed is the case or not. I can't see what more could be said about that.

So with "appears to say", I meant something like "appears to us", "can be interpreted as", and so on. And I mean those expansively, provisionally and contingently. I think part of what makes Moore's Paradox interesting is because it invites us to bracket a normal functioning of language and thus throws it into relief. — fdrake

Here is my edit in case you didn't see it:

I would say that the context-independent interpretation is clearly contradictory, and that it doesn't make much sense to present it as context-independent and expect the hearer to place it in some idiosyncratic context. The additional context could be as simple as, "Sally, a deeply intelligent woman, said..." — Leontiskos

For me the paradox is too obviously contradictory to be a good candidate for exercises regarding misinterpretation or ambiguity of meaning. It's not a coincidence that you will go through your entire life without ever once hearing someone say, "X is true but I believe it is false."

I would say that the detective who has only Sally's statement in front of him is not a good detective if he multiplies all sorts of theories without any evidential basis for those theories. The exercise feels like being put in the place of a sophist's pupil who receives the task, "Politician Sally said this and nothing more. Find a way to spin it so that she didn't contradict herself. It won't be easy."

And what is the difference between Sally saying, "It is raining but I don't believe it is raining," and Sally saying, "It is raining but it is not raining"? I think the difference is only minor, and the same maneuvers that saved the first could equally well save the second. If this is right then on your approach every statement is unfalsifiably noncontradictory.

And moreover, I'd bet that this conflict of appearances is commonplace and essential out in the wild. — fdrake

When we hear something like this in the wild we either ignore it given our dearth of information, or else we try to gather more information in an attempt to account for the seeming contradiction. But the fact that it is a contradiction at face value will not go away.

Is it a contradiction? Yes. Could there be some extenuating circumstance or idiosyncratic use of language or intent that renders it non-contradictory? Yes.

In the wild I'd be inclined to read "believe" somewhat figuratively, like an exasperation, or an alternatively that Sally is experiencing a disconnect between whatever engenders her to assert statements and whatever engenders her to assert her own belief in statements. Basically I want to trust Sally rather than calling her out. — fdrake

Supposing it never occurs in the wild, does that matter? In that case, "What would an unspoken sentence mean," is a bit like the question about the sound of the tree falling in the unoccupied woods. Perhaps there is a good reason why Moore's sentence is never actually spoken.
Language is flexible, but there are limits to this.

there is a kind of default or prima facie intentional sense of every proposition, given the fact that there is no way of interpreting or even apprehending a proposition without assuming some intentional context or another. — Leontiskos

I'm going to recast what you're saying in this post. (That's just a representative quote.)

I think what you're actually circling around is this: given a sentence (not even necessarily a statement, though that's been the main focus), what would a speaker of this language use it for? You're taking about use, and the default use of statements is assertion.

Yes, and it need not even be limited to logical sentences. It applies to any piece of language. I am saying that we do not have any notion of what a piece of language means without a background of intentional sense and implicit speaker. And yes, the default for statements would seem to be assertion. It is something like asking what a speaker of the language would use it for, but pre-critically.

I suppose a question that arises is whether the material symbol of a pun or ambiguous reference can itself be pointed to. The answer is probably: Yes, but only as a material symbol and not as a proper linguistic sign. To take an old example: 'bank'. In the river sense or the money sense? More to the point, can we reference the single bearer of both separate senses? Sure, but that bearer is a material token rather than a proper and meaningful word: b-a-n-k.

I must be missing something. I can see no more of a problem with fictional assertions than I can with fictional imaginings, fictional events, fictional places, fictional characters and so on. — Janus

Are fictional assertions true? Here is Frege:

In myth and fiction thoughts occur that are neither true nor false. Logic has nothing to do with these. In logic it holds good that every thought is either true or false, tertium non datur. — Frege Reader, 300

The very next sentence of the unabridged text begins the section on dissociating the assertoric force from the predicate.

To take an old example: 'bank'. In the river sense or the money sense? — Leontiskos

To stick with Frege, this is the motivation for the context principle: never ask for the meaning of a word except in the context of (as it is used in) a sentence.

Let's come back to @J 's issue (instead of doing all of philosophy of language).

If you think of something people use, you might think of a tool. Tools capture the problem we face pretty well.

Take a screwdriver. It is a designed artifact, with an intended and, in practice, overwhelmingly common use. That's what it's for. (You might also say that obvious use is what it does, if you're comfortable saying things like that.) When you want to drive a screw, you reach for a screwdriver (of the right sort) because it is the right tool for the job.

But you can use a screwdriver for an unknown number of purposes improvised in the moment. And you can use an unknown number of other objects to drive a screw in a pinch.

The activity of driving screws and screwdrivers are only loosely coupled, though they are indeed coupled and statistics on how screws have been driven and how screwdrivers have been used would certainly show that.

And the analogy to words should be clear, although we're really aiming for sentences, and I think it does no harm to pass the analogy up a rung.

So we have a statement, which, like a screwdriver, carries in its very design its fitness for being asserted; on the other hand, we have the act of assertion which makes use of the appropriate statement. But this coupling is loose: the sentence has other uses as well, and the assertion can be made using other sentences.

For the duration of an act of assertion, there may be a temporary tightening of the coupling --- to produce an utterance you have to commit to a particular sentence. But that grip is immediately slackened: this forum consists almost entirely of people trying to express the same thoughts using different words.

All of which, I think, explains both @J 's sense that statements display assertoric force without themselves being assertions -- in much the way a screwdriver has a clear and unambiguous purpose ---but also why Frege distinguishes them, because the coupling of a statement to the assertion it would naturally be used to make is loose.

We could also note that Frege's sense/reference distinction is in some ways an acknowledgement of such looseness: "3 + 2" and "5" are different expressions denoting the same object; that means you have some freedom in choosing what expression to use to pick out an object, and both proofs and methods for solving equations rely on this possibility of rewriting a mathematical sentence using expressions that can be substituted salva veritate.

If you think of something people use, you might think of a tool. Tools capture the problem we face pretty well. — Srap Tasmaner

Yes, I was thinking of the same analogy.

All of which, I think, explains both J 's sense that statements display assertoric force without themselves being assertions -- in much the way a screwdriver has a clear and unambiguous purpose... — Srap Tasmaner

Right, good.

--but also why Frege distinguishes them, because the coupling of a statement to the assertion it would naturally be used to make is loose. — Srap Tasmaner

I think it is worth asking this question of why Frege distinguishes the assertoric force from the predicate. Your idea seems to be that it is because different predicates can be used to make the same assertion. Here is Frege, and @J may also be interested:

Dissociating assertoric force from the predicate

We can grasp a thought without recognizing it as true. To think is to grasp a thought. Once we have grasped a thought, we can recognize it as true—make a judgement—and give expression to this recognition—make an assertion. We need to be able to express a thought without putting it forward as true. In the Begriffschrift I use a special sign to convey assertoric force: the judgement-stroke. The languages known to me lack such a sign, and assertoric force is closely bound up with the indicative mood of the sentence that forms the main clause. Of course in fiction even such sentences are uttered without assertoric force; but logic has nothing to do with fiction. Fiction apart, it seems that it is only in subordinate clauses that we can express thoughts without asserting them. One should not allow oneself to be misled by this peculiarity of language and confuse grasping a thought and making a judgement. — Frege, Posthumous Writings, 192

(Of course "assertoric force" is here binary, as on/off or true/false)

This is what I would want to question:

So we have a statement, which, like a screwdriver, carries in its very design its fitness for being asserted; on the other hand, we have the act of assertion which makes use of the appropriate statement. But this coupling is loose: the sentence has other uses as well,... — Srap Tasmaner

Now in that bolded phrase you switch from 'statement' to 'sentence', but regardless, I would question the idea that a statement has other uses than assertion. Appealing again to my idea of intentional senses and implicit speakers, I don't think statements are ever wielded while wholly prescinding from their assertoric nature. I don't know whether Frege would think of a quoted phrase as a "subordinate clause" (or the equivalent German), but suppose I put quotation marks around, "The grass is green." This is sufficient to indicate that I am expressing a thought without asserting it to be true. Is that proposition assertoric? No and yes. No, insofar as I am explicitly indicating that I do not intend to assert it. Yes, insofar as the intentional sense and the implicit speaker associated in my mind with the proposition both have everything to do with assertion; or in other words: it is a statement, albeit quarantined and muted. Similarly, tigers are dangerous. If we put a tiger in a cage does it become non-dangerous? Yes and no, for the cage only exists because it is dangerous, and the cage is what holds the danger at bay.

This is part of what I understand @J to be saying in the OP and elsewhere, but as I said in my first post, I am not yet convinced that it is something Fregian logic must worry about.

Are fictional assertions true? — Leontiskos

They are fictionally, as opposed to actually, true or false, or their truth or falsity may not be sepcified in the work. I still don't see a problem, just a matter of different contexts.

cage — Leontiskos

sandbox — bongo fury

?

@Banno @Leontiskos @Srap Tasmaner @RussellA @fdrake
Whew, we’ve got the makings here of a solid weeklong conference on Frege and Kimhi!

Impossible to address all the interesting points and questions, but I’ll do my best to respond to folks one by one. Leontiskos up first!

(1) Assertion is (a person, an agent) claiming that the possible state of affairs, let's say, described by a statement does in fact hold.
(2) Assertion is (a person, an agent) affirming the claim about the world made by a statement.
— Srap Tasmaner

I am struggling to see the difference here, but maybe that is just me. — Leontiskos

The difference is that (1) is an assertion, couched of course in language, about something in the world, e.g. the green grass. (2) is an assertion, couched as affirmation or denial (which could be in symbolic language rather than words) of the sentence used in (1) about the grass.

The irony is that Kimhi claims there is no difference – this is his monism. He says there’s no “logical gap” between (1) and (2). But in order to appreciate how he could say such a thing, we first have to get clear on what appears to be the difference. Hope this helps.

I thought J was saying, "This thing has assertoric force even before you pick it up and assert it." — Leontiskos

I know, this is really hard to be clear about. When I suggested “adding a nuance to the vocabulary” that would separate force from assertion, I was suggesting a possible way to clarify. My idea was that we could then talk about “displaying force” without “asserting.” So, to respond to your paraphrase: No, not exactly. I‛m suggesting that we should stop thinking of “force” as something that only an assertion can create. The term “assertoric force” kind of twists our arm into thinking that there’s no force without assertion. So instead, “This statement has force [positive or negative predication] even before you pick it up and assert it.”

For Frege we don't quantify things and then go on to decide whether they actually exist — Leontiskos

Right, but it’s the introduction of the argument into the function that allows us to claim it exists. I see how you could have read my “before we can say whether it exists or not” to mean that there would be a further decision process. But no, all I’m positing is that, for Frege, ontological commitment can only be shown through his predicate logic.

Okay. I can see how Frege mandates a dissociation between sense and assertion. Is that the same as mandating a dissociation between sense and force? Or sense and assertoric force? Kimhi seems to believe that something can have assertoric force without being asserted. It seems like Frege wants to make one big distinction (between propositions and their truth values), and Kimhi wants to make lots of small distinctions (between different kinds of force, or different levels of assertoric force). — Leontiskos

Good questions. If you accept my proposal to disambiguate “force” from “assertion,” then we need to clarify the relations among all these terms, which is a headache, not just for Kimhi -- much less so than for Frege, as you point out. Just to repeat the point from above, though: I think Kimhi believes that something can have force (not assertoric force) without being asserted.

I think @J's confusion about Frege may stem from a similar place. J may be thinking, "Kimhi criticizes Frege for divorcing the sense of a proposition from its assertoric force; 'Fido exists' is a proposition; therefore Frege divorces this proposition's sense from its assertoric force; therefore Frege thinks we can quantify over Fido before predicating existence of Fido." At the same time, J knows that Frege does not accept the idea that existence is a predicate, and so there is a tension. — Leontiskos

Hmmm. Well, ‛Fido exists’ isn’t a proposition, if I understand Frege. So for that very reason, we don’t have to do anything with Fido other than use him in a function in order to claim he exists. We do have to do that much, though.

Can you say more about this point? It’s possible I’m not following you.


Oh, and about the Novak paper: Your link didn’t seem to take me there. Mind verifying and posting it again? Thanks.

Now in that bolded phrase you switch from 'statement' to 'sentence', — Leontiskos

And I'm usually so careful about that. At any rate, I'm just using 'statement' to mean 'indicative mood sentence.'

I would question the idea that a statement has other uses than assertion — Leontiskos

Are you using 'statement' here the same way I was, or as 'a sentence that is being asserted'? (Or something else? Everyday terminology does not lend itself to the distinctions we're discussing.)

Sentences is in the indicative mood are of course used to ask questions, give commands, suggest doubt, make wishes, and so on.

Is that proposition assertoric? No and yes. — Leontiskos

This is the whole point of my screwdriver discussion. Driving screws is an activity, like making assertions. The favored use of screwdrivers is driving screws, as the favored use of indicative mood sentences is making assertions. But, as I argued, this coupling is loose. So we are right to recognize that a screwdriver is longing to drive screws, and this is the most joy it can find in life, but we still might drive screws without it, or use it for something else. What I'm not sure there's grounds for saying is that the screwdriver itself is always kinda driving screws; it's really not, though that is its special purpose and we are right to recognize it.

Oh, and about the Novak paper: Your link didn’t seem to take me there. Mind verifying and posting it again? Thanks. — J

Here is the link that was buried in the original post, which I have verified is working: "Can We Speak About That Which Is Not? Actualism and Possibilism in Analytic Philosophy and Scholasticism," by Lukáš Novák. (Pages 155-188)

So when Frege 'wrote that his most important contribution to philosophy was “dissociating the assertoric force from the predicate”', he could not have been talking simply or explicitly about illocutionary force . . . — Banno

Right.

. . . but had in mind at least partly something of the sort given here, where the assertive force of "S is Richard's sister" is simplified by treating it extensionally as S={Ruth} — Banno

In other words, by identifying the two extensional sets as the same, we're able to "make the assertion" that S is Richard's sister without any appeal to some actual act of assertion (i.e. illocutionary act). Have I understood you? And if I have, do you see the addition of the judgment stroke as referring to assertion in this sense? This is one place where Frege confuses me. When he says that the judgment stroke marks "a true thought," does he mean a thought asserted to be true, or one that actually is true? So again, different senses of "assertion" might arise here.

We can grasp a thought without recognizing it as true. — Frege, Posthumous Writings, 192

Not to beat a dead horse but this is another exemplar of the pattern I've been talking about. We do generally credit candid speech -- and there's an argument that it must be so for a language to have any consistent semantics -- and it is also true that by and large when we speak we intend and expect others to believe us, but for all that the coupling of understanding and belief is loose: you can understand what someone says without believing it or endorsing it yourself.

I'm also not clear on what an "exact connection between assertion and truth values" could be. If I claim something is the case I am either right or wrong depending on whether what I've claimed is the case or not. I can't see what more could be said about that. — Janus

Perhaps nothing more, in that simple case. But as this thread demonstrates, "assertion" gets used in some much more complicated and ambiguous contexts. As @Banno points out, above, Frege didn't think in terms of actual illocutionary acts such as the one you're using as an example. And Russell talks about a "non-psychological sense" of assertion whereby we can say that "If p then q" asserts an implication without asserting either p or q. And I would add, though Russell doesn't, that the implication "If p then q" can be asserted on paper, so to speak, without anyone claiming it's true.

All of which, I think, explains both J 's sense that statements display assertoric force without themselves being assertions -- in much the way a screwdriver has a clear and unambiguous purpose ---but also why Frege distinguishes them, because the coupling of a statement to the assertion it would naturally be used to make is loose. — Srap Tasmaner

I like the screwdriver analogy. ‛The grass is green’ is all set, ready to go -- its "purpose" is what I'm calling its force -- but someone does have to pick it up and use it. A lot of the interest in this thread centers on how best to understand picking it up and using it. The Fregean “loosening” is also good – for him, it is logically necessary that assertion does nothing to sense.

The distinction [between arguments/subjects and predicates in Frege] is total and fundamental. — Srap Tasmaner

This passage from Julian Roberts is worth quoting in full:

Psychologism rests on the assumption that we can coherently say things about thinking itself. To rule this out, something stronger than asymmetry of the function-argument relation is needed. Frege achieves this strengthening by interpreting ‛functions’ as ‛thinking’. This semantic interpretation involves an ontological commitment. The distinction between functions and arguments becomes an ontological one. So the asymmetry of the relation is ‛explained’ by saying that the entities denoted at either side of the relation are themselves, ontologically, distinct. There are function-entities, and there are argument-entities. As a result of this ontological move, the rule ceases to be a formal rule of calculation and becomes a declaration about the necessary properties of certain entities. Thoughts, in a word, are ontologically distinct from their objects; and that is why thoughts may never be arguments. — Julian Roberts, The Logic of Reflection

I agree with this, and it seems to support your understanding as well. Notice, though, that Roberts puts “explained” in scare-quotes. Fair enough: Is this really an explanation or just an “ontological move”?

I think we're all on the same page, I'm just using the word "claim" instead of "assert", and also drafting the word "say", all three of which have considerable overlap in everyday speech. — Srap Tasmaner

Yeah, and it’s this overlap that’s driving us all slightly crazy! So far we lack a rigorous analysis of how the terms relate.

There's a little bit of a puzzle about the "affirming" language, because it makes asserting sound like it has an extra step, so that it strongly resembles indirect discourse. — Srap Tasmaner

That’s a big chunk of Kimhi’s argument as well. I was going to save the whole “believes that” / “thinks that” / “judges that” issue for a different OP, but (sigh) of course it’s relevant here too.

As if a person making an assertion were "channeling" a spirit guide: there's an internalized claim presented, which you speak on Ephraim's behalf, and by so speaking endorse it. — Srap Tasmaner

Nice to meet another Merrill fan!

Nice to meet another Merrill fan! — J

I didn't consider for a moment anyone would get that reference!

Is this really an explanation or just an “ontological move”? — J

This phrase, "just an ontological move", is interesting.

My first instinct, as suggested earlier with force and content, is to suggest that what's at work here is not the agent adapting her behavior to the facts of the world, but adopting a strategy to treat this as force and that as content, this as function and that as object. (Frege admits as much when he says talking about functions is treating them as objects.)

The explanation on offer, then, would be an explanation of "how I intend to proceed", how I intend to treat these "things".

Now if you think there is one true ontology out there, you'll naturally ask if functions and objects are really distinct ontological categories; that distinction may be overlooked or abstracted away when it's not of interest, but it lies in wait as an aspect of reality, and this is why we are even able, when we so desire, to distinguish them. The contrary position, I suppose, would be that we "impose" the distinction, it's all in our head, or in our language, or in our culture, whatever. I guess you just deny that any explanation for how we do this is necessary, or you you reach for the magic of consciousness, or something complicated and Peircean.

Obviously I don't know what to say about all that, but I like getting the facts we're trying to explain clear, rather than passing as quickly as possible to the contest of ideologies. To that end, I prefer looking at what people do, and how they think and talk about what they do, to speculating about how things are.

And here I'll add to my previous suggestion: people are very aware of the subtleties of assertion, whether you meant what you said, whether you have committed to a certain claim and can be held to account, etc., and recognize that there are degrees of belief and degrees of conviction in speaking. We put considerable effort into managing these uncertainties. (Think of the conventions surrounding agreement to a contract.)

So I don't really expect an absolute, almost mechanical explanation of what assertion is and how it works, and when faced with a deliberately mechanical system like Frege's I'll tend to see it as a strategy that must be useful for some purposes, with no expectation that it is some total solution to a supposed problem.

Wow, kinda rambled there. Sorry about that.