Yeah that's quite interesting, and I think both (yours and mine) represent types of triangulation.

A further curiosity is that parasitic reference has to be self-consciously contrastive, so it's the sort of thing a parent can engage in; on the other hand, children are said to be learning when they manage this sort of "playing along," "calling things what you call them," but they lack the distinction between the two ways of doing this.

Cheers.

AI tells me it's in Ruth Barcan Marcus’s Modalities: Philosophical Essays (Oxford, 1993), especially in the early papers and appended discussions. I can only see a limited preview.

This is interesting, thanks. My question to @Banno focused on something a little different. If we say that reference, as a matter of fact, requires triangulation, then it would follow that whatever it is I'm doing, privately, is not an example of referring. That's one way of setting it out conceptually. The second way would be to say that the question is not a factual one at all. We have a term, "reference," and we're considering how best to use it in order to carve up the conceptual territory. So it might be that we want to reserve "reference" for the cases where triangulation is involved. In that case, we need another term to describe what I'm doing, privately. I was asking Banno which of these outlooks he favors -- hope that makes sense.

Going back over this, it seems to me that the reference is now fixed by the indexical, "the man over there", and not by the description "He has champagne in his glass". — Banno

I think it's yet a third thing: The reference is fixed by the description "the man about whom I say . . ."

We have a term, "reference," and we're considering how best to use it in order to carve up the conceptual territory. So it might be that we want to reserve "reference" for the cases where triangulation is involved. In that case, we need another term to describe what I'm doing, privately. I was asking Banno which of these outlooks he favors -- hope that makes sense. — J

I'm jumping in here without an clear understanding of how "triangulation" is being used here (a relation between the scribble, the writer and the reader, or the scribble, what the scribble references and the scribbler). I don't see a triangulation unless we leave out the reader/listener as we have three things in the scribbler, the scribble and what the scribble refers to. So it appears there may be some trying to fit a square into a triangle-shaped hole.

Say you were the only person in the world. Why would you even consider drawing scribbles to refer to other things that are not scribbles? Well, maybe you might want to keep track of time, like how many days passed since the last rain, or when the deer migrate, etc. Maybe you want to remember the past so you might make scribbles that remind you of something in the past. I have spreadsheets containing information that I've never shown to anyone. They are for my eyes only, and only I would understand the relationship between the information within the cells. So it seems to me that a reader/listener is not necessary for reference, just some rule where some scribble refers to something else and the rule-maker. Agreement with others about the rules of reference come later.

whatever it is I'm doing, privately, is not an example of referring. — J

I just don't think that follows from anything.

Everytime someone argues that blah is born out of social practices which continue to support and inform it, someone will say, "So if I privately blah, in my mind, you're saying it's not really blah?!"

No, of course not. It's why I tried to make clear in that post that both views of language at least attempt to end up with both social and private uses.

Here, consider reading. Famously, reading used to only be done aloud. To this day, children are overwhelmingly taught to read aloud: your teacher tells you, out loud, what sounds the letters make; the student demonstrates their ability by making those sounds out loud. It is how this knowledge is transferred to the next generation. It makes clear the relation between our use of oral and written language.

Would anyone then conclude that reading silently is not "really" reading? No.

whatever it is I'm doing, privately, is not an example of referring.
— J

I just don't think that follows from anything. — Srap Tasmaner

That's certainly how it seems to me, and for the same reasons you cite, but I want to understand why @Banno might think otherwise.

Say you were the only person in the world. Why would you even consider drawing scribbles to refer to other things that are not scribbles? Well, maybe you might want to keep track of time, like how many days passed since the last rain, or when the deer migrate, etc. — Harry Hindu

Same point. Surely Robinson Crusoe did some private referring! But again, let's hear more about the "no private reference" case. I do think it's crucially different from "no private language," which may be where Banno is coming from.

It depends on how one is defining "refer" and "language". Is referring and language the same thing, or can you have one without the other?

Surely Robinson Crusoe did some private referring! — J

I know you're kidding, but that's clearly the wrong test case. He was taught to refer to things using first oral and then written language. Even gesturing at things is learned behavior.

I don't think anyone has made the claim that reference is ever done using a private language. The claim you made is that someone has to comprehend the speaker's reference in order to for there to be any reference. As @J pointed out, that's an odd usage of the term "reference." I don't think it's what Kripke is talking about.

What I'm saying is that we only have something we call "reference", the thing that we do with referring expressions like names and descriptions, so that we can talk about things with other people. More than that, our individual cognitive capacities are shaped by our interactions with other people, so the sorts of things we want to talk about are already the objects or potential objects of shared cognition.

And I think our referring practices are shaped by the goal of achieving shared cognition. In conversation, both speaker and audience contribute: the speaker says what they believe will be enough to direct the audience's attention, expecting the audience to draw on whatever they can to "fill in the blanks" (context, shared history, reason).

Why does any of this matter? Because words are a "just enough" technology that evolved for cooperative use; a word, even a name, is not something that carries its full meaning like a payload. Words are more like hints and nudges and suggestions. They are incomplete by nature.

And so it is with using them to refer. We should expect that to be a partial, incomplete business.

I think it's tempting here to think of this on the analogy of regular human finitude: in our minds we pick out objects to talk about and we do so perfectly, completely, but words are imperfect and ambiguous and are kind of a lousy tool for communicating our pure intentionality.

I doubt that story, but about all I have in the way of argument is that our cognitive habits and capacities are shaped by just this sort of good enough exchange. My suspicion is that we largely think this way as well. And this makes a little more sense if you think of your cognition as overwhelmingly shared, not as the work of an isolated mind that occasionally ventures out to express itself.

It depends on how one is defining "refer" and "language". Is referring and language the same thing, or can you have one without the other? — Harry Hindu

I think this highlights the question we're discussing. I'm just thinking this through myself, but there has to be a difference between "private language" and "private reference," doesn't there? As frank says, we don't need a private language to refer privately. We can use the community language we all know. That's not what's private about private reference -- rather, I'm arguing that it's the independence from "triangulation" or the need to have a listener comprehend the speaker's reference. I read Srap as talking about language, not reference, and if that's so, then what Srap says is clearly true: Robinson Crusoe needs to have inherited and practiced a non-private language before he can make up any designations for the flotsam that washes up on his beach. But once he does that, why would we deny that he's referring to said flotsam when he thinks about it, or perhaps makes a list of tasks?

we only have something we call "reference", the thing that we do with referring expressions like names and descriptions, so that we can talk about things with other people. — Srap Tasmaner

Our posts just crossed! So, as to this: That's what I'm questioning. Why couldn't it be true that we need reference equally to talk to ourselves? I'm not even sure that your version would be true as a genetic account -- who knows which came first, private naming or public discourse, or whether they were simultaneous?

Yeah that's quite interesting, and I think both (yours and mine) represent types of triangulation.

A further curiosity is that parasitic reference has to be self-consciously contrastive, so it's the sort of thing a parent can engage in; on the other hand, children are said to be learning when they manage this sort of "playing along," "calling things what you call them," but they lack the distinction between the two ways of doing this. — Srap Tasmaner

That's right, and therefore I think the interesting question asks how parasitic or triangulated reference fits into reference in general. There is certainly a sense in which parasitic or triangulated reference is secondary, and this is seen in the way that the child does not begin with it. I think it is also true that lying requires parasitic or triangulated reference, and lying too is a secondary form of reference.

More generally, people are doing slightly different things when they refer, but it would seem that all acts of reference have commonalities.

What I'm saying is that we only have something we call "reference", the thing that we do with referring expressions like names and descriptions, so that we can talk about things with other people. More than that, our individual cognitive capacities are shaped by our interactions with other people, so the sorts of things we want to talk about are already the objects or potential objects of shared cognition. — Srap Tasmaner

That's fairly persuasive as a theory of the origin of speech, but I don't think it necessarily indicates that we can't speak meaningfully while alone. The part of the motor cortex that orchestrates speech is separated from the portion that handles comprehension. It's not clear that the unity of consciousness we enjoy today is the way humans have always been. It may be that talking to ourselves has been around as long as talking to each other has.

It's important to remember that skills don't necessarily arise for a need, but having arisen, they find a need (can't remember who said that, Democritus?) It may be that speech just randomly emerged as a continuous stream accompanying experience. In time, it became valuable for group dynamics. We really don't know.

Why does any of this matter? Because words are a "just enough" technology that evolved for cooperative use; a word, even a name, is not something that carries its full meaning like a payload. Words are more like hints and nudges and suggestions. They are incomplete by nature.

And so it is with using them to refer. We should expect that to be a partial, incomplete business. — Srap Tasmaner

Absolutely.

I doubt that story, but about all I have in the way of argument is that our cognitive habits and capacities are shaped by just this sort of good enough exchange. My suspicion is that we largely think this way as well. And this makes a little more sense if you think of your cognition as overwhelmingly shared, not as the work of an isolated mind that occasionally ventures out to express itself. — Srap Tasmaner

Sure. I think there's convincing evidence that speech capability is innate, but interaction is necessary for development.

That's not what's private about private reference -- rather, I'm arguing that it's the independence from "triangulation" or the need to have a listener comprehend the speaker's reference. — J

And I'm suggesting that this "independence" is to some degree illusory, in two senses: the sorts of things you think are the sorts of things you could express, whether you do or not; and secondly, they are that way because you learned how to think from other people.

Roughly, I want to convince to feel, behind every thought you have and every word you utter, millions of years of evolution and hundreds of thousands of years (at least) of culture. The thoughts and words of countless ancestors echo through your thoughts of words. Everytime you choose as the starting point for analysis "What am I doing all by myself?" that's a mistake. It's the tail wagging the dog.

We really don't know. — frank

I agree.

I hope no one will take the forcefulness with which I'm expressing my view to indicate dogmatism. I could be entirely wrong.

Honestly I think I'm inclined to push this sort of inside-out approach just because so much of our tradition presumes the opposite. I'm curious to see if other approaches might be enlightening.

Honestly I think I'm inclined to push this sort of inside-out approach just because so much of our tradition presumes the opposite. — Srap Tasmaner

And that is by and large a good idea, which I appreciate. We don't want to be taking words like "private" or "mental" to imply some lonely kingdom we inhabit and populate by ourselves, as if it (and we) were something new on the Earth. That never happens.

Why couldn't it be true that we need reference equally to talk to ourselves? I'm not even sure that your version would be true as a genetic account -- who knows which came first, private naming or public discourse, or whether they were simultaneous? — J

It should be clear from other posts that I agree we do not know, and may not be able to know.

But I am still a partisan of the communication first view, or, rather, shared intentionality and cognition first. A lot of that I get from Tomasello. I was playing with my granddaughter last year after watching one of his talks and it's shocking how obvious this is once you look for it: I roll the ball toward her and she glances up at me then back at the ball until she traps it in her pudgy little hands and immediately her face pops up to look at me. (Did I do it right? Is this how we do it?) Then she focuses on the ball so she can roll it toward me and as soon as she lets go, her face pops up again to see, again, if she's doing it right. It's constant. We start as early as possible learning to see the world through the eyes of our caretakers. I think talking builds on and elaborates this fundamental orientation of ours toward communal cognition.

That's fairly persuasive as a theory of the origin of speech, but I don't think it necessarily indicates that we can't speak meaningfully while alone. The part of the motor cortex that orchestrates speech is separated from the portion that handles comprehension. It's not clear that the unity of consciousness we enjoy today is the way humans have always been. It may be that talking to ourselves has been around as long as talking to each other has. — frank

Well, there is a theory that reading in the ancient classical world was always reading out loud. Reading to oneself in sllence developed later. Sadly, I have lost my note of where I got this story. However, one can see this process at work by watching small children as they learn to read. Even it is not true, it seems to me to be a plausible myth of the origin of talking to oneself.

That's not what's private about private reference -- rather, I'm arguing that it's the independence from "triangulation" or the need to have a listener comprehend the speaker's reference. — J

Is there any reason why we can't distinguish two phases of reference? The speech act and the hearer's response, which acts as feeback to bring into line any misunderstandings.
Where speaker and hearer are one and the same person, we have, so to speak a limiting case. One of the limitations is that the tendency, over time, of language to wander from its original starting-point. A solitary speaker has, and requires, no feedback.
The involvement of other people puts a brake on this for a solitary individual. Of course wandering still occurs, but occurs as the result of many individuals communicating with others, so the changes are controlled by consensus.

We assign an interpretation to this syntax by assigning an individual to each of the individual variables, a to "a", b to "b", and so on.
So, assigning a property to an individual happens in a different part of the logic to assigning a name to an individual. — Banno

This quote from @Banno is from the other thread, explaining to me how formal logical systems are constructed. This process seems to me to assume that assigning properties to individuals presupposes the assignation of names to their references. But perhaps I have misunderstood.
Of course, that's not a problem if we are simply using natural language as opposed to constructing one. But it would be nice to be able to say that referring and describing are interdependent activities. They really need each other.
Incidentally, ostensive definition is the traditional way of escaping from the endless circle of descriptions (I believe). Wittgenstein's point about this is, as I understand it, that there is no guarantee of success. But if we can sort out misunderstandings, why do we need a guarantee of success?

Sadly, I have lost my note of where I got this story. — Ludwig V

Saint Anselm? I'll have to google now.

Ambrose!

Roughly, I want to convince to feel, behind every thought you have and every word you utter, millions of years of evolution and hundreds of thousands of years (at least) of culture. The thoughts and words of countless ancestors echo through your thoughts of words. Everytime you choose as the starting point for analysis "What am I doing all by myself?" that's a mistake. It's the tail wagging the dog. — Srap Tasmaner

I think this is why people like @J talk to themselves aloud. It creates a quasi-externalization and a quasi-triangulation.

We start as early as possible learning to see the world through the eyes of our caretakers. I think talking builds on and elaborates this fundamental orientation of ours toward communal cognition. — Srap Tasmaner

Yes, but we see the world through their eyes. Your granddaughter asks implicitly, "Am I doing it right?" I don't think she is implicitly asking, "Am I doing it your way?" That's why I think—at least according to a prominent rational aspect—the parasitic move is secondary. If I tell my nephew what a spider names, or how many legs a spider has, he is immediately thinking in objective terms. He thinks, "A spider has eight legs," not, "My uncle says/thinks a spider has eight legs." The shift to the latter is actually quite complex and difficult. In fact the concept and basis of error eludes most of the TPFers on a regular basis.

Well, there is a theory that reading in the ancient classical world was always reading out loud. Reading to oneself in sllence developed later. Sadly, I have lost my note of where I got this story. — Ludwig V

St. Augustine found St. Ambrose's silent reading strange and abnormal, which is one evidence we have for that thesis.

Saint Anselm? I'll have to google now. — Srap Tasmaner

Thanks very much. Perhaps I should have paused before posting.

St. Augustine was considered strange in that he practiced silent reading. — Leontiskos

Thanks. It's good to know I was not wrong.

Thanks. It's good to know I was not wrong. — Ludwig V

I miffed that a bit. It was actually St. Augustine writing about St. Ambrose, who practiced silent reading. Augustine found it strange.

I will try to get to the big can of worms you opened later tonight.

I am familiar with this passage. It's in one of the earlier books of Confessions.

That people used to often read aloud might also be a reason for the heavy preference of verse up until the modern era, the epic poem being for them what the novel is for us, and even scientific and political topics were covered with poetry. Another common hypothesis is that it is easier to remember text that is in rhyming meter and books were so expensive that memory was essential. Also, you can often do more emotionally and thematically with clever verse using less text, and when you have to kill a bunch of oxen to make a single book, economy is key.

Even gesturing at things is learned behavior.

Is it? I suppose in some sense it must be, because it requires some stimulus input, but it also seems about as innate as almost any behavior can be. Honey bee dances would seem to refer, but bees are not "taught," although they do have a "critical period" in development where lack of exposure to other dancers will hamper (but not remove) their dancing abilities. For humans, pointing seems similar, being fairly universal and innate (as much as anything can be).

Unrelated comment for the group:

Animal "reference" is an interesting example. Can dogs or dolphins refer? It seems they can in at least some basic sense. Dolphin pods send out scouts and use signals to direct the pod during attempts to catch fish, etc.

I'd question though whether it makes sense to talk about "bee's dances" or "dolphin clicks and whistles" doing the referring. The bees and dolphins refer, the dances and whistles are their means of doing this. Certainly, we can look at the means of communication in abstraction, but zoologists are pretty good about always trying to understand these in terms of the animals' particular biology. With philosophy of language, this focus is sometimes lost. Language itself does the referring in some views, rather than acting as a means. That's a crucial difference.

From an information theoretic perspective though, you can never ignore that data source. If a random text generator just happens to spit out a coherent English sentence, all it has really given you information about is the pseudo-random process underpinning the generation of the text. If it says: "Rome is burning," nothing about this ties back to Rome or fires, except wholly extrinsically.

Just a rambling thought about analysis. There is conventional meaning and intended meaning I guess.

There is a similar, maybe more influential issue where we tend to think that the unconscious processes undergirding empathy and language must work something like conscious inductive inference. Indeed, formal versions of induction are often called on to explain language, and "Bayesian Brain" theories certainly imply this sort of thing. But is it so? I don't see how any purely computational, formal inductive process can ever entail feeling or understanding (the old Hard Problem). Empathy seems to involve more than just induction, yet empathy also appears to be key to language acquisition and successful interspecies interactions.

There is no intentionality, emotion, and first-person experience in formal models of induction. I am not sure how there ever could be. Hence, using this as the starting point for philosophical analysis might be condemning our project to behaviorism (perhaps against our wishes). If the idea is that "what is empirically observable (and measurable)" is what fits in such models, then we might end up being forced to reevaluate this epistemic criteria.

I miffed that a bit. It was actually St. Augustine writing about St. Ambrose, who practiced silent reading. Augustine found it strange. — Leontiskos

Thanks.

That people used to often read aloud might also be a reason for the heavy preference of verse up until the modern era, the epic poem being for them what the novel is for us, and even scientific and political topics were covered with poetry. Another common hypothesis is that it is easier to remember text that is in rhyming meter and books were so expensive that memory was essential. Also, you can often do more emotionally and thematically with clever verse using less text, and when you have to kill a bunch of oxen to make a single book, economy is key. — Count Timothy von Icarus

Yes, and I think the recovery of reading aloud would be a good thing. Simpson talks about the development of poetry as part of the natural progression of culture's orientation towards a comprehensive end:

Further, methods and principles will be looked for everywhere, and those who, because of advancing community, have leisure from necessities through being served by others, will look for principles and methods in the things of leisure, particularly in counting and numbers, and in the regular motions of the heavens. Records of the past will be examined and ways to preserve memory fostered, in particular by forms and patterns of words that in their rhythmic features lend themselves to easy recollection. Poetry will develop, and those skilled at composing or remembering poems will be prized and honored. At some point, ways of recollection that do not rely on living memories will be invented, such as by marking shapes on long-lasting material objects: walls of caves, pieces of wood, cured animal skins, baked clay, or beaten metal. Pictures that are direct copies of visible objects will likely come first, but the need for more abstract shapes, such as to record numbers or the sounds of human speech, will be felt and find varying solutions. But throughout all will remain the mysterious riddles of the universe and whence it came and how it always remains, and where and what are the hidden things, the dreams and visions of the night, strange foretellings of the future, sudden intimations of events far away. The riddles of man himself will figure largely among these mysteries, the mysteries of love and peace, hatred and war, success and failure, advance and decay, birth and death. — Peter L. P. Simpson, Political Liberalism, 72

-

- Sounds good.

What happens if we change the designation to "The man over there who I think has champagne in his glass is happy"? That's where Kripke himself winds up: "The speaker intended to refer . . . to the man he thought had the champagne in his glass." Has the speaker still made a mistake in reference? I think we have to say no. The reference is now based on something the speaker thought, not something that is the case about Mr. Champagne. The speaker can point out Mr. Champagne to me, explain that the man is being designated according to a belief the speaker has about him, and we can both usefully talk about that man and no other. Whether or not Mr. Champagne really has champagne coudn't be relevant. — J

Building on what said, this essentially contravenes your earlier observation, 'I shall use the term "Glunk" to refer to the man that I call "Glunk".' Note that part of the problem here is that 'Glunk' is not a term that is serviceable for triangulation, because it has no common meaning between the two speakers. This is the same problem with your, "The man who I think..." Mere thoughts are not common between speakers, and therefore are not serviceable for triangulation. This is why your revision does not add anything - because your interlocutor cannot read your thoughts. The reference still succeeds or doesn't succeed on the basis of the claim that the man is holding a glass of champagne.

Has the speaker still made a mistake in reference? I think we have to say no. The reference is now based on something the speaker thought, not something that is the case about Mr. Champagne. — J

It is now based on something thought to be the case. There is no escaping the assertion of what is the case (and this is equally true on TPF, where it is unavoidable to give opinions). The revision is only useful insofar as it introduces an explicit margin of error. It is only useful insofar as it says, "The man over there who is holding a glass which appears to contain champagne." The idea here is that the interlocutor will be helped if the possibility of an erroneous appearance is pointed up. But apart from that, revising to thoughts is no help at all, given that thoughts are private. "That man over there of whom I am thinking," is not going to suffice for a common reference. We say, "The one I am looking at," or, "The one my right foot is pointing at," but not, "The one I am thinking of." So, "I am thinking of the one I am thinking of," is an infallible statement, but it won't be helpful when it comes to public reference.

My question to Banno focused on something a little different. If we say that reference, as a matter of fact, requires triangulation, then it would follow that whatever it is I'm doing, privately, is not an example of referring. That's one way of setting it out conceptually. The second way would be to say that the question is not a factual one at all. We have a term, "reference," and we're considering how best to use it in order to carve up the conceptual territory. So it might be that we want to reserve "reference" for the cases where triangulation is involved. In that case, we need another term to describe what I'm doing, privately. I was asking Banno which of these outlooks he favors -- hope that makes sense. — J

This plays well on my dithering between Davidson, Austin and Wittgenstein.

Referring as a speech act is public and communal, as are all speech acts. If we are to make sense of thinking about something to oneself, we might well do so as a back construction, a re-application of the public act to the equivocal "private" world.

One argument for this is that the "private" discussion might be made public - you can tell someone what you are thinking. The act of thinking about something to oneself is not inherently private in the radical sense; it can be translated back into the public domain. That's what distinguishes it from the kind of "private" experience Wittgenstein critiques in the private language argument. So it's not private in the way that the sensation "S" is for Wittgenstein.

In answer to your question, I don't see that we must either deny that a referring to something unvoiced is not a reference proper, because it well might be made public.

The private language argument shows the incoherence of a language that in principle cannot be shared. It remains that something – a reference – may be in fact unshared yet not unsharable.

So I do not think I am caught in the dilemma of having to choose to rule some references as not references.