This topic has been discussed and/or skirted around for centuries in one form or another. The result is that it is extremely complex discourse to navigate. There are numerous different notions of reference throughout history, not all of which are compatible with one another. The sheer quantity of subtopics and/or otherwise tangential aspects is downright daunting.

The irony(or difficulty depending) is that what we're doing here - the activity we're all involved in - is the referent of our own discourse. That holds good for all historical accounts as well. Via discourse, we are drawing an other's attention to the fact that we draw an other's attention to some thing in particular that our attention is already upon. We are accounting for how that happens. We do not need to draw an other's attention to the fact that we're drawing an other's attention to some thing in particular in order to draw an other's attention to that. In other words, we do not need to talk about the fact that we can draw an other's attention to some thing or other in order to be able to do so. The other need not know that that is being done. We need not know that we're doing it.

Drawing an other's attention to the same thing in particular that our attention is already upon is something that all of us began doing long before we came to realize that we were doing it. Drawing an other's attention to the same thing in particular that our attention is already upon is not existentially dependent upon our awareness of it. It does not require being named. Rather, it is something that exists in it's entirety(it is something that we do) prior to our knowing that we're doing it. Drawing an other's attention to the same thing in particular that our attention is already upon is not existentially dependent upon our ability to take an account of it.

We can draw an other's attention to the same thing in particular that our attention is already upon by virtue of ostension and/or description, despite the fact that the description need not be true. What we cannot do is take account of the fact that we're drawing their attention to some thing that ours is already upon by virtue of false description. What we're doing when we draw an other's attention to some thing that ours is already upon cannot be adequately taken into account by a single name/term. It just cannot be done. Rather, what we're doing can only be taken into account by virtue of true description.

That's the irony here.

There is a clear distinction between the necessary and sufficient conditions for drawing an other's attention to some thing that our attention is already upon and our accounts thereof. Drawing an other's attention to some thing that our attention is already focused upon is not something that is capable of being true/false. It's something that happens. The necessary and sufficient conditions for drawing an other's attention to the same thing in particular that our attention is already upon do not include our thought/belief and/or terminological jargon about the fact that we're doing it. Rather, our thought/belief and/or jargon(our reports/accounts) about drawing an other's attention to some thing that ours is already upon can be true/false.

We cannot use false descriptions about what we're doing when we're taking an account of drawing an other's attention to some thing that ours is already upon and expect to arrive at an acceptable conception and/or notion of reference. Any and all acceptable conceptions/notions of reference must be able to properly take account of the fact that in all cases of referring we are drawing an other's attention to some thing that ours is already upon. This can only be arrived at by virtue of true descriptions about how that happens.

That sheds a bit of much needed light upon the elephant in the room.

There's a difference between what successful reference takes and what a proper account of successful reference takes. The former can happen with the use of false descriptions. The latter cannot consist of such.

I wanted to ask anyone interested in their thoughts about this assertion.

Strictly speaking, counterfactuals can only be de dicto, and not de re.

Is that true?

One is about the report of the thing, the other is about the thing. Sometimes we talk about the report. Is that not the thing in those cases?

The distinction is very nuanced however, and that rendering cannot capture all that convention says about it. The SEP article ought be confusing enough for anyone's joy.

I like the idea of substitution without losing meaning as a means for adequate translation. It's handy for checking on the coherency of terminological use. It also often renders the notion of 'logical' entailment useless. Gettier plays on it when the substitution quite simply does not hold. What is claimed to be entailed by 'X' can have different truth conditions... different meaning.

One is about the report of the thing, the other is about the thing. Sometimes we talk about the report. Is that not the thing in those cases? — creativesoul

Yes. I'm quite confused. I was under the impression that rigid designators can only be talked about what is said, not the thing itself. The same or similar would apply to counterfactuals.

Well early on Kripke says that 'designator' is a term that covers both, names and descriptions.

What does that mean?

So, we refer by ostension and/or description. We do so prior to taking what we're doing into account. Our accounts are existentially dependent upon our doing so, but not the other way around. What our doing so is existentially dependent upon, and what our accounts of doing so are existentially dependent upon have two distinct sets of necessary and sufficient conditions, and/or elemental constituents. Doing so does not require thinking about thought and belief. Accounting for doing so does. Doing so does not require true description of what we're referring to. Accounting for what we're doing does. Better have thought/belief right...

Jane can refer to Joe by saying "the man who killed Bob" even when Joe did not, because she believes Joe killed Bob. She need not know Joe's name. It is arguable whether or not she needs to know something else(have some other true belief) about Joe. We cannot take proper account of what Jane is doing by virtue of using description if our descriptions of what she's doing are false.

What she's doing is drawing an other's attention to the same thing that her attention is already upon by virtue of expressing her own thought and belief. Those belief statements need not be true in order to be about Joe. They need not be true in order to draw an other's attention to Joe. That is because they need not be true in order to be meaningful. A listener of Jane's(when Joe is not present) can know who she is referring to by sharing her belief about Joe, or by knowing that she believes it. Neither person here needs to know Joe's name. Both would need to know what Joe looks like in order for the description to successfully refer.

This is the interesting part...

Knowing what "the man who killed Bob" means is quite simply inadequate for successful reference if knowing that is equivalent to knowing which man uniquely satisfies that description. Knowing that Allen killed Bob does not help a listener at all when it comes to knowing who Jane is talking about.

This might help:

https://pdfs.semanticscholar.org/31d4/81ec126c3baef191ea0990832bf81727de87.pdf

Abstract: The problem of counterfactual attitudes de re was identified by Ninan (2008) as a challenge for standard theories of de re. I show that once counterfactual non-de re attitudes are properly analyzed, trivial composition of such an analysis for them with the analysis of doxastic de re provides the solution to the problem. Thus there is no independent phenomenon of counterfactual attitudes de re, and therefore no problem as such.

I imagine Kripke would argue that the impossibility of de re counterfactuals only obtains for rigid designators. Of course, one might reasonably speculate that de re statements are a fortiori rigid.

So, expanding on my topic about quantifying counterfactuals. I don't think they can be stated without adherence to a de dicto assertion of their existence dependent on our world. If that is the case, then de re assertions are nonsensical for counterfactuals.

Then what do you think about quantification across possible worlds? Is it nonsensical to do so?

Formally, there is no problem with it. I have never seen a philosophical criticism that was compelling either.

Formally, there is no problem with it. I have never seen a philosophical criticism that was compelling either. — Snakes Alive

So, bear with my confusion! If we are to quantify over possible worlds then, we can only "measure" (quantify) counterfactuals by an accessibility relation to our own world. Therefore how can we assert something as necessarily true in all possible world's if quantification of modal relations (counterfactuals) is/are restricted to only our world?

I already posted this in a separate topic, but, my confusion hasn't ceased since.

What I found out was that "actualist" interpretations of QML (Quantified Modal Logic) tend to agree with this sentiment.

And, this is where I stumbled on the Barcan Formula, and then from there I started reading about de re and de dicto propositional attitudes. The Barcan Formula seems to support actualist interpretations of QML.

Quine would disagree, but the SEP entry on QML asserts that his criticism holds little weight nowadays.

Knowing what "the man who killed Bob" means is quite simply inadequate for successful reference if knowing that is equivalent to knowing which man uniquely satisfies that description. Knowing that Allen killed Bob does not help a listener at all when it comes to knowing who Jane is talking about. — creativesoul

Furthermore, knowing who Jane is talking about does not help a listener at all to know which man satisfies that description...

"The man who killed Bob" is a definite description as a result of the fact that there is only one man who satisfies the conditions of the description. Jane's case shows that one can use a definite description in order to successfully refer even when it is the case that the language user does not know who actually satisfies the conditions therein(even when their belief about who satisfies those conditions is false).

And yet, the definite description "the man who killed Bob" sets out conditions that only Allen satisfies.

One can use a definite description to successfully draw another's attention to the same thing that one's attention is already upon even when that thing does not satisfy the description. That's because doing so is belief based. It is also the case that the DD is satisfied by a unique individual. That's because doing so is truth based. Belief presupposes truth. Hence, Jane can successfully refer to Joe by virtue of using a definite description that only Allen satisfies because she believes that Joe satisfies it.

So...

Jane's case clearly shows the irrevocable role that belief plays in all successful reference. It doesn't matter whether or not the definite description(belief) is true of Jane's referent when it comes to her being able to successfully draw an other's attention to the same thing that her attention is already upon.

However, Jane's case does not warrant concluding that definite descriptions are inadequate for successful reference. Rather, it shows that definite descriptions are capable of being used in more than one way as a means for successful reference.

A comment on the discussion over the last few pages, first.

Seems to me that there has been considerable loose play between truth and belief in the preceding critique of Kripke. That is, perhaps folk have failed to notice just how different "Paris is in France" is from "Jenny believes that Paris is in France". It introduces a second modality on top of necessity, and scampers the possibility of direct substitution.

And "knowing" suffers the same fate, since it involves both belief and truth.

It all adds up to a bit of a mess.

Respectfully Banno...

While I am all for using the simplest framework possible for taking proper account, I am also all for taking proper account. If the possibility for direct substitution is hampered by virtue of taking proper account, then it is not an issue with the proper account my friend. It's an issue with the inherent inability of formal logic to offer a proper account of belief.

More importantly, as a result of not taking the speaker's belief into proper account, if we are claiming that definite descriptions are adequate for picking out a unique individual, and we're not drawing and maintaining the crucial distinction between statements of belief and definitive descriptions, then we are conflating truth and belief. This is clearly shown because we are forced to say things like Jane's referent is Allen because Allen uniquely satisfies the conditions within "the man who killed Bob". Jane is not referring to Allen. So our saying that is quite simply not true. That notion of reference consists of false descriptions about what we're doing when we draw an other's attention to the same thing that ours is already upon in cases like Jane... and the champagne case as well. That account is unacceptable.

Kripke points this out, or so I am told, as a problem with versions of descriptivism. I would agree. While Kripke did not attempt to clear up what was going on. I have been.

So far as I know, Quine isn't taken seriously on this matter.

So, bear with my confusion! If we are to quantify over possible worlds then, we can only "measure" (quantify) counterfactuals by an accessibility relation to our own world. Therefore how can we assert something as necessarily true in all possible world's if quantification of modal relations (counterfactuals) is/are restricted to only our world? — Wallows

This paragraph just doesn't make sense.

What do you mean by "measure" or "quantify" counterfactuals?

There is no such thing as "an accessibility relation to our own world." Accessibility relations hold among a set of worlds – it doesn't matter which one is actual, and the standard modal logic does not even mark an actual world.

Modal logic's semantics determines the truth of a formula relative to a model, world and variable assignment – this also makes no mention of the "actual world." If you want to include a special, designated actual world to the model, you can do this, but it's just not needed for the semantics. The whole point of the modal logic is that any arbitrary formula can be evaluated fro truth or falsity relative to any world. And once you have a semantics for counterfactuals, you can plug this into your modal logic.

Accessibility relations hold among a set of worlds – it doesn't matter which one is actual, and the standard modal logic does not even mark an actual world. — Snakes Alive

This depends on whether you are an actualist or possibilist for QML.

he whole point of the modal logic is that any arbitrary formula can be evaluated fro truth or falsity relative to any world. And once you have a semantics for counterfactuals, you can plug this into your modal logic. — Snakes Alive

Again, I am professing an actualist interpretation of QML. If you assume my position then Counterfactuals can only be truth apt relative to our world. This is an assumption that I understand applies to both actualist and possibilist interpretations.

This depends on whether you are an actualist or possibilist for QML. — Wallows

No it doesn't. The modal logic is a formal device, indifferent to metaphysical interpretations of modality.

Again, I am professing an actualist interpretation of QML. If you assume my position then Counterfactuals can only be truth apt relative to our world. This is an assumption that I understand applies to both actualist and possibilist interpretations. — Wallows

There is no modal logic that in principle only allows the evaluation of a formula for truth relative to the actual world (you could create a vacuous frame with only one possible world, but this would be a pointless exercise, and says something only about the frame, not the logic). Indeed the entire point and expressive power of the logic is that it allows evaluation relative to multiple worlds. If you remove this, then you have a vacuous modal logic, i.e. one that only has the expressive power of a non-modal logic.

What's your take on the Barcan Formula?

The validity of the Barcan formulae follows independently from ordinary, independently plausible semantics for the universal quantifier and the box. If one objects to it, one had better have a pretty good reason, and I'm not aware of one.

I suspect that resistance to it is due to the confusion that distinct worlds 'have' distinct domains of individuals associated with them, over which quantifiers operate. You can make your logics this way, but it's probably a bad idea. Many bad ideas in logic come from philosophers having qualms independent of the logic, and trying to force their prejudices back into the logic, with bad results. Presumably, in this case it has something to do with the idea that the domain of quantification represents what 'exists,' which is not right.

The validity of the Barcan formulae follows independently from ordinary, independently plausible semantics for the universal quantifier and the box. If one objects to it, one had better have a pretty good reason, and I'm not aware of one. — Snakes Alive

I believe the point I'm trying to make is the following and bears some semblance to the Barcan Formula in restricting the domain of truth-aptness to the actual world (note that the range can span to an infinite amount of counterfactuals, whilst the domain can be restricted to the actual world):

I can stipulate a possible world where an event might have happened otherwise; but, the framing condition for doing so, to sound technical, will always be restricted to the world where the stipulation was made with respect to that event or state of affairs. A roundabout way of positing counterfactuals.

I believe the point I'm trying to make is the following and bears some semblance to the Barcan Formula in restricting the domain of truth-aptness to the actual world: — Wallows

The Barcan formula doesn't 'restrict the domain of truth-aptness,' whatever that's supposed to mean. It is just a formula, valid on an ordinary modal logic, and its validity follows from the way quantifiers and modal operators are ordinarily interpreted.

The issue you're talking about is that the Barcan formula's validity makes it impossible that worlds accessible from a world have 'larger' domains than the world from which they're accessed: in other words, domains don't 'grow' across accessibility relations. This is fine, however, not because of commitments to modal actualism, but because to think that distinct worlds are associated with distinct domains in the first place is a mistake. One can make a logic this way, but it is probably a bad idea. There is just one domain of individuals, and it is not anchored to worlds to begin with.

Definite description is capable of being used to successfully refer to something other than the unique object which satisfies the conditions therein. That is always the case regarding false descriptions of someone regardless of whether or not anyone else actually satisfies the conditions of the description.
The conditions of the description "the man who killed Bob" could not be satisfied if a woman was the murderer. Yet, "the man who killed Bob" can be used to successfully refer to someone other than the murderer, regardless.

Jane's case exemplifies this.

Definite description is also capable of being used to successfully refer to the unique object which does satisfy the conditions of the description.

A proper account of Jane's case(including Allen) shows that.

What all this clearly shows is that the unique individual satisfying the conditions of a definitive description is not always the referent of a speaker using that definitive description, and thus... the referent of a speaker using definitive description is not always determined by the truth conditions of their belief statement(definitive description), and/or the unique individual satisfying those conditions.

The issue you're talking about is that the Barcan formula's validity makes it impossible that worlds accessible from a world have 'larger' domains than the world from which they're accessed: in other words, domains don't 'grow' across accessibility relations. This is fine, however, not because of commitments to modal actualism, but because to think that distinct worlds are associated with distinct domains in the first place is a mistake. — Snakes Alive

So, when we talk about possible worlds, and specifically make stipulations about counterfactuals, then we are restricted to the domain of the actual world? Does that make sense?

One can make a logic this way, but it is probably a bad idea. There is just one domain of individuals, and it is not anchored to worlds to begin with. — Snakes Alive

What do you mean?

So, when we talk about possible worlds, and specifically make stipulations about counterfactuals, then we are restricted to the domain of the actual world? Does that make sense? — Wallows

Not really. Are you talking about the domain of individuals?

What do you mean? — Wallows

In a standard quantified modal logic, there is a domain of individuals, and a set of possible worlds. Each world does not have 'its own' domain of individuals associated with it.

Are you talking about the domain of individuals? — Snakes Alive

I guess so. What is it?