Grice’s Maxims of Conversation(effectiviology.com)
effectiviology.com
Grice’s Maxims of Conversation
https://effectiviology.com/principles-of-effective-communication/
23 comments
> virtually all speakers intuitively understand and follow whether they've heard of them or not
Do they? I regularly speak with people who can't stick to the point, ramble, start talking about something whose first half of the conversation occurred only in their heads, and are prone to providing unnecessary detail.
Haven't you never experienced a three way conversation where you notice one of the parties used an ambiguous term and the other party chose the wrong meaning? I find this happens pretty frequently; and of course, I've been either of these parties at times!
Do they? I regularly speak with people who can't stick to the point, ramble, start talking about something whose first half of the conversation occurred only in their heads, and are prone to providing unnecessary detail.
Haven't you never experienced a three way conversation where you notice one of the parties used an ambiguous term and the other party chose the wrong meaning? I find this happens pretty frequently; and of course, I've been either of these parties at times!
All the time. Some stuck have stuck in my head. Once in a fraught technical exchange, an example was contrived regarding goegraphical place and temperature. One party talked about handling degrees and meant either latitude or temperature (I forget which), the other party took it as the other (temperature or latitude), and argument ensued over bogus handling of data. I attempted to alert them as to what the misunderstanding was, and they both glared at me and said to stay out of it. Took a bunch of time in a meeting involving at least 13 people to finally converge.
There is a missing maxim about being welcome to engage in the conversation in the first place.
There is a missing maxim about being welcome to engage in the conversation in the first place.
Pobody's nerfect. Intuition isn't a precise skill.
Humans also intuitively solve differential equations when catching a ball, but they can't land rockets on mars off the top of their heads.
Humans also intuitively solve differential equations when catching a ball, but they can't land rockets on mars off the top of their heads.
Yes, yes. I meant that people do not intuitively understand or follow these rules, as asserted by the post I was replying to. People routinely fail to follow them, and lots of people I have direct contact with don't even understand why being ambiguous when giving directions for a job is a bad thing.
I have a perfect example of this: one person I know, when he says "next Monday" only truly means the next immediate Monday when it's more than 3-4 days away and never when next Monday is tomorrow. In all other cases he'll say "this Monday" when he means the next one, and "the 'other' Monday" when he means the one after the next one. It seems to be a very imprecise function of the distance to the actual next Monday.
Now, some considerations: some people actually talk like this in informal contexts, though not everyone and there's no way of always knowing the context. Also, I'm translating from Spanish; presumably the English vernacular has similar imprecise conventions.
What drives me crazy is that this imprecise language is used when arranging a date or scheduling some job. "Next Monday I have a doctor's appointment", or "This task will be finished next Monday". This person is always puzzled when I ask him to clarify what he means by "next Monday" -- is it the actual next Monday or the other Monday?
Of course many times you can guess by context, especially if you're familiar with this convention, but it's not always possible and it catches people unfamiliar with the convention off-guard.
And this is just an example. It happens all the time, with lots of people, with lots of ambiguities that have real impact. People absolutely do NOT follow these maxims of conversation.
Another example: I could have explained all of this way more concisely. I chose to rant instead ;)
I have a perfect example of this: one person I know, when he says "next Monday" only truly means the next immediate Monday when it's more than 3-4 days away and never when next Monday is tomorrow. In all other cases he'll say "this Monday" when he means the next one, and "the 'other' Monday" when he means the one after the next one. It seems to be a very imprecise function of the distance to the actual next Monday.
Now, some considerations: some people actually talk like this in informal contexts, though not everyone and there's no way of always knowing the context. Also, I'm translating from Spanish; presumably the English vernacular has similar imprecise conventions.
What drives me crazy is that this imprecise language is used when arranging a date or scheduling some job. "Next Monday I have a doctor's appointment", or "This task will be finished next Monday". This person is always puzzled when I ask him to clarify what he means by "next Monday" -- is it the actual next Monday or the other Monday?
Of course many times you can guess by context, especially if you're familiar with this convention, but it's not always possible and it catches people unfamiliar with the convention off-guard.
And this is just an example. It happens all the time, with lots of people, with lots of ambiguities that have real impact. People absolutely do NOT follow these maxims of conversation.
Another example: I could have explained all of this way more concisely. I chose to rant instead ;)
What's interesting to me is how "technical" concepts might predict someone's inability to adhere to Grice's Maxims.
A person in a manic episode might be affected by Logorrhea and therefore may give too much information (violation of Maxim of Quantity)
Someone with Asperger's or ADHD may be inclined to tangents (violation of Maxim of Relation) or misuse figures of speech, misunderstand implied meanings, or be too detailed in an explanation (violation of Maxim of Manner)
A person in a manic episode might be affected by Logorrhea and therefore may give too much information (violation of Maxim of Quantity)
Someone with Asperger's or ADHD may be inclined to tangents (violation of Maxim of Relation) or misuse figures of speech, misunderstand implied meanings, or be too detailed in an explanation (violation of Maxim of Manner)
If curious, past thread:
Grice's Maxims - https://news.ycombinator.com/item?id=18458332 - Nov 2018 (10 comments)
Grice's Maxims - https://news.ycombinator.com/item?id=18458332 - Nov 2018 (10 comments)
> While these maxims were originally meant to describe how people intuitively communicate, they can be used to actively guide the way you communicate in various situations.
This isn't something I'd endorse doing often. Grice was making a descriptive statement about the linguistic pragmatics of human conversation; he was absolutely not saying that this is how people 'should' speak. Reversing the principles like this turns them into listicle fodder instead of the rather insightful framework for understanding pragmatics that they are.
A parallel exists in linguistic syntax. The goal of linguistic syntax is to describe which sentences are recognized by speakers of a language as grammatical, rather than prescribe some notion of 'correct' grammar. If you take syntactic descriptions and interpret them as rules, you are profoundly missing the point - see "Don't end your sentences with a preposition!" nonsense.
There's more reading on Grice's maxims in their original, descriptive linguistics/social science context is here: https://en.wikipedia.org/wiki/Cooperative_principle
This isn't something I'd endorse doing often. Grice was making a descriptive statement about the linguistic pragmatics of human conversation; he was absolutely not saying that this is how people 'should' speak. Reversing the principles like this turns them into listicle fodder instead of the rather insightful framework for understanding pragmatics that they are.
A parallel exists in linguistic syntax. The goal of linguistic syntax is to describe which sentences are recognized by speakers of a language as grammatical, rather than prescribe some notion of 'correct' grammar. If you take syntactic descriptions and interpret them as rules, you are profoundly missing the point - see "Don't end your sentences with a preposition!" nonsense.
There's more reading on Grice's maxims in their original, descriptive linguistics/social science context is here: https://en.wikipedia.org/wiki/Cooperative_principle
> This isn't something I'd endorse doing often. Grice was making a descriptive statement about the linguistic pragmatics of human conversation; he was absolutely not saying that this is how people 'should' speak.
Advice is directional, not absolute. If someone struggles with correctly identifying and adopting the shared interpersonal protocols most people use in society, a documented, descriptive guide to those protocols can be used as a guide.
Even if Grice’s project was to document and model the set of rules that “everyone” implicitly knows and follows, if you’re not a part of that “everyone” and you struggle to reach understanding when communicating with others, maybe Grice’s maxims can explain what you missed.
Advice is directional, not absolute. If someone struggles with correctly identifying and adopting the shared interpersonal protocols most people use in society, a documented, descriptive guide to those protocols can be used as a guide.
Even if Grice’s project was to document and model the set of rules that “everyone” implicitly knows and follows, if you’re not a part of that “everyone” and you struggle to reach understanding when communicating with others, maybe Grice’s maxims can explain what you missed.
If "don't end your sentences with a preposition" was an actual grammar rule, no one would ever talk about it because every native speaker would intuitively follow it anyway.
> Don't end your sentences with a preposition!"
That's based on a misunderstanding of syntax, not an excessive attention to it.
That's based on a misunderstanding of syntax, not an excessive attention to it.
> This isn't something I'd endorse doing often.
Why not? They seem like helpful principles.
Why not? They seem like helpful principles.
I think particularly for autistic people that struggle to pick up the implicit rules of the game, they can be useful.
Are there situations where it would be harmful to follow these rules?
These rules can be a hinderance during social conversations, where the goal is often to show that you understand the beliefs and practices of the in-group and to demonstrate your superiority through a command of humor or charisma, where the implied and non-verbal communication is possibly more important than the actual spoken words.
Sure. If you are talking to a woman in a romantic context, the purpose of the conversation is not to convey the most information in the fewest words possible.
Software people with philosophy backgrounds (there are a lot of us!) will be quick to note that Grice did not intend these to be understood as rules of thumb for regular people. But if it gets ordinary citizens interested in philosophy of language, theories of rationality, or game theoretic ideas about convention (cf. David Lewis), then I don't mind some innocent misappropriation.
When I was training people in sales, this is the first thing I would talk about.
Grice was a god tier philosopher of language. Best known for this: https://en.m.wikipedia.org/wiki/Implicature#:~:text=An%20imp....
My favorite maxim is the ironically worded:
> Be brief (avoid unnecessary prolixity)
The parenthetical is in Grice's original article, but it doesn't make it into most of the secondary sources (presumably because the humor might mislead or confuse readers). The parenthetical is in the wikipedia entry, however. [1]
1: https://en.wikipedia.org/wiki/Cooperative_principle
> Be brief (avoid unnecessary prolixity)
The parenthetical is in Grice's original article, but it doesn't make it into most of the secondary sources (presumably because the humor might mislead or confuse readers). The parenthetical is in the wikipedia entry, however. [1]
1: https://en.wikipedia.org/wiki/Cooperative_principle
In that vein this made me laugh out loud too:
Be perspicuous.
I had to look up perspicuous, it means ‘clear in communication’.
Be perspicuous.
I had to look up perspicuous, it means ‘clear in communication’.
This site ironically illustrates a principle of group communication dynamics that Grice himself never considered. No matter the quality or relevance of any single information source, the human brain simply can't process or filter hundreds or thousands of simultaneous channels. Hence, no matter how great your publication is, I am never going to give you my email because I don't want to increase the hour a day I already spend deleting emails from every vendor I have ever made a purchase of anything from so I can get to the three or so from my neighbors or coworkers I actually need to read. I was about to bookmark this site until I scrolled a quarter inch and that popup blocked my screen asking me to sign up for a newsletter.
Somebody needs to formalize the principles of communication at scale in order to capture the reality that whatever works when it's one on one no longer works when it's one on forty thousand. We seem to understand this with non-human request processors. If you run one instance of a single-threaded server, you're going to get DOSd and be unable to handle any requests at all if you get too many at a time. But everybody who publishes a web page with textual information doesn't seem to understand that this same thing happens to human readers as well, except human reading is necessarily single-threaded. No matter how important your request is, I can't process it if I'm getting DOSd, and I can't clone myself or fork my ability to read.
At some point, content creators and sellers are going to have to realize their relationship with consumers has fundamentally changed. You can't proactively push information. There is simply too much of it. You need to switch to a pull only model, which is scary because now consumers have to intentionally seek you out and you need to rely on discoverability, but sticking with push marketing is a collective social disaster. The attention economy is overwhelming and destroying our ability to figure out for ourselves what is actually worth spending our time on.
Somebody needs to formalize the principles of communication at scale in order to capture the reality that whatever works when it's one on one no longer works when it's one on forty thousand. We seem to understand this with non-human request processors. If you run one instance of a single-threaded server, you're going to get DOSd and be unable to handle any requests at all if you get too many at a time. But everybody who publishes a web page with textual information doesn't seem to understand that this same thing happens to human readers as well, except human reading is necessarily single-threaded. No matter how important your request is, I can't process it if I'm getting DOSd, and I can't clone myself or fork my ability to read.
At some point, content creators and sellers are going to have to realize their relationship with consumers has fundamentally changed. You can't proactively push information. There is simply too much of it. You need to switch to a pull only model, which is scary because now consumers have to intentionally seek you out and you need to rely on discoverability, but sticking with push marketing is a collective social disaster. The attention economy is overwhelming and destroying our ability to figure out for ourselves what is actually worth spending our time on.
The article lacked an example of the "Maxim of Relation", so here's one:
Alice: "Where can I buy some tire chains around here?"
Bob: "There's a gas station down the block."
If Bob knows that the gas station down the block does not in fact sell tire chains, he has not technically lied. However, he has violated the Maxim of Relation by providing information irrelevant to Alice's question, leading to a misunderstanding if Alice assumes it is relevant, implying that the station does sell tire chains.
Alice: "Where can I buy some tire chains around here?"
Bob: "There's a gas station down the block."
If Bob knows that the gas station down the block does not in fact sell tire chains, he has not technically lied. However, he has violated the Maxim of Relation by providing information irrelevant to Alice's question, leading to a misunderstanding if Alice assumes it is relevant, implying that the station does sell tire chains.
Where they're really useful, though, is in identifying when people are being deceptive or misleading. It happens all the time with politicians taking stuff out of context - "My opponent said xyz is true, which is terrible!" when the reality is that your opponent said xyz is true only when abc happens, they've violated Grice's Maxim of Quantity.
Oftentimes people get tripped up when they're speaking to someone who is saying things that are technically true but feel misleading, but they're unable to really pinpoint what's wrong with them - a lot of the time it's because the other speaking is violating the the Maxims of Quantity or Relation. If you can identify that, then it's a lot easier to figure out what the other person is doing (and to try to understand why they're doing it).