Podcast: Play in new window | Download (Duration: 56:11 — 51.5MB)
Continuing from part one on “Meaning” (1957), “Utterer’s Meaning and Intentions” (1969), and “Logic and Conversation” (1975) with guest Steve Gimbell.
We tie Grice’s initial project about meaning into this apparently new project in “Logic and Conversation”: What are the rules that people tend to follow in conversation to actually be engaged in the cooperative enterprise of exchanging meanings? Grice identifies this general framework that we follow as the “Cooperative Principle” and identifies several maxims within that: Providing the right amount of information in an utterance (not too much, not too little) for the context, providing high-quality conversational input (meaning not false, unsupported, or irrelevant), and speaking clearly. When a speaker appears to violate one of these principles, then there are systematic ways in which we as listeners try to interpret the speaker so as to still think of them as acting according to the Cooperative Principle. For instance, if someone says something blatantly false, we might look at it as irony or metaphor or some other non-literal speech. If someone provides too little information to be helpful, we might charitably think that the maxim to provide enough information was in conflict with the maxim to only provide high-quality information, which includes only information that the speaker has good evidence for, or perhaps the speaker has been sworn to secrecy or is otherwise trying to maintain decorum by not revealing private business.
We attempt to relate Grice’s project to various activities in cognitive science (e.g. computer language modeling) and continental philosophy (e.g. the critique of writing from Plato through Barthes). Some of these connections relate to whether Grice can unproblematically restrict himself to talking about what a speaker intentionally means as opposed to what meanings (natural meanings!) may be exhibited by a statement, whether this means the literary meanings that may be extracted from it beyond what the speaker meant, or the unconscious motives of the speaker who thus betrays certain meanings unintentionally through the utterance. Wes and Mark will delve into this issue further in the subsequent Closereads treatment of “Logic and Conversation” (a three-part conversation to be released over the next weeks. Subscribe now at patreon.com/closereadsphilosophy).
Next episode: We’re interviewing Michael Tomasello about his new book, The Evolution of Agency.
Transcript
0:07 Mark: This is the Partially Examined Life, episode 325, part 2. We've been talking about Paul Grice with Professor Steve Gimbel.
0:15 Steve: Welcome back.
0:16 Mark: You should look a little more at the main concepts in the third paper from Grice that we had on the table, Logic and conversation from 96. I mean, you could see it as a completely different project, or you could see it as the culmination of his earlier project on meaning. Where do you actually stand, Steven? You presented it as if it were the culmination, the latter. But it seems like it's. As far as the scholarship goes, it might be sort of an independent launching of a thousand ships, just like the 57 paper was.
0:45 Steve: That's fair enough, but I think it really is. He takes a huge step in a new direction, but in order to answer the old question. So the original question that he's dealing with is, how do we understand each other? How do I get what you mean when you talk? And yet we do. Somehow we, despite the fact that we can't get inside of each other's heads, I understand your intentions most of the time. And what's weird is that when you look at how we talk, if you were to take a normal conversation and then just type it out, most of what you see does not follow in any way from what was previous.
And yet it's a smooth flowing conversation. And somehow when you look at responses, they make perfect sense if you know how to speak the language. But if you were just, say, a computer, this would seem like a complete non sequitur. And so what we're seeing in this third paper is sort of this ultimate answer to how do I understand what. What you mean when what you say isn't what you mean?
1:53 Seth: Okay, so in your email response, Steve, where you said you would come on and do the show, you made a mention of kind of like a general criticism that we can bring up later of the Gricean perspective, if you will. But you just raised a really interesting point. So we exist in the world of audio. That's what we do. That's what podcasting is. Even though I'm looking at you now on a screen, the video will disappear forever. Much to the chagrin of our adoring fans, I'm sure. But we frequently get asked for transcripts, and there's lots of reasons why, from SEO and all kinds of other perspectives, you get transcripts.
When you see a transcript of us having this conversation, I don't want to say it's literally unintelligible, but it requires a massive amount of editing to make any kind of sense of. And so I'm wondering if, when, you know, what did they call it? The linguistic turn in analytic philosophy, you know, and yes, we've talked about positivism versus, you know, versus some kind of, like, performative. But, like, did analytic philosophy in your perspective. And by the way, Mark and Dylan, you chime in here, did we just lose sight of the ancient Greek insight and the difference between the written and the spoken?
Because it seems like the whole concept of fixed meaning, timeless meaning, conventional meaning, whatever you want to call it, semantic meaning, all comes from some kind of ahistorical, a temporal, or at least no specific speaker versus audience. Kind of like, are we just talking about. Ultimately, are we talking about the difference between the written and the spoken? And that's what this whole Gricean move and the Austin move and all that stuff are revisiting. I don't know. I'm just curious about your perspective on that.
3:43 Steve: I'm working right now on a new project which is a sort of popular history of early analytic philosophy telling the stories. And you're exactly right, that early analytic philosophy really was striving to, in a certain sense, dehumanize philosophy, which seems really weird until you remember the context from which it's emerging, which is it's really being written in Berlin and Vienna between the world wars, when you have a rising tide of fascism, which is being bolstered by Nazi race theory, which purports to be scientific. And so the idea that we could, in a certain sense, let's take the human passion completely out of this, because human psychology is so easily corrupted.
Let's just get down to what we can know with absolute scientific certainty, or as close as you can get with an inductive method. That is science. Give us scientific knowledge, give us rigor in such a way that we can look at this nonsense that's coming to dominate Europe and go, that's garbage. And here's why it's nonsense. When we understand what meaning is, when we understand how inferences really work. Now what we can do is see what the nature of scientific evidence is and understand that this stuff is nonsense. So where analytic philosophy is coming from is intentionally dehumanizing thought in the hope that it gets us to a sort of truth that could ultimately save humanity.
But what happens to it is, as you know, was mentioned at the very beginning of the last episode, is it reads very dry. It seems very distant from lived experience. You know, you take something as simple as having a conversation, which everybody does all the time, and now suddenly you're using symbols for it, so it seems alienated from the thing it was ultimately trying to say. And I think you could see Grice as this sort of figure who's trying to Rehumanize without losing the methodology?
5:51 Mark: And was he pretty important in establishing the kind of thing that cognitive science would be doing in how do we understand human understanding, like of a story, for instance, from utterances? Well, let's try to program a computer to do those same things. And it's very difficult, as we would see by looking at this logic in conversation paper, because there are all these background assumptions, there are all these contextual things. Is there any way to do that with a computer unless you actually got, or even if you actually got an actual robot with a body to walk around and interact with things and gain some sort of embodied knowledge?
Right. This was Dreyfus's Heidegger kind of critique that this is what's essentially wrong with a series of sentences. But that seems to be what Grice is at least trying to do is when I understand as an audience, as a hearer, understand a speaker, what exactly is going on? Can we somehow really formalize this so that we are including only those cases that we would count as authentic communication and ruling out everything else?
6:50 Dylan: I just want to disabuse one thing, right, Is he is emphatic about there being meaning. But one really important point is it's not always clear what the meaning is. And so part of what he does is he provides a framework for it. So the fact is you can be wrong about what the meaning was in a given interaction, even if you still can analyze why you were wrong within the framework of the maxims and stuff like that. And the computer question, I didn't read very much, hardly any secondary literature on this, but I did see one comment made which was, grice is super important for people who are doing AI and language recognition or ChatGPT type stuff.
And I had no idea what that meant until I read language and conversation. Then I thought, okay, now I get it. What you're doing here is that if you want to write a. I don't want to use the word ordinary language because that's overloaded in philosophy, but a ordinary language type analysis and machine learning within a computer is that you would embody it in terms of Grice's maxims. That would be one of your first. You would develop an AI that was parameterized in terms of Grice's maxims, and then you would train it according to examples so it would begin to recognize and also produce things that were ironic or ambiguous and have multiple meanings.
And then you would have multiple possibilities of interpreting them. And so you would go down that path. Inevitably, there's going to be Some kind of probabilistic kind of interpolation kind of thing going on. But that's what you would do to make it so it worked like a human being. And one point is that human beings don't get these interpretations right all the time either. We're often wrong about it. And I think Grice in his third level cases, he goes through a lot of things, well, you could be wrong because you violated this maxim. You could be wrong or you could be misinterpreting this, or you could be communicating in such a way that, that you're violating on purpose as maxim to communicate this on purpose.
Both those things are going on.
9:12 Seth: So I think you explained to Steve our various connections. But I don't know if you remember, Mark, there was an AI company here in town. I'm trying to remember the name, I'm blanking on it. But a number of the graduate students who went to school with us at the same time worked at this. And part of what they were doing was trying to codify natural language into a formal framework, et cetera, et cetera. It was another one of those, like almost too early for its own good. Like maybe it was like 10, 15 years too early. Trying, trying to. Do you remember Dragon naturally speaking?
So they tried to do something and you somehow it just never. Now YouTube just does what Dragon was trying to accomplish on every single video, a billion videos a minute or whatever it is that they do. So the point being that I don't think, Dylan, you might say, like Grice is influential to some extent in this area, but the actual mechanics, it's not that Grice is influential because of phi or psi, X, Y, whatever, you know, it's that, oh, we can't handle Lacan or Derrida. We can't figure out how to turn Lacan or Derrida into artificial intelligence and large language learning models and all that.
But we can figure out Grice and these kinds of guys.
10:36 Dylan: So I don't know. I mean, I'm speculating on a comment that I saw that Grice was important for this. And I was speculating on the way in which it would be true. And the reason that I think it would be true is that Grice formulates in this paper principles and ways of interpreting meaning and goes on various axes for how to do that. I mean, he gives a self acknowledged Kantian structure for how to understand meaning in conversation. And I don't think that Derrida or Lacan have any ambition about doing anything of the sort. At best, they're Kind of naysaying about meaning.
11:16 Mark: Let's not get pulled into Derrida right
11:19 Dylan: now, but I understand. Okay. But my only point is that Grice seems to be. I would see as particularly potent in this. Putting aside whether or not you could pull something out of Lacan or Derrida or someone else in this respect, because Grice structures it explicitly this way.
11:38 Seth: Before I go down, I'd just say I was not intending to take us on a divergent path. I was talking about a radically different interpretation of what meaning is or how meaning works and how that would not be computational or something that would be easily consumed by people who are trying to computationalize. That being said, I don't think Grice is a major factor in anything that you're seeing going on with AI and all that kind of stuff now, anyway. I think all those things are radically different than what analytic philosophers of the language in the 80s and 90s were doing.
12:11 Steve: If you look at what Derrida is doing is part of what he is saying is that everything is infinitely ambiguous, that we can unpack the meanings of utterances in all sorts of ways if we use different, you know, sort of bases of interpretation. We can, you know, come up with your favorite dichotomy, and we could pull it apart this way, we could pull it apart that way. Whereas what Grice is doing, and I think this is where Dylan is right, is he's saying timeout. When we look at human interactions, when you look at the transcript of one of your podcasts, that seems just completely unintelligible, and yet when you listen to it, it made complete sense to a consciousness listening.
What that means is that on the one hand, it is infinitely complex, which does seem like it rules out the possibility of automation altogether. But then Grice has the insight, and that's what this paper, I think, really is crucial in doing in the history of thought, that maybe there are certain simple principles that we can use to structure human conversation. But what's even more important is that when we have these unusual occurrences, which are actually quite common, of conversational contributions that seem on the surface meaningless, but in the context make complete sense, how is it that we understand them?
A computer would seem to have a very difficult time because this seems to be an utter non sequitur. But to a human listener, Ah, I see what you're doing. So what is it? And here's where the AI comes in. How can I pull apart these discussions so that. Oh, I see how you got from this sentence to this sentence. So Speaker A says this Speaker B responds with this, which seems utterly nonsensical, but Speaker A then goes, exactly. And so what is that? How is it that we can understand each other even when we're speaking in ways that don't seem connected, but to us are perfectly smooth, ordinary conversations?
And that's what this paper is about.
14:27 Seth: Right. I don't know if this is a distraction or. But I know enough about ChatGPT and some of these other technologies that they don't ask you who your audience is for who you're building it for. You say, like, give me a three paragraph summary of Herman Melville's whatever. You know, it doesn't say like for whom.
14:45 Mark: You can, you can say, yes, aim it at a third grade level.
14:48 Seth: You can add that, but you don't have to.
14:50 Dylan: No, but also where it gets aimed at. And the audience is also going to be incorporated into how it was trained. Right. So we're going to quickly get down. Probably would be a much more interesting. It would be more interesting to have this conversation in the context of Talking about how AIs are and are not revolutionary. And that might be a separate conversation. It made complete sense to me when I read logic and conversation why someone pointed to Grice as being part of that. Now, I don't know if that's a fact at all or not. I have no idea if the folks who wrote ChatGPT are Grice fans.
15:28 Mark: Let me follow this. A few of the past threads I want to knit together before we actually get into the logic conversation.
15:34 Dylan: Mark is the tapestry guy. How are we going to make a tapestry out of the threads?
15:39 Mark: So Seth was asked at the beginning, you know, how have we forgotten that speech should be primary and writing should be secondary? And then we brought up these postmodernists and Lacan and it seems like those folks. It is almost a reductio against meaning as the written word. Because if you take the written word as the primary source of meaning, that even when human beings think that they mean something, and I'm sort of, again kind of reviving this pseudo behaviorist attack, well, it's because we are latching on to public meanings. We're unable to think unless we are thinking in these public meanings that have been stitched by society.
But then these postmodernists, they give us sort of a reductio against that actually meaning anything at all, that it means a multiplicity of things. It's merely a tissue of citations. I believe was a quote from Barthes that we talked about or the death of the author. We had the whole episode on authorial intent. That there's a whole group of these folks that say, you know, figure out, what does this work of literature actually mean? Is not a matter of asking, well, what did you intend when you wrote that? And Grice actually acknowledges that at one point, just with.
Well, if you ask a philosopher, what did you mean by this passage in your paper? Well, you should expect them not to just explain, well, I meant to write this. And then they read the sentence aloud, you know, no, they give a re. Explanation. They're actually doing something creative. And maybe that is what we're actually getting into in this logic and conversation paper is that in addition to the sort of so called literal meaning, the conventional meanings associated with this, there's all these things that actually turn out to be more symptoms.
We're actually getting back to the natural meaning in the first place. What did you mean when you said, I look nice? Well, if you ask me that, are you feeling okay? Are you depressed? What's wrong? If you ask me that question, I'm going to make something up. Like, there's lots that I can find in my life to be upset about, but I don't know if that actually is the thing that caused me to be upset right now. And I think the same thing goes with utterances. And meaning is always sort of larger than what any human being intended to put into an utterance.
17:43 Dylan: Let's be careful here. Your responding to that question is different than what the question meant. And what we're talking about was the utterance of the question, not your response. Those are two different things to respond to about what the meaning is. So, you know, asking the question, are you depressed? We're talking about the meaning of that now. Your response to that question is a completely different issue.
18:09 Mark: I guess I was just treating. Any utterance can be given the same treatment that you could give the same. Why do you ask me if I'm depressed? What am I doing that is making you feel like you're asking that? You know, you need to ask that question. And I could interrogate you and you have to make up some bullshit about,
18:24 Dylan: you know, how many of you came to mind. Vizzini's speech with the Dread Pirate Roberts and the Princess Bride over the Iocaine powder question. Like that whole back and forth that immediately came to mind when I was.
18:39 Mark: I have to know that you understood that I meant that. I knew that you meant that you.
18:44 Dylan: Exactly.
18:45 Steve: I will admit that I thought of exactly that scene as the perfect way of understanding Grice.
18:51 Mark: Yes.
18:52 Seth: And this is where it turns into a thing where we're all white men of a certain age with a certain shared. This is where people complain, Steve. But they're like, you need to get more diversity on your podcast because those Princess Bride references can't last forever.
19:11 Dylan: Princess Bride is timeless.
19:14 Seth: We'll find out. We'll find out when I finally get a chance to watch it with Sky. I have a five year old. I've been waiting to introduce this. She's not ready for it yet, I don't think.
19:23 Dylan: But because you brought it up, Mark, I wanted to just clarify, is this question about me meaning in art or meaning authorial intent? Is that out of scope here? Because I felt like when we were clarifying what Grice means by meaning, we were restricting it to utterances and not the meaning of life, for instance. And it strikes me that asking a question about, well, what does the art mean? Maybe it's not quite the same as understanding what's the author intent would be one thing, but what does the novel mean? It's more like a what is the meaning of life Question. And I don't think that Grice is tackling that question and that his discussion of meaning doesn't apply.
20:14 Steve: Yeah, I think you're one step back. You're exactly right. Is that there is what did the author mean in saying this as opposed to what does the novel mean? And those are two different meanings of meaning. And the first one can, I think, be given a Gricean treatment, but the second one not.
20:30 Mark: I just thought that again, one of the things that grace supposedly innovated or fought back against is the verbal is primary. And then the timeless meanings of sentences are all derived from what people actually meant by those phrases on particular occasions. They caught on, they ossified. And so I was generalizing from that to what is a book but a series of sentences where the author is no longer there speaking to you directly, it has become this disembodied voice, Right? This is the thing that Foucault and Barthes then were worried about. What is this weird audience lists or uber audience?
Am I writing for posterity but author lists, essentially, once it's out on a page, is language itself somehow autonomous so that once it is codified, once it is written down, it is autonomous in some way? Right. It has a life of its own based on how meanings change over time of a phrase within language, you know, so, yeah, we could talk about the meaning of Moby Dick now versus when it came out or whatever.
21:32 Seth: This is part of my issue. Again, started off the conversation by saying, I think Grice should have circumscribed the discourse that he thought he was or the universe of problems that he thought he was trying to solve for. I feel very strongly about this. Numerous times over the last 13 years on this podcast, I have talked about
21:50 Mark: how you almost said, this stupid fucking pod. You're.
21:54 Dylan: I.
21:54 Mark: Is that what you're doing, people?
21:56 Seth: You're right. I almost did say this fucking podcast. But no, you know, I've talked about how intent matters. Not at all. It's like, I do not give a. If I take a book and I use it as a doorstop, then that's what it is for me. It's like whether the author intended me to read it or not. This account is seriously flawed with respect to the unintended audience or the unknown audience in a variety of ways. But the point is, he's thinking, whether he thinks about it or not, in terms of Austinian acts, performances, utterances, things that are not timeless.
When you write something down and if there's a possibility that somebody 3,000 years or 200 years or 100 years or 50 years later is going to read it, there can be literally no intentional interaction between you as the author and the writer and the audience. It's just not possible. And we've come across this problem, I think, in much more sophisticated analyses when we talk about art and beauty and aesthetics and trying to understand artist intention and artist productions in a certain time frame or with a certain audience. And then, like, why is something timeless?
Like, why do Platonic dialogues still have meaning for us? But why does the pieta? Answering that question, I think, goes a lot further than Grice's analysis to help us understand meaning. But it breaks the convention that he's put in place because what he's conceiving is a certain linguistic convention in a certain historical epoch. So, you know, when I was reading him, I was thinking about Tomasello. I was thinking about language acquisition, language acquisition versus language use. We just went through, Steve, this thing where we're talking about, how does language become acquired, and how do we come to learn what words mean, how we use them?
And in a certain sense, Grice has been vindicated, if you believe this guy Tomasello and his account, which is language is acquired socially, like it's about shared attention. Tomasello's point was the way that children learn language is they come to be aware that you are trying to put a thought into their head, you are trying to get them to do something, and then they in turn say, I want you to pay attention to me, it's all about understanding that there are internal mental states in this social interaction. And if you cannot picture the being across from you has a mental state, internal, that you can't see or read off their face, you will never acquire language.
That's basically Tomasello's point. And Grice had that insight in 1957, even though it wasn't cashed out quite that way. So in a certain way, he's been vindicated. I just think trying to apply a certain set of rules to the way that language is used, it's just too limiting for the variety and the expansiveness in the way in which we use language.
24:52 Steve: And I think that would be absolutely right if Grice said that we were constrained by those rules. And I think you have to understand, you know, for Grice, the rules are flexible and viable. So he's not, you know, a Wittgensteinian in the sense that this is a game, there are rules, and you follow the rules, and if you don't follow those rules, you're no longer speaking the language. What's. I think the most insightful part of this piece is the idea of an implicature. That is that we will often, within the confines of a conversation, intentionally violate the rules in such a way that we're still very much playing the game.
So if you think about basketball, basketball is a really weird sport in that if you're losing and there's. It's a close game, two minutes left on the clock, what are you supposed to do? You're supposed to foul, you're supposed to break the rules, but those rules are constitutive of the game. If you break the rules, you're not playing the game. But in basketball, if you're losing a close game with less than two minutes and you're not fouling, then you're not playing the game. That is part of playing the game is breaking the rules. And what Grice does is set up a general framework of rules that generally tend to structure how we converse with each other.
But what he says is those rules are meant to be broken, and when we break them, how we break them contextually will then often convey our actual meaning to the other conversant. So we're not limited by those rules, but those rules set up at least the beginning of how we're going to play this verbal game of Calvin Ball.
26:37 Seth: Thanks.
26:37 Mark: Thanks.
26:37 Steve: Which is another old reference. Sorry.
26:40 Mark: Which we use regularly. Thank you for transitioning us back to the text. Yes, we should finally characterize this set of rules, the cooperative principle. CP so we assume when we're having a conversation with someone that the other person is having a conversation with us. We assume if I say something to you and then you say something that I'm not quite. Quite sure how that follows. Then I try to figure out, okay, you're not literally answering my question, but we just assume you're not just giving a non sequitur. So that's the sort of rules each person is expected.
27:13 Dylan: What.
27:13 Mark: Be relevant, avoid ambiguity, obscurity, don't say too much, and don't lie or say something that you just have no basis for. So it's. It's quantity, quality, relation, and manner. Are those four rules in. Not in the order that I describe them.
27:29 Seth: Right.
27:30 Steve: Those are the maxims. And you know, and we all know people who violate them.
27:33 Seth: Right. This is kind of where, I mean, I don't know how fast the train was moving when I jumped off, but this was the point at which it happened. I mean, I can think of nobody, absolutely nobody, who follows them. The maxim of quantity. Like, either the people don't speak enough or they just. They provide too much. They provide absolutely too much.
27:57 Dylan: I completely disagree.
27:59 Seth: Okay.
28:00 Dylan: I think that that maximum quantity, both in terms of it just being generally followed and just ordinary everyday conversations like going to the grocery store. Most of the time, I go to the grocery store and the response to my question is the right amount of quantity. I'm not getting too much detail about the steak, and I'm not getting not enough detail about the steak. You know, and same thing with, like, going to the restaurant. Right. And my daily interactions with my spouse or by people at work. It's very easy for me to identify someone who is speaking with too little quantity or too much quantity. I can give you a list of them, and it's just extraordinarily easy. And the fact that the idea of quantity as being a fundamental characteristic of the way in which we engage in conversations, that the violation of which leads to understanding of meaning and not meaning, I think is spot on.
28:54 Seth: Okay, so without trying to set myself up, without trying to contradict or demean your experience. My experience is quite different. I am constantly engaged by people who usually it's in the excess quantity. And maybe it's just who I am
29:14 Dylan: and how I am, that says a lot about you.
29:17 Seth: It says, but here's the thing.
29:18 Mark: You're a very good listener, Seth.
29:20 Seth: I am a great listener. And my wife and I together, we have a superpower. Like, if we are anywhere on earth and I'M not this. That's not an exaggeration. Every place we've traveled, like, we're at lunch or breakfast, and we're like, how's it going? Whoever it is is going to open up and give us their life story, like, that's just who we are. That's the vibe we give off. So we as an audience have a very. It's particular. People say, like, oh, this looks like a couple of people that I can talk to. And that is just our experience. That's just how it goes with us. So that maxim gets violated. And that's the thing, is we look at each other often and we're like, okay, I guess this is where it's going today. This is what's about to happen. And I mean example after example of example, but I'm kind of like, okay, so that maxim gets violated for me all the time.
30:15 Steve: Maybe see the maxims not as descriptive, but as normative.
30:19 Seth: Mm.
30:20 Steve: So, yeah, those people are violating it. Aren't they obnoxious?
30:24 Seth: No. The way we see it is, like, this is clearly somebody who has something that they're having a moment.
30:30 Mark: Well, then it sounds like that speech.
30:32 Seth: We are there for that moment, and that's okay. Like, that's my purpose in life, is to help them transition out of this.
30:39 Mark: So it's not violating. You're describing a language game in which you are properly the receiver of as much as they want to unload on you. So there is no room in that for there being your utterances are driving
30:52 Dylan: their quantity unless they follow you to the hotel.
30:56 Seth: Listen, the Gricean account does not account for the differences in tolerance between or the difference in, like, aura colors or whatever it is that you want to distinguish between me and Dylan. The Gricean account is. I will give you just as much information as is necessary to convey this meaning and nothing more. It doesn't matter. You're saying that my whatever vibe I'm giving off is the context. Like, it's.
31:21 Steve: Yes.
31:22 Mark: If you're checking out.
31:23 Seth: Oh, interesting. If you're.
31:25 Mark: If you're checking out.
31:26 Seth: Enriching the Gricean account. Because there's no way that anything he wrote talks about the difference between me and Dylan when we're at.
31:33 Dylan: No, it's not true.
31:35 Steve: The next paper.
31:38 Dylan: Well, we should read all the maxims because we don't have that much more time that we're going to talk about this.
31:44 Mark: But I think this is what maybe Seth is trying to argue against, is these maxims are what make conversation understandable in the first place. And so, in general. In general. So to the extent that somebody is violating this. And I think it's completely contextual that, you know, it is very much. If you're on the stand in a courtroom, how much information you're going to give is very different than if you're having a casual conversation, which is very different if you're in line at the grocery store. And I so hate if the checkout person and the person in front of me at the grocery store decide to have an actual conversation.
I'm very keenly. There has to be. As you were saying, Steve? These are rules that are made to be broken because then those become clues for the listener of how to interpret the sentence. So the one about quality. So don't say things that are not true, and don't say things for which you don't have evidence. But saying things that are not true. All of art, you're just talking about metaphor. We're talking about. There are so many reasons why you would want to say something that is not literally true and that would be entirely appropriate given the context.
But it feeds to the interpretive strategies, the heuristics that the listener will use to figure out, oh, I'm in a poetry reading. I'm in the audience of a poetry reading, and so I'm gonna understand what this person is laying down with the background of my ordinary conversation. So I understand how the poetry reading differs from that. And I guess I will carry those norms over so that it is possible that even within the poetry setting, you could be like, wait, why are they just saying the same word over and over again for 20 minutes? This is not avant garde in a way that I appreciate.
33:29 Steve: And in violating the maxims, we're often actually communicating, whereas it seems like if we take these rules as descriptive. By violating the rules, that should eliminate even the possibility of. Of my conveying meaning. But it's often in violating the rules how we violate the rules that we're able to convey. So one of the examples he gives is, you know, let's say the four of us are talking about, you know, that other host who. Who you had mentioned earlier. And, you know, did you see him with that other woman? And next thing you know, he pops up on Zoom.
And I'll say something like, boy, it's been awfully hot here in the East. Which seems to have absolutely nothing to do with the conversation. It violates the maxim be relevant. Why would Steve do that? Maybe Steve's being a jerk and doesn't want to actually know. Steve's been having a reasonable conversation with us. Why would he violate. Oh, he's here. Right. So the idea is that by violating the rule in a way that you recognize that I, who am usually perfectly comfortable with in a normal conversation, obeying all the rules, the fact that I just violated the rules, big red flag.
Oh, this means something else, by the way.
34:45 Seth: I don't disagree with that point. I don't disagree with the point that there are conventions. Maybe this is a semantic distinction where I don't believe that people communicate according to rules having to do with those four maxims. At least with quantity, but maybe with the others, I don't want to spend time on it. But there are definitely conventions that if somebody gives you more information than you're expecting, you're like, okay, well that's one thing. Or if they give you less, if they give you less information like you say you're checking out.
And these people, they're. They're trained, right? The people who check you out are trained to say, oh, hi, how's your day going? And you go, great. Conversation stops, right? And it's deep. You go, great. How's yours? At that point, right. There's a convention that's like, fine, thanks for asking. Right? Many places it'd just be like, fine, thanks for asking. Again, in my case, it's likely to lead to a 10 minute personal confession. I don't know why that is. It just happens to be. But there's no doubt that they're breaking convention. I'm just not sure that they're violating a rule of quantity.
It sounds semantic, but you could ask,
36:01 Mark: what are they communicating? Are they communicating? You know, I'm actually an assembly line here. And the fact that you're asking me
36:08 Seth: to be like, communicating that they're a human being and they desperate for connection,
36:13 Mark: which seems like a symptom rather than an intentional non natural meaning. It's not like my telling you the life story that one of the meanings of that is that I'm trying to tell you I am reaching out for some human. But you could read that as a symptom of what they're doing. Right? So this is where I'm wondering about when Grice admits that the illocutionary, the perlocutionary, whatever it is that is. We're now not talking about what you. You literally meant anymore. We open things up to figurative. It seems like we're opening a very, very big bag of ways of attributing intentions to people that they didn't know that they had Right. Freud and Lacan can now enter the
36:51 Steve: picture because we make inferences, we have to reason to it. And this is where, you know, someone like Freud or Lacan, who's picking up on Freud is going to say, well, this is indicative of something in the subconscious that maybe we can tease out. Whereas this is different in that I'm trying to tell you something in a way that doesn't use the usual words of telling you, but yet I'm conveying it.
37:13 Dylan: So let's be a little bit careful about this because it may be that what Grice is talking about opens up to this sort of subconscious communication, but he frames this in terms of cooperative principle and where he is framing this, first of all, this conversation. So I want to also put a little bit of an asterisk again, about this whole concept of novels meaning something and him not talking about this. Even the abstract notion of there being a conversation in philosophy over years and stuff like that. I don't think that is what he means. I'm not a grace expert.
But it's a cooperative activity between two entities that's going back and forth, actually going back and forth. Not like hypothetically, because there is not a response that the book makes to me, okay? It's always, at best, a manufactured set of conversations between me and the book and me. What I think the book said and what I said back to the book and stuff like that. He's talking about meaning that comes out of actual conversations between interlocutors that are going back and forth. And I think that that cooperative principle, he at least intends to be scooping it towards understanding what you meant.
Now, there might be an extension of. There's the thing that you didn't even know you meant, but you actually meant this unconscious activity, right? That somebody like Lacan or Freud is going to be super interested in your subconscious meaning. What they're going to want to say. That's your true meaning, right? But the thing that you didn't know that you meant. I don't think Grice is talking about that at all. I think you might get open to it and might want to go down that road to thinking about that other piece of meaning. But he's not talking about that kind of meaning.
He's talking about just the meaning about us understanding what you said.
39:07 Steve: And it brings it back to the original question that he was asking, which is, how do I understand what you're saying when you don't say exactly what you're saying? And yet I am perfectly capable of understanding what you're Meaning I can't get inside your head, but yet I understand what you're meaning even when you don't say it. It's a weird phenomenon. And this is how he explains it.
39:31 Dylan: Part of the challenge here is the word meaning is overloaded in the way we talked about earlier about, you know, the meaning of life and stuff like that. Meaning is an overloaded term. Maybe in some ways it's just say, what are you saying?
39:47 Mark: Sure, I was again, bringing back the Wittgenstein thing. Some of these descriptions of unconscious motives in something are to say, you might not know it, but this conversation that is a cooperative enterprise is some kind of conventional language game. It is a dance. Maybe it's a dance of seduction. And so, you know, when you said, you know, you told this story, sad story, to your date about your father's illness or whatever, you were trying to produce an effect, or at least this is part of the dance. This is.
40:16 Steve: Look how vulnerable I am.
40:18 Mark: Is it part of the meaning of the story that you told according to the 1957 paper? Clearly not. It is merely an indirect effect. Maybe I even intended in telling that story. But it's still not the literal meaning. Once we open up, I'm not sure by the time we get to the 1975 paper, there are definitely something. Some of the examples that he give that are, you know, I told you this story, but what I really mean is, I want to go home with you.
40:46 Dylan: Yes. I think that that's right to me, that that's the road he's gone down in the 75 paper. And it makes me think about comedy, which is why when I heard that Steve was gonna come on and that he wrote this book about comedy and that Grice was a big deal in that book, it made sense to me because the example you gave Mark seems to me perfect example. When you do that wrong, we recognize it as audience members because it ends up being funny because of that. So Rom com has a scene in which Joe is trying to seduce the woman, and he's having that conversation that has those alternate meanings to it, and he just does it wrong.
So he manifestly doesn't communicate in a way that is successful, but that ends up being hilarious. And I submit that one of the reasons it's hilarious is because of Grice's analysis about the way in which conversation works, that we recognize when it is working and not working. And sometimes the manifestation of that lack of working is funny. I mean, things are funny because they end up being misappropriately aligned. But in a way that's close enough to be recognized. Now you're getting the whole interpretation of why things are funny, but that's at least one of the pieces.
42:13 Seth: So I agree with you, Dylan, and the notion that meaning is a relationship. Meaning involves the relationship between the speaker and the audience and that there are intentions with respect to the speaker and there are expectations. A word that is not anywhere in any of these documents, but it is crucial. Expectations on the part of the audience. And the mismatch between intentions and expectations is what generates tragedy, comedy. There's a variety of things with respect to the way that Grice tries to formalize all this or make it rule based or whatever.
I think that's the level of the insight is that there's an agreed upon set of conventions that are. And you can violate those conventions in a number of different ways. Both sides can violate those conventions and that can lead to comedy, tragedy, or just nothing, you know, misunderstanding. All of that I agree with. I'm not sure that the Grice and Mac, I certainly am not going to spend any time worrying about the last three pages of whatever it was. Logic and utterance, or logic and meaning, whatever, you know, with all the utterance, meaning.
But when you think about there's a lot of great comedy, that's exactly the point you're making. It's building up the expectation that the move is going to go here and instead it goes here. Right.
43:42 Mark: As we wrap this up, there are so many great examples here from any of the papers. And let's just kind of put forth one each to treat, however briefly or not that we want. What's a particularly noteworthy thing that you remember from these papers that we have not yet picked up on?
44:02 Steve: The example I adored the most was from the second paper, which we didn't talk too much about. But he starts with the utterance, if I shall be helping the grass to grow, I shall have no time for reading. Which we later come to learn means that, yeah, if I'm dead, I don't have to worry about all the awful things that happen in the world. But we first try to understand literally. And so here we are in 1968, he's giving a talk at Oberlin where he initially starts by saying, well, what does the word grass mean? He says, well, it's, it's. It's lawn material. Or he says, it could mean marijuana, which in 1968 at a college would have absolutely killed the room with a very upper class British accent. It would not have been what was expected according to the maxims of conversation.
44:57 Mark: He makes a big deal in that paper about the difference between that we're only talking about non conventional conversational, not conventional.
45:06 Steve: Yes.
45:06 Mark: Okay, so that there are plenty of things that just have conventional multiple meanings, but we sort of already understand that. It's. How do we use language in atypical ways, like that particular artistic turn of phrase?
45:20 Steve: How does linguistic improv work? And it does.
45:24 Mark: I remember the one. This is also from the second paper, where it's, an American soldier is captured by the Italians and wants to convince them that he's a German officer, but he only knows one sentence in German,
45:36 Steve: which is not, I am a German officer.
45:38 Mark: I don't remember what it was. It's like the long grass is very, you know, something from a poem or something. So the question is, does that count as him meaning, hey there, I'm a German officer, or does he mean the literal content of the German words that presumably the Italians do not know? And it sort of comes down to, well, what else is he doing? Like, is he trying to give them evidence? Again, that sounds like natural meaning. If he just is sputtering in something that sounds like German, you wouldn't have to know any German, and hopefully the Italians would be dumb enough to think that was real German. There seems to be a difference here between giving evidence in general. Like, here's another example that I'm sorry if I'm stealing this from somebody else, but the difference between I show you a picture of your wife having an affair with somebody versus I draw you a picture.
46:28 Steve: Right.
46:29 Mark: Both of them are giving you the information. Both of them are trying to get you to think this sentence. Yes, there's an affair going on, but simply giving the evidence is different than me drawing a picture where you could ask, like, wait, are you. Are you trying to tell me that this is what's happening? Are you just doodling? Are you trying to do a piece of fan art about my wife? What exactly is going on here?
46:51 Steve: So, right.
46:51 Mark: Similarly with the soldier example. Well, you could, if you knew the literal meaning of the German sentence of, you know, the grass is green, you could be pointing at the grass and like, all right, but if you're not doing that, then probably you either correctly are communicating to them, hey, I'm a German officer, even though neither of you knows the German word for that. Or you're giving just evidence and you're not communicating anything at all. You're just trying to put on an act and make them see you like, you're a measle. That is a symptom. Of being a German officer. Did you have a way of disambiguating those two options, Steve with that particular
47:25 Steve: example, I don't think I could add anything to that at all.
47:29 Mark: All right. Are you too worn down or do you have something?
47:32 Seth: SETH well, in the last paper, this is the thing is I bring up the point that if you narrow the scope of discourse and say, look, I'm not accounting for cases of this or cases of that, because there's examples of torture and bribery, and I feel like to try to have a theory of meaning that encompasses those types of use cases, intentional deception, persuasion. I don't think a theory of meaning or language use needs, if you were just like saying, I'm trying to explain how we mean things when we're trying to communicate to others, like if you could even figure that out without having to bring in all of the non standard cases, that would be something that could be an accomplishment, particularly in the context of trying to solve some kind of propositional discourse around philosophy and logic and science and whatever, that would be an accomplishment.
And part of me thinks you could section it off and you could actually come up with a reasonable account for how this particular thing happens. And then part of me thinks, like trying to section it off like that is exactly the wrong thing, because it is a view of language that completely misunderstands how language works and how we use it in this variety, we can go from satire to farce to persuasion to all these things, even within the context of one or two paragraphs, right? And so Grice is the epitome of what analytic philosophy of language was to me anyway, I think, and to many others.
And I asked myself where that project is now, and it kind of, you know, that'll answer most of your questions.
49:26 Mark: This is the question I asked initially, which now I think is comprehensible, is why does he want a theory that looks like this? And what I mean by looks like this is a set of necessary and sufficient conditions that will include all legitimate cases of a speaker means something and the hearer gets that meaning, but will rule out every other kind of thing. So that when we're talking about all these, well, if I'm torturing you and I want you to give me the information, so this is a, this is a command case. So it's not exactly the same as the informational cases, but we've said how they're related.
I want you to give me the information, and I want you to give me the information because, you know, I have the intention of getting that information out of you, but the fact that I'm using thumb screws on you. It's not that my use of the thumb screw means please give me the information, at least according to this analysis. It's fun to look at those examples, but if that is just feeding into Wait a second, let's modify condition 3 or add condition 5 to the list of exact conditions. One could wonder of like, what is the point of that project exactly.
As opposed to we want to have, I think you were asking for, Seth, a more general, humanistic understanding of what it is when we understand each other, as opposed to something that you could program in a computer, which is what I think this precision is shooting. Steve, do you have any insight on this?
50:49 Steve: Well, I will say that the thumb screw example is why I'm no longer allowed to have a lab section when I teach philosophy of Dylan, did you
50:56 Mark: have a final example or thought to shut us down here?
51:01 Dylan: I think that the last paper really goes actually a lot further, and I basically disagree with Seth on this is I think that it's a lot more fluid and open, but while being a structured way to understand how conversation works and provides some pretty interesting mechanics for how to do that. And I think there's a reason that he calls them maxims as opposed to rules or laws and stuff like that. And so I think the straight jacket of quashing the human spirit that I hear emanating from Seth regarding the activity of conversation, I think I just don't think he's guilty of it.
What I don't know about, I do find, like his categorization of things like, well, the beginning of how is irony working? How is ambiguity working? And what is happening in a conversation that allows me to detect that happening and get meaning out of it. It was pretty persuasive. Even, like, given that I have there's like a paragraph on half a paragraph in each of those, I feel like I'd want to road test it more. You know, I'd rather poke at it more to see if there's just kind of a happy path that this is going through, which is maybe what if I'm more generous to stuff.
What he's saying is that this seems almost like a just so account and that it's not quite living up to the full dynamics of a conversation. And for him, I'm interpreting it as just leaving out too many things. I found it much more comprehensive than I expected it to be.
52:38 Mark: I should ask Steve, as being more familiar with Grice's corpus was like the 1982 paper, Meaning Revisited. Did he? I think we're reading as if the 1975 paper is almost like later Wittgenstein to early Wittgenstein, that early Wittgenstein, he was trying to do this very precise thing, you know, that was in vogue in his time, but by the time he gets to the investigations, he's opened up and created something else. And it's a much more humanistic project. And I'm seeing the same thing. But it could just be because the 1975 paper, like the 57 paper, is short, that he's setting up a framework and that maybe He Then, in 1976 and 77, through the end of his life, did the equivalent of the Utterer's Meaning 1969 paper, to the 1957 paper.
And we do get, you know, analytic philosophy up the wazoo, precision about this framework that feels very open and humanistic and helpful in the way that Seth is asking for in the logic and conversation paper.
53:39 Steve: Your wazoo is safe. That this 75 paper really is sort of the pinnacle. And I think that the Wittgenstein analogy is probably pretty apt in that. Yeah, what he really is doing here is setting out a fascinating project that puts pragmatics at the center of understanding language. Before that, you've got syntax, you've got semantics. And he's really the one who launches this idea along with Austin. I think you see it reach its pinnacle in the 75 paper, that this is about the pragmatic elements, it's about the context. And when you add context, then you get meaning.
54:21 Mark: Well, thank you so much, Steve, for joining us.
54:24 Seth: Yes, thank you, Steve. Especially on such a short notice.
54:28 Steve: Oh, thank you, guys. Always happy to talk about Grice.
54:30 Mark: All right, next time, we're supposed to interview Michael Tomasello about his new book, the Evolution of Agency. Also, Wes and I recorded a part three on this episode where we begin reading line by line through the latter article, Logic and Conversation. So that will be available to all partially examined life citizens who support us via one of the methods@partiallyexaminedlife.com support. However, one hour was not enough for us to get through that article. So that Close Read is actually a three part Close Reads series. And there is a different place if you want to hear parts two and three of that series.
Patreon.com Close Reads Philosophy, right? Pel and Close Reads are different projects, but do some things overlapping. We make some of the Close Reads available to PEL supporters, but we're putting in extra work. And so it has its own place where you can support it and get all of it. Thanks to everybody who supports us through either or both of those channels. You could do things that are free as well. You could follow us on Facebook. You could follow us on X, formerly Twitter, Instagram, etc. Reach out to us pl partially examinedlife.com let us know what topics you'd like to cover.
There's the blog@ partialexaminedlife.com where you can comment on this episode. You could say, yes, I really agree with Dylan and and Seth was so, so wrong. Grice is very humanistic and and fulfills my soul with with glee, as he apparently does for Steve here.
55:57 Seth: Seth was actually intending that happy path interpretation that Dylan generously gave him to.
56:03 Mark: All right, good night, everybody.
56:04 Seth: Good night.
56:05 Dylan: Good night.
Leave a Reply