I think abstract objects are products of analysis. — frank

It seems to me rather that abstract objects as artefacts of generalization are products of recognition and synthesis.

Please write the OP and I’ll join the conversation.

I don't it's a case of ambiguity. You just don't have enough information to know what's being asked. You'd have to go back and get the request clarified, right? — frank

Because it's ambiguous. I'm struggling here to guess how you understand that word.

But communication failure is not the main point here; it's that you have the option to treat shared abstract properties as singletons, but you do not have that option with the objects instantiating them.

Sorry, wasn't trying to be dense. You mean the marbles are identified separately, but the colors, being universals, can merge?

Yeah there you go.

I thought of this because if instead you were told to sort the marbles by color, there's no ambiguity -- well, not this particular ambiguity. Someone might distinguish the reds and blues more finely, but then we'd be back at the first ambiguity, which is really along a different axis, right?

I wouldn't even throw this little puzzle out there if it weren't for what Kimhi says about propositions versus actual occurrences. The options you get dealing with marbles are different from the options you get dealing with the colors of marbles. — Srap Tasmaner

The relevant part in Kimhi seems to be this (p. 39)

"In what follows, I shall call the correct understanding of Frege’s observation Wittgenstein’s point, and I shall call the conclusion Geach and Frege draw from it—that assertoric force must be dissociated from a proposition’s semantical significance—Frege’s point. We shall see that Frege’s point is mistaken. It only seems necessary if we accept certain functionalist (and more generally, compositionalist) assumptions about logical complexity. Correctly understood as Wittgenstein’s point, Frege’s observation concerns actual occurrences of a proposition and amounts to the full context principle; misunderstood as Frege’s point it runs together the symbolic and actual occurrences of a proposition and limits the context principle to atomic propositions."

My rough intuition about this, at this stage, is that the full context principle assigns meanings (Fregean senses) to subsentential expressions (e.g. names predicates and logical connectives) not only in the context of whole sentences but also in the context the other sentences a sentence relates to in a language game. (And this is not exhausted merely by inferential relations). This is the point about actual occurrences of propositions since, unlike propositions considered in abstracta (with regards only to linguistic form), actual occurrences always make moves in language games (be they games that we play with ourselves). Negating a proposition is one such language game, which Martin analyses in great details, and asserting it is another. It is tempting here to assume that there ought to be a common content that is being either asserted or (assertively) denied, since you may be asserting something and I am denying it. (It must be the content, and asserting it or denying it are the extraneous forces). Martin denies this, I think, but I must read him further to understand how contents are differently understood or differently individuated within different language games that warrant different ways to mark the content/force distinction (while my asserting that p is the case still logically bears on your denying "it").

@Srap Tasmaner

Here are a few further thought regarding the Wittgensteinian full context principle and the distinct ways in which propositional contents (Fregean thoughts) are being individuated in assertion or negation.

Consider the issue that I raised at the end of my previous post: '...while my asserting that p is the case still logically bears on your denying "it" ' In order for your denial to logically bear on my assertion, the game of making assertions must bear on the game of denying asserted contents. But, surely, denying a claim made by someone else isn't equivalent to denying a claim made by oneself, which it at worst irrational and a best could be construed as a retraction. So, maybe we can say that although asserting p and denying p individuate the thought content p (as distinguished by the common linguistic form "Satz" p) differently, that is not a problem for rational engagement since the issue precisely is that there is a rational dispute regarding which move should be made (in light of evidence, etc.) in the language game with this linguistic form, and hence a dispute regarding the proper way to individuate the content of the claim (to be either asserted or denied).

Someone might still puzzle over my seemingly gratuitous assertion (which I tentatively ascribe to Martin) that, specifically, the content p, when asserted, isn't the same as the content p, when denied (i.e. when its negation is asserted). And that isn't reducible to a mere distinction in force, although it can be explained, non-reductively, as such a distinction. But indicating the force amounts to gesturing to a specific sort of language game. The letter p that is common to both therefore merely signifies the common linguistic form. Why not say that there is a common purported content (how things are, state of affairs) that is here denied and there asserted? Consider the content of the assertion "this apple is red" or "p". When one asserts that, what one thinks is that the apple is red. When you use the same linguistic form to deny that the apple is red, one says "it is not the case that this apple is red" or "not p", this can not be construed as you standing in the different relations (with the "force" of negation, say) to the same state of affairs consisting in the apple being red since, from your point of view, there is no such state of affairs in the world. So, it is necessary for you to characterise the content of your denial — i.e. what it is that you deny — differently. This appears to be related to the issue of object dependent de re senses where I say that this apple is red (while I purport to refer to it demonstratively) but there is no apple owing to some optical illusion. In that case, Evans or McDowell would say that the thought content one might express in other circumstance with p (when there is an apple in sight) simply isn't available to be thought, in this case.

Maybe what might further help dissolve the puzzle (or the counter-intuitiveness) in the claim that when you assert that there is an apple on the table and I deny it there isn't a common content of thought that you assert and that I deny would be to place this activity within a more fully specified embodied linguistic game of finding apples in an environment where illusions occasionally occur (or apples move around) such that our capacities to deictically refer to them, and remember where they are, are fallible. In the context of such a game, asserting that there is an apple on this or that table is the actualization of a capacity to locate apples in point of communicating to other where a particular apple is (and hence transmitting the content of a de re sense). Denying that there is an apple on the table is a move in the very same language game that aims at signifying that someone's general capacity to locate an apple on the table (de dicto) has failed to be actualized properly, owing to the occurrence of an illusion or owing to the apple having unexpectedly moved away since it had last been looked at. Within this context, it seems much more apparent how claims and denials about the same linguistic contents actually bear on distinct propositional contents (distinct Fregean thoughts) since the successfully (true) asserted contents are de re while the successfully denied (false) contents are de dicto.

the full context principle assigns meanings (Fregean senses) to subsentential expressions (e.g. names predicates and logical connectives) not only in the context of whole sentences but also in the context the other sentences a sentence relates to in a language game — Pierre-Normand

I have mixed feelings.

Yes, this is the natural way to go, but I think there's a risk that it flattens communication too much. We need different levels to play off against each other, unless you intend words to drag other language-games along with them to provide the necessary contrast. (Which, okay, and I know LW talked that way sometimes, but there are several issues with that.)

The examples I have in mind are irony, sarcasm, exaggeration, implicature (and we could go on) -- none of these are at all intelligible without something that counts, even if temporarily, as the literal meaning to play off of. That could just be some "other" meaning, but it has to be something widely enough available to count as "literal" for the relevant speech community.

So I lean toward what I take to be Grice's approach: logic is all we need for the semantic connections between sentences; pragmatics can't even get started unless that analysis stands as it is. For instance, to trigger the recognition of implicature, a response has to violate the principle of cooperation if taken literally. If you assign as meaning the use being made of the sentence (the speaker's meaning rather than the sentence meaning), you undermine the whole process. What's more, the sentence used to trigger the recognition of implicature (or irony, exaggeration, etc.) has to have a particular literal meaning for it to be suitable for the job in the first place, and thus selected and uttered. (Obligatory chess analogy: being used to block check doesn't change what a piece is, only its role in this position in this game.)

how contents are differently understood or differently individuated within different language games that warrant different ways to mark the content/force distinction — Pierre-Normand

Indeed.

It seems like I'm very nearly giving content to logic and force to pragmatics, and I'm not sure that's a terrible first thought, but I'd be interested in alternatives.

And nothing I've said does much to ground or explain the content/force distinction ...

---- I'll try to catch up so we're not having two conversations, but I'm going to hold off on your post just after this one for a bit.

---- I'll try to catch up so we're not having two conversations, but I'm going to hold off on your post just after this one for a bit. — Srap Tasmaner

:up:

Why not say that there is a common purported content (how things are, state of affairs) that is here denied and there asserted? Consider the content of the assertion "this apple is red" or "p". When one asserts that, what one thinks is that the apple is red. When you use the same linguistic form to deny that the apple is red, one says "it is not the case that this apple is red" or "not p", this can not be construed as you standing in the different relations (with the "force" of negation, say) to the same state of affairs consisting in the apple being red since, from your point of view, there is no such state of affairs in the world. — Pierre-Normand

(my bolding)

I can't provide an adequate response to this, with appropriate citations and such, but I can say this: I am deeply, deeply suspicious of the model suggested here. It's the alternating monologues model of conversation, where the speaker simply expresses their thought out loud within earshot of an audience; language exists to mediate the connection of my mind to the world, and my audience more or less eavesdrops on my review of that relation.

I'm not going to deny the sentence I bolded, but I think this is entirely backwards. It's the other person who mediates my connection to the world, and I hers, and language is an important part of how we connect to each other, for that purpose.

Force, in particular, if we can define such a thing adequately, is meaningless absent an audience. I'm not denying that we internalize this dynamic; we very much do. But the origins of assertion, question, command, and so on, clearly --- to me, at least --- lie in our relations with others. We can play at making assertions to ourselves, asking ourselves questions, giving ourselves orders, but nothing could be more plainly derivative of what we do with others, and it is necessarily a make-believe sort of business.

(Oh, you could, if you wanted, cobble together some sense of force appropriate to our dealings with the non-minded world --- putting nature to the question, imposing your will on her, blah blah blah --- but I don't have any idea why anyone would speak a language for that purpose, since nature isn't listening.)

This doesn't directly address the scenario you described, much less provide an alternative analysis, but the starting point strikes me as completely hopeless.

I wonder if the impetus for separating content from assertoric force is related to the 'view from nowhere.'

It's language that's detached from any particular human. It's like the narrator of the novel we inhabit.

It's the alternating monologues model of conversation, where the speaker simply expresses their thought out loud within earshot of an audience; language exists to mediate the connection of my mind to the world, and my audience more or less eavesdrops on my review of that relation. — Srap Tasmaner

I was rather arguing for something that is very much the polar opposite of that view. I alluded in passing to the case where we assert something to ourselves (as in making a judgement in the course of practical deliberation) as just a special case, or personal use, that we can make of the resources of a language game that is essentially public. In alluding to this, I wanted to highlight that even in this kind of personal use that we make of moves of asserting and denying — that are at home in paradigmatically public, embodied and interpersonal language games — the propositional contents (Fregean senses) or our asserting, or judging that p, or of judging that p isn't the case (i.e. that not-p is the case) also mark out distinct contents despite both of them being linguistically expressed as "p". Showing this was the point of my elaboration of the public game we might play with apples in my last paragraph, where the aim of the game of asserting and denying is to coordinate our joint engagements with the apples, and internal monologues or private representations drop our of the picture entirely.

I wonder if the impetus for separating content from assertoric force is related to the 'view from nowhere.'

It's language that's detached from any particular human. It's like the narrator of the novel we inhabit. — frank

It may indeed be indirectly related to the view from nowhere. When arguing against representationalism, or the idea that our thoughts or the propositional contents of our utterances consist in a private inner realm that is problematically communicated to other or problematically grounded in the "external" world, John McDowell speaks of the "sideways-on view that places empirical reality beyond the space of concepts". Once we conceive of being, or of things as they are in themselves, as residing outside of the space of concepts, we indeed are conceiving of them as viewed from nowhere.

I was rather arguing for something that is very much the polar opposite of that view. — Pierre-Normand

Which is what I had expected! Obviously I must have misread you.

I suppose we can leave it there for now, though I'll certainly go back through your posts.

Several posts back it felt like we were finally dealing with the central issue of this thread, so I would feel bad if we lost sight of that.

Great references again, thank you. I've just gotten onto McDowell and Rodl, and now I can add Martin and Conant. That was the good thing about grad school, you could actually read all day . . .

What still remains an open question to me (even though I lean towards the Wittgensteinian quietism of McDowell) is whether their accounts of this self-conscious propositional unity constitutes an improvement over the charitable accounts, put forth by Evans and McDowell, of what Frege was trying to accomplish when he sought to individuate thought/proposition at the level of sense (Sinn) rather than at the level of extensional reference (Bedeutung) in order to account both for the rationality of the thinking subject and for the objective purport of their thoughts. — Pierre-Normand

Well put. What you're calling the charitable accounts do seem to capture what Frege wanted to do, and why (especially the desire to explain the objectivity of thoughts), but I'll have to read more of McDowell.

Me too.

The passage from Kimhi you cite was key to my OP. I appreciate your analysis. Interestingly, Kimhi seems to equate "Wittgenstein" with the Tractatus. There is very little indication that Kimhi has absorbed later Witt, and he never speaks in terms of language games.

The full context principle assigns meanings (Fregean senses) to subsentential expressions (e.g. names, predicates and logical connectives) not only in the context of whole sentences but also in the context of the other sentences a sentence relates to in a language game. (And this is not exhausted merely by inferential relations). This is the point about actual occurrences of propositions — Pierre-Normand

I think Kimhi would accept this as a paraphrase of his thought. If we want to continue to speak Kimhian, we could replace "in a language game" with "in an actual occurrence of p."

I think it is, very much so. The "view from nowhere," on Thomas Nagel's understanding, is precisely the longed-for objective view that, in this case, would grant thoughts some autonomy from being "merely about us," in whatever way you want to take that idea. Everything I understand about Frege says that he would have heartily endorsed the view from nowhere as essential even to the simple self-identity of one instance of 'p' with another.

:up:

one way of buttressing the idea that propositions have independent existence is to align them with the mental, rather than the physical. — Srap Tasmaner

Yes, and Kimhi wants to do neither. He’s after a unity of thinking and being. And no, I still haven't completely grasped how he gets there.

the question is in what sense "what he said" is a thought, while the actual words spoken were merely a physical "expression" or even representation of that thought. — Srap Tasmaner

To that I would add a third discrimination: “thought” in the Fregean sense of “proposition” or “propositional content.” So we have:
- “Thought” as a mental event. Mine is not yours, etc.
- “Thought” as the content of that mental event – this is the Fregean sense. Here, my thought is the same as yours, or can be.
- The language in which thought-as-content may be expressed. This is important because we need to leave room for synonymy between languages.

I often find it helpful to invoke Karl Popper’s “Three Worlds” idea in this sort of discussion. Popper’s World 1 is the physical world, the world which would exist, in a rough-and-ready way, without any consciousnesses. World 2 is the mental world, understood strictly as mental events, or brain events. Each occurs at a Time T-1, -2, etc. and is unique in the sense that no one can have the exact same event. World 3 is the world of our mental constructions, our ideas, propositions, theories, poems. As I said above, it’s basically the content of thoughts – it’s what the World 2 events are about.

There is a point you can make about content, the proposition, and a different point you can make about actual occurrences, in which that content features, but Frege, I think it is claimed, forgets what he's about and tries to make a single point about both, or tries to make a point about one that can only be made about the other, and somehow tricks himself into thinking he has not mixed up the two. — Srap Tasmaner

I read Kimhi to be saying that both the symbolic and actual occurrences of p are World 3 objects – they are contents of thought, one unasserted, one asserted – but the actual occurrences can’t be understood if we strip away any connection with assertoric force/truth value and try to treat them as if they were synonymous with symbolic occurrences.

That makes sense. I read this in the SEP article on Lotze (he lived from 1817 to 1881):

"The new model of the academic researcher demands the replacement of prominent personalities—i.e., romantic ‘geniuses’—by the new exemplar of the sober scientist."

Lotze navigated a transition from German idealism to a philosophy that focuses on values and analysis. Maybe we could say the whole western culture was shifting toward an emphasis on quiet objectivity rather than an emotional idealistic view. Frege was a part of that?

We live on the other side of the 20th Century, so we know what kind of terrain that shift was headed toward: eliminative materialism, functionalism, inscrutability of reference. Complaining about the view from nowhere seems to target something that's foundational for us: scientific objectivity.

I love this topic, but all I'll say here (it really needs its own thread) is that any plausible account of the view from nowhere would have to explain why it's a desideratum of rational thought, while at the same time leaving room for subjectivity, for me being a Kierkegaardian individual rather than merely an item in the Hegelian System or, worse, the victim of "rationalization" in Weber's sense. I agree, the 20th century was a rough time for objectivity -- it doesn't look as attractive as it used to -- but I'm not sure this forum would make sense without some faith in the view from nowhere.

I wouldn't sell him short. — J

Fair enough. I misspoke, since what I have at hand is not Kimhi's book but the reports of his book found hereabouts. The ambiguity may not be his. Similarly it's the discussion of PM on this thread that I found problematic, not the account given by Kimhi. I'm not sure what to make of the paragraph you quote, in particular the conclusion: " Therefore, Geach's understanding of Frege's observation conflates the two senses of propositional occurrence: symbolic and actual." Not at all sure what "symbolic and actual" is doing here.

If I were to comment it would be to point out the benefit of treating Modus Ponens as a rule of inference. We might do well to avoid trying to work out what an "actual argument of the form modus ponens" might be, apart from one that makes use of the rule of inference.

It's common for those versed in axiomatic or ealry logics to think in terms of "Laws Of Thought", such as Modus Ponens. It's worth seeing what "Modus Ponens" looks like in the Open Logic text. There's the rule of "→ elimination" for Natural Deduction,
This is understand as that if in a natural deduction you can write "φ" and you can write "φ→ψ" then this rule entitles writing "ψ". and

Proposition 12.19. If Γ ⊢φ and Γ ⊢φ →ψ, then Γ ⊢ψ. — Open Logic, p.180

or the subsequent Deduction Theorem: Γ ∪{φ}⊢ψ if and only if Γ ⊢φ →ψ. The first of these is about what can be written in Γ, the second about what happens when Γ is extended. In the sequent calculus, Modus Ponens is derived. Cut provides a generalisation of Modus Ponens, but the cut elimination theorem displaces it from any centrality. Modus Ponens is not any one expression, and is not always fundamental.

It's worth asking what sort of thing a "Law of thought" might be. Presumably a Law of thought must be such that it hold true in all cases. If there are such entities, then there is "one true logic", that which uses these Laws. Recent work has questioned this, asking why there could not be a variety of correct logics. And some investigation of the view that there might be no correct logics. Don't be too quick to reject logical pluralism. Given any system of rules, it is a simple question of imagination to set out what things would be like if those rules are contradicted.

But if logical pluralism is even a possibility, then the idea of "Laws of Thought" would be undermined, and with it, perhaps the psychologism that Kimhi would reintroduce into logic.

To this mix we might add more general questions about following rules. Language does not only function by observing convention, but also by rejecting convention, by hitting the nail right on the thumb, putting the cart before the hearse, or bringing your chickens home to roast.

We must give up the idea of a clearly defined shared structure which
language-users acquire and then apply to cases. And we should try again to say how convention in any important sense is involved in language; or, as I think, we should give up the attempt to illuminate how we communicate by appeal to conventions. — Davidson, A Nice Derangement of Epitaphs

In this regard it seems to me that Frege, Kimhi and others have a view of philosophy as seeking a closed, consistent and complete account of how things are. We know from other considerations that we can easily achieve completeness usually at the expense of coherence. I'll opt for coherence over completeness.

While I'm rabbiting on, it's worth taking a bit more. care with notions of objectivity. Einstein's great contribution wasn't to set out a view from nowhere, but to give an account that worked anywhere. He recast the Laws of physics so that they worked for anyone, regardless of their frame of reference. Objectivity is about finding explanations that work from multiple points of view; about having explanations that work regardless of where one stands, and so for the many rather than for the one. it's not about the view from nowhere, but about the view from anywhere.

Added: Frege's contribution was in this direction. By setting the assertive force aside he permitted wider agreement. Some might disagree that p is the case, while agreeing that if p were the case, q would follow. Logic allows us to examine the consequences of our options without commitment. Who was it said " "It is the mark of an educated mind to be able to entertain a thought without accepting it"?

Who was it said " "It is the mark of an educated mind to be able to entertain a thought without accepting it"? — Banno

This is commonly (miss-)attributed to Aristotle who was making a different point in Nicomachean Ethics, Book I, Chapter 3. "In the same spirit, therefore, should each type of statement be received; for it is the mark of an educated man to look for precision in each class of things just so far as the nature of the subject admits; it is evidently equally foolish to accept probable reasoning from a mathematician and to demand from a rhetorician demonstrative proofs." (Ross translation)

However, I can't fail to notice the resemblance of Aristotle's idea here bears to the "full context principle" Kimhi attributes to Wittgenstein (as interpreted by Martin)!

Well, I hope it's clear that I have more sympathy for game formalism than you have expressed, and also that I am sceptical that there is a thing that is the meaning of an utterance, such that it might be got at by the "full context principle" - not that I am too sure what that might be.

the full context principle assigns meanings (Fregean senses) to subsentential expressions (e.g. names predicates and logical connectives) not only in the context of whole sentences but also in the context the other sentences a sentence relates to in a language game — Pierre-Normand

Should we picture the meaning of a sentence as something we approach only asymptotically, as our comprehension of the context improves?

Suppose we go along with Frege and think of thoughts (propositions) as objective, in his sense, not the personal property of anyone.

What about judgment? What about inference?

If I judge P true, and so do you, aren't we making something we'd want to call "the same judgment"?

If I infer Q from P, and so do you, aren't we making something we'd want to call "the same inference"?

We can go further.

Suppose I forget to "carry the 1" in a piece of simple arithmetic, and so do you. Aren't we making "the same mistake"?

How far can this analogy go? Couldn't we have the same taste in music? The same fear of snakes?

Or is there some reason all of these things aren't just as objective as Frege's propositions?

Nice.

If I judge P true, and so do you, aren't we making something we'd want to call "the same judgment"? — Srap Tasmaner

If we make the same judgment, we agree on the same proposition. We aren't agreeing on a sentence or an utterance. Would you like to work through the argument that establishes that?

That post reads like I addressed @J as a defender of Frege, but it was meant to be addressed to @J as someone who believes Kimhi is making a point about Frege, but it keeps coming out as just the sort of thing we'd expect Frege to say.

The truth cannot be told in such a way as to be understood and not be believed. — Blake, not Frege or Kimhi

Should we picture the meaning of a sentence as something we approach only asymptotically, as our comprehension of the context improves? — Banno

As the quote from Aristotle's Nicomachean Ethics suggest (analogically) getting clear on the meaning of a sentence must rather be relativized to the context (the specificity and pragmatic point of the language game within which it occurs). Each language game furnishes a different context. But I am getting ahead of myself. Before inquiring about the nature or unicity of the allegedly forceless "proposition" (conceived as the content p that is common to thoughts that may occur with different forces), we can also inquire, with Kimhi, about the source of the unity of a propositionally complex judgement within which a simple predicative proposition occurs. In On Redrawing the Force Content Distinction, Christian Martin sets up the problem thus:

"The idea that thought is inherently forceful can only become an insight if it is concretely shown how that idea is compatible with the fact that embedded thoughts and dependent acts of thinking must do without a force of their own. If thoughts as such are tied to some force or other, while embedded thoughts (e. g. p qua part of not-p) do not directly come along with a force of their own, it must be clarified how the indirect connection to force, which embedded thoughts must indeed come along with, is to be understood. That is, it must be clarified how dependent logical acts that have an embedded thought as their content, and the overarching logical act that does indeed bear a force of its own interlock with each other such as to provide for the unity of a propositionally complex thought."

I don't want to say much more before reading the second part of Martin's paper, and I'm currently rereading the first part. I likely will need to revise substantially my tentative analysis of the content of negated propositions in the specific context of the game of locating apples that I described earlier, before drawing more general conclusions about the context dependence of force/content distinctions.