Samenvatting Pragmatiek
Week 1 Introduction:
Readings, The dark matter of pragmatics:
1. What is pragmatics and why does it matter?
Pragmatics is the study of how language is used to communicate.
Pragmatics is new, only 50 years old.
Reason: We take usage patterns for granted and there were no recording
techniques.
Pragmatics are important for understanding how language can cause and
prevent situations (airplane example).
2. How to find out what we don’t know we don’t know
The pragmatic dark matter is everything we don’t know about pragmatics.
For example: how we come to interpret particular utterances in the way
we do in a specific context.
Reasons why there is so much dark matter in pragmatics:
o The subject is 50 years old
o The basics and ground works are old and need to be refreshed. A
great deal of theory dates back to the foundational period (1960-
1985)
o Most theories are based on Western philosophy, no input from for
example Asia
o Mostly focused on European languages and English
o Unsophisticated tools were used
o There are many under-developed topics of research, example: sign
language
To find the known unknowns, we must look at the known knowns and
discover the edges of our known universe. Everything beyond that is
unknown and could be explored.
3. The human communication bottleneck and the niche for
pragmatics
There is a bottleneck on speech production maximum of 7 to 8 syllables
(lettergreep) per second.
The speed of human language production is extremely slow!
However, the speed of human understanding is way faster.
The bottleneck is a coding bottleneck.
The gap between the coding rate and the comprehension rate is the
pragmatic niche the zone that can be filled by additional means of
communication.
Language use tends to optimal efficiency even when engaged in the
practice of bullshit humans are optimizers in communication.
4. A design perspective on human communication
,An engineer will try to build a language with maximal expressive or
communicative effectiveness. He will try to maximize the data
transmission rate.
The effective bit rate depends on the noise in the channel and on the
construction of the language.
The best way to handle noise is to add backup or repetition.
When the bit rate is fixed, the engineer will maximize semantic
informativeness.
We need the full apparatus that natural languages have to express logical
relations using a general vocabulary.
Then, the engineer will use 5 tricks to surround that speech production
bottleneck and utilize the full potential gap between slow speech
production and fast comprehension.
Circumventing = omzeilen
5. The first trick to circumvent the bottleneck: Multiplying
channels
5.1 The known knowns: Multimodality
We have a strict coding speed limit on the speech channel so we will use
other channels as supporting devices, for example gesture.
You can multiply channels (example: speech + gestures + facial
expressions) and withing channels multiply layers (within one channel,
example: speech pitch, speed).
5.2 The corresponding dark matter: an orchestra with 150 instruments
We have no idea how many layers or channels are reliably in use across
cultures.
Speaking isn’t just words; it’s a whole-body performance with many
instruments working together to communicate meaning an orchestra
with 150 instruments.
Multimodal binding problem = How do we know which signals on each
of these layers belong together? How do we unite them into a coherent
message?
6. A second trick to circumvent the bottleneck: dual content, word
and deed
6.1 The known knowns: Action potential
Steganography = the art and science of hiding one message within
another.
You can double the content if you can send 2 messages simultaneously
and we do this all the time.
Austin: When we utter a sentence, we do not only say things (locutionary
act) but we also do things (illocutionary act).
Perlocution = indirect action.
Searle: divided different speech act classes:
, o Representatives (assertions)
o Directives (requests, questions)
o Commissives (threats, promises)
o Expressives (thanking, apologies)
o Declarations (blessings, namings)
6.2 The associated dark matter: Finding actions
Main domain of pragmatics: How are actions mapped onto words?
It is hard to understand this because:
o It is unclear whether there is a finite list of potential actions.
o There has not been a systematic comparison of speech act types
across a good sample of unrelated cultures.
o We don’t know how people ascribe actions to utterances.
2 possible answers to how mapping actions onto utterances is so efficient
and fast:
- We interpret behaviour on the assumption that it is goal driven. We are
continually in the business of goal reconstruction.
- We build up a vast association network between utterance forms,
contexts, and actions.
The whole point of language is to deliver actions. The sequence of these is
the thread that ties together conversational activity.
7. A third trick to circumvent the bottleneck: Choice of message
form
7.1 The know knowns: Utterance-type meanings
We can convey extra meaning through the form or type of utterance itself,
rather than the literal content.
Presuppositions can help with this implied assumption that must be true
for an utterance to make sense.
This part is mostly about methods of amplifying coded content by virtue of
prearranged rules of thumb or pre-packaging of default assumptions.
Inferences depend on contrast and expectation.
7.2 The corresponding dark matter: How many principles?
There is a huge open question about how we process inferences
(gevolgtrekkingen, we concluderen iets)
Potential additional principles that structure our communication:
o Acknowledge the presence of another. Example: eye contact
maintains social coordination.
o Match the channel, medium and tone. People modulate their own
speech patterns to more closely match the other’s.
o Stay within the topic and/or activity, or signal otherwise.
Conversation flows smoothly because people generally mean
relevant, or mark when they change the subject.
These and other tacit norms make human communication remarkably
efficient and cooperative. Yet, Levinson points out that we still lack a clear
, theory of how many such principles exist, how universal they are, and how
they are learned or represented. Understanding these hidden regularities
— the “dark matter” — is essential to fully explaining how humans
interpret and produce meaning far beyond literal language.
8. A fourth trick to circumvent the bottleneck: Non-literal uses of
language
8.1 The known knowns: How language use goes on holiday and does
figures of eight
We can mean a great deal more than we say, even indefinitely more. Our
ability to use irony, metaphors, and indirectness is what makes language
flexible, creative, and deeply human.
It helps pass the bottleneck by packing multiple layers of meaning into
simple expressions.
8.2 The corresponding dark matter: Finding the message
We still don’t fully understand how people interpret non-literal meanings
because this process seems effortless but actually involves very complex,
hidden mental work.
9. A fifth trick to circumvent the bottleneck: Leveraging the
context
9.1 The known knowns: Trading on common ground
The strategy of relying on the context. Utterances both conform to
contexts and project them, allowing us to smuggle in our presumptions
through the encoding bottleneck.
Ways to exploit the context:
o Through deictic reference.
o Presupposing whenever you can.
o Assuming the listener will add what’s just been said to what they
already know and use it to draw new conclusions.
9.2 The black hole of context
The dark matter of context that is interesting studying is the way in which
factors that might be relevant to understanding a talk exchange are
structured.
9.3 Summarizing the tricks for circumventing the production bottleneck
The 5 tricks use the residual cognitive capacity to understand much more
than what is actually said.
1st trick = multiply channels and modalities. Dark matter = how and which
signals are combined into one clear massage
2nd trick = adding action content to each utterance. Dark matter = how we
actually do action attribution.
3rd trick = how something is said can suggest certain meanings. Dark
matter = what is agreed on? Are there more rules? Where do
presuppositions come from?
Week 1 Introduction:
Readings, The dark matter of pragmatics:
1. What is pragmatics and why does it matter?
Pragmatics is the study of how language is used to communicate.
Pragmatics is new, only 50 years old.
Reason: We take usage patterns for granted and there were no recording
techniques.
Pragmatics are important for understanding how language can cause and
prevent situations (airplane example).
2. How to find out what we don’t know we don’t know
The pragmatic dark matter is everything we don’t know about pragmatics.
For example: how we come to interpret particular utterances in the way
we do in a specific context.
Reasons why there is so much dark matter in pragmatics:
o The subject is 50 years old
o The basics and ground works are old and need to be refreshed. A
great deal of theory dates back to the foundational period (1960-
1985)
o Most theories are based on Western philosophy, no input from for
example Asia
o Mostly focused on European languages and English
o Unsophisticated tools were used
o There are many under-developed topics of research, example: sign
language
To find the known unknowns, we must look at the known knowns and
discover the edges of our known universe. Everything beyond that is
unknown and could be explored.
3. The human communication bottleneck and the niche for
pragmatics
There is a bottleneck on speech production maximum of 7 to 8 syllables
(lettergreep) per second.
The speed of human language production is extremely slow!
However, the speed of human understanding is way faster.
The bottleneck is a coding bottleneck.
The gap between the coding rate and the comprehension rate is the
pragmatic niche the zone that can be filled by additional means of
communication.
Language use tends to optimal efficiency even when engaged in the
practice of bullshit humans are optimizers in communication.
4. A design perspective on human communication
,An engineer will try to build a language with maximal expressive or
communicative effectiveness. He will try to maximize the data
transmission rate.
The effective bit rate depends on the noise in the channel and on the
construction of the language.
The best way to handle noise is to add backup or repetition.
When the bit rate is fixed, the engineer will maximize semantic
informativeness.
We need the full apparatus that natural languages have to express logical
relations using a general vocabulary.
Then, the engineer will use 5 tricks to surround that speech production
bottleneck and utilize the full potential gap between slow speech
production and fast comprehension.
Circumventing = omzeilen
5. The first trick to circumvent the bottleneck: Multiplying
channels
5.1 The known knowns: Multimodality
We have a strict coding speed limit on the speech channel so we will use
other channels as supporting devices, for example gesture.
You can multiply channels (example: speech + gestures + facial
expressions) and withing channels multiply layers (within one channel,
example: speech pitch, speed).
5.2 The corresponding dark matter: an orchestra with 150 instruments
We have no idea how many layers or channels are reliably in use across
cultures.
Speaking isn’t just words; it’s a whole-body performance with many
instruments working together to communicate meaning an orchestra
with 150 instruments.
Multimodal binding problem = How do we know which signals on each
of these layers belong together? How do we unite them into a coherent
message?
6. A second trick to circumvent the bottleneck: dual content, word
and deed
6.1 The known knowns: Action potential
Steganography = the art and science of hiding one message within
another.
You can double the content if you can send 2 messages simultaneously
and we do this all the time.
Austin: When we utter a sentence, we do not only say things (locutionary
act) but we also do things (illocutionary act).
Perlocution = indirect action.
Searle: divided different speech act classes:
, o Representatives (assertions)
o Directives (requests, questions)
o Commissives (threats, promises)
o Expressives (thanking, apologies)
o Declarations (blessings, namings)
6.2 The associated dark matter: Finding actions
Main domain of pragmatics: How are actions mapped onto words?
It is hard to understand this because:
o It is unclear whether there is a finite list of potential actions.
o There has not been a systematic comparison of speech act types
across a good sample of unrelated cultures.
o We don’t know how people ascribe actions to utterances.
2 possible answers to how mapping actions onto utterances is so efficient
and fast:
- We interpret behaviour on the assumption that it is goal driven. We are
continually in the business of goal reconstruction.
- We build up a vast association network between utterance forms,
contexts, and actions.
The whole point of language is to deliver actions. The sequence of these is
the thread that ties together conversational activity.
7. A third trick to circumvent the bottleneck: Choice of message
form
7.1 The know knowns: Utterance-type meanings
We can convey extra meaning through the form or type of utterance itself,
rather than the literal content.
Presuppositions can help with this implied assumption that must be true
for an utterance to make sense.
This part is mostly about methods of amplifying coded content by virtue of
prearranged rules of thumb or pre-packaging of default assumptions.
Inferences depend on contrast and expectation.
7.2 The corresponding dark matter: How many principles?
There is a huge open question about how we process inferences
(gevolgtrekkingen, we concluderen iets)
Potential additional principles that structure our communication:
o Acknowledge the presence of another. Example: eye contact
maintains social coordination.
o Match the channel, medium and tone. People modulate their own
speech patterns to more closely match the other’s.
o Stay within the topic and/or activity, or signal otherwise.
Conversation flows smoothly because people generally mean
relevant, or mark when they change the subject.
These and other tacit norms make human communication remarkably
efficient and cooperative. Yet, Levinson points out that we still lack a clear
, theory of how many such principles exist, how universal they are, and how
they are learned or represented. Understanding these hidden regularities
— the “dark matter” — is essential to fully explaining how humans
interpret and produce meaning far beyond literal language.
8. A fourth trick to circumvent the bottleneck: Non-literal uses of
language
8.1 The known knowns: How language use goes on holiday and does
figures of eight
We can mean a great deal more than we say, even indefinitely more. Our
ability to use irony, metaphors, and indirectness is what makes language
flexible, creative, and deeply human.
It helps pass the bottleneck by packing multiple layers of meaning into
simple expressions.
8.2 The corresponding dark matter: Finding the message
We still don’t fully understand how people interpret non-literal meanings
because this process seems effortless but actually involves very complex,
hidden mental work.
9. A fifth trick to circumvent the bottleneck: Leveraging the
context
9.1 The known knowns: Trading on common ground
The strategy of relying on the context. Utterances both conform to
contexts and project them, allowing us to smuggle in our presumptions
through the encoding bottleneck.
Ways to exploit the context:
o Through deictic reference.
o Presupposing whenever you can.
o Assuming the listener will add what’s just been said to what they
already know and use it to draw new conclusions.
9.2 The black hole of context
The dark matter of context that is interesting studying is the way in which
factors that might be relevant to understanding a talk exchange are
structured.
9.3 Summarizing the tricks for circumventing the production bottleneck
The 5 tricks use the residual cognitive capacity to understand much more
than what is actually said.
1st trick = multiply channels and modalities. Dark matter = how and which
signals are combined into one clear massage
2nd trick = adding action content to each utterance. Dark matter = how we
actually do action attribution.
3rd trick = how something is said can suggest certain meanings. Dark
matter = what is agreed on? Are there more rules? Where do
presuppositions come from?
