Modules: Difference between revisions

From the Logical Languages Wiki
Jump to navigation Jump to search
(retrieved and converted document from https://docs.google.com/document/d/1R9K740crNFTCJ0ah-L3kyJYNHd15fWFsCk_4JGAiaMU/edit)
 
(→‎Predicates and their arguments: Adding Eberban as an example of predicate-only syntax.)
 
(23 intermediate revisions by 2 users not shown)
Line 1: Line 1:
== Goals ==
== Goals of the Logical Language Modules Project ==
 
Build a Library of Logical Language Modular Components and Tools.


These components should aim to:
These components should aim to:


* A full logical language can be assembled wholly or partly from existing modules.
* A full logical language can be assembled wholly or partly from existing modules.
 
** Innovators can focus on developing a new module rather than a whole language.
Innovators can focus on developing a new module rather than a whole language.
 
* Enumerate the design space of Logical Languages.
* Enumerate the design space of Logical Languages.
*: Provide alternate modules for different design choices.
** Provide alternate modules for different design choices.
* Describe modules' strengths, weaknesses, and compatibility.
* Describe modules' strengths, weaknesses, and compatibility.
*: Do not apply value judgements or advocate for any particular option.
** Do not apply value judgements or advocate for any particular option.
* Be systematic and well documented.
* Be systematic and well documented.
*: A method or a choice without documented justification will be lost.
** A method or a choice without documented justification will be lost.
* Provide software for working with these components.
* Provide software for working with these components.


== Modular Language Architecture ==
== Related Work ==
 
This section will explore how a language could be broken down into modules.
 
=== Module Breakdown ===
 
; [[wikipedia:Phonology|Phonology]]
: Different sets of [[wikipedia:Phoneme|phonemes]] which can be used.
 
; [[wikipedia:Phonotactics|Phonotactics]]
: Different rules for how a phonology can be assembled into words.


; Writing System
There are a number of existing [https://www.frathwiki.com/Software_tools_for_conlanging software tools for constructed languages]. Some of these include related input datasets, such as vocabulary.


; Structure, Syntax, and Semantics
This project differs from these existing tools, because it is focused on producing logical languages.


; Vocabulary
Some functions, such as generating word morphologies (absent of meaning), are largely arbitrary decisions and do not need to be done differently for logical languages (although we wouldn't want to prevent a logical language from controlling word morphologies more closely).


* Provides set of non-core/non-syntactic/non-structural words with defined meanings
Some functions, like providing an explicit mapping to a semantic foundation, are a key focus for logical languages. This will need to be a focus of software tooling specific to logical languages.
* Not attached to specific phonological forms
* Different vocabulary module instances could incorporate different philosophies


* e.g. world view for constructing composites:
== Modular Logical Language Architecture ==
*: Have a word for 'tooth', or have a compound like 'mouth-stone'?
 
=== Module Hierarchy ===


* Morphology
* Morphology
Line 49: Line 34:
**** Phonology
**** Phonology
* Vocabulary
* Vocabulary
* Structure, syntax, semantics
* Syntax
** Predicates and Arguments
** Numbers and Counting
* Semantic Foundation
 
== Module Design Space and Features ==
 
<!-- To editors: please pay attention to the exact structure of these entries. -->
 
=== Morphology ===
* Self-segmentation strategies
*; Continuation marker
*: certain feature or features of a syllable is used to determine if it is a continuation of the previous word
*:* [[Toaq]] uses the flat toneme for this<ref>http://toaq.org/#compound_predicates</ref>
*; Word length marker
*: the first syllable of each word determines how long the word is
*:* [[X-1]]<ref>https://mw.lojban.org/papri/X-1#Morphology</ref>
*; Initial consonant cluster
*:* [[Lojban]]<ref>https://lojban.github.io/cll/4/3/</ref>
*; Sentinel value
*: [[wikipedia:Sentinel value|programming concept]]: special predetermined value as a signal for termination
*:* [[Lojban]]: name words (''cmevla'') are booked with pauses<ref>https://lojban.github.io/cll/4/8/</ref><ref>https://mw.lojban.org/papri/BPFK_Section:_cmevla</ref>; foreign quotes triggered with ''zoi''/''la'o'' start and end with any chosen one-word terminator<ref>https://lojban.github.io/cll/19/10/</ref>
* Relation between Written and Verbal
*; Audio-Visual Isomorphism
*:* [[Lojban]]<ref>https://lojban.github.io/cll/3/1/</ref><ref>https://mw.lojban.org/papri/audio-visual_isomorphism</ref>
 
==== Written ====
* Writing System
*; Alphabet
*; Syllabary
*; Logography
 
* Structure
*; Linear text
*; Cognitive Maps
 
==== Verbal ====
* [[wikipedia:Phonology|Phonology]]
*: Different sets of [[wikipedia:Phoneme|phonemes]] which can be used.
* [[wikipedia:Phonotactics|Phonotactics]]
*: Different rules for how a phonology can be assembled into words.
 
==== Gestures ====
* Gesture System
*: [[wikipedia:Sign_language|Sign language]] or [[wikipedia:Manually_coded_language|Hand gestures]] could replace or supplement verbal communication.
*; [[wikipedia:Manual_alphabet|Hand gesture alphabet]]
*; Hand gesture logography (signs)
=== Vocabulary ===
 
Provides set of non-core/non-syntactic/non-structural words with defined meanings
Not attached to specific phonological forms
Different vocabulary module instances could incorporate different philosophies, such as world view for constructing composites - have a word for 'tooth', or use a compound like 'mouth-stone'?
 
* Vocabulary for Opposites
*; One word per dimension, use negation
*: Smaller, more atomic vocabulary.
*:* ''happy'' vs ''unhappy''
*:* [[Lojban]]: ''gleki'' vs ''tolgleki''
*; One word per dimension direction
*:* ''happy'' vs ''sad''
*:* [[Lojban]]: ''gleki'' vs ''badri''
*: (Note: it may be debatable that happiness and sadness are opposites, for this example, assume they are)
* Vocabulary for Intensities
*; Multiple words for different intensities
*:* ''happy'' vs ''elated'' vs ''ecstatic''
*; Modifier for different intensities
*:* ''happy'' vs ''very happy'' vs ''extremely happy''
*: (Note: you may need to ignore some connotations for these examples to fit better)
 
 
=== Syntax ===
 
==== Predicates and their arguments ====
* Sentence functions
*; Predicate–argument
*: predicates, together with arguments, form predications; both must be present (at least implicitly)
*:* <code>see(man, cat)</code>
*:* ''present in most logical languages''
*; Predicate–variable
*: predicates only accept variable labels as terms; these, in turn, can be bound by quantifier expressions and restricted with subsequent predicate clauses
*:* <code>∃xy[man(x) ∧ cat(y) ∧ see(x, y)]</code>
*:* [[Xorban]]; [[X-1]]<ref>https://mw.lojban.org/papri/X-1#Sample</ref>
*; Predicate
*: no arguments are exposed; predicates are implicitly connected with quantified variables
*:* <code>man₁ = see₁; see₂ = cat₁</code>
*:* [[Eberban]]
* Ad-hoc predicate composition
*; Serial predicates
*: the embedding of one predicate's structure inside another
*:* [[Toaq]], using the falling tone<ref>http://toaq.org/#serial_predicates</ref>; [[Gua\spi]]<ref>http://www.jfcarter.net/~jimc/guaspi/acmpaper.html</ref>, using the high even tone
*:* [[Lojban]] of at least one dialect: ''turtau''<ref>https://mw.lojban.org/papri/User:Gleki/turtau</ref>
*; Compound metaphors
*: predicate apposition as modification of one predicate's meaning by the other's
*:* [[Lojban]] has a highly developed appositional ''tanru'' grammar
 
==== Numbers and counting ====
* Representation
*; Numerals as a part of speech
*: numbers constitute a separate grammatical class
*:* [[Lojban]]: the ''selma'o'' (grammatical class) PA<ref name="lojban-numbers">https://lojban.github.io/cll/18/2/</ref>
*; Numbers as predicates
*: numbers are cardinality predicates (‘x₁ is three in number’ etc.)
*:* [[Toaq]]<ref>http://toaq.org/#numbers</ref>
* Composition
*; Concatenation
*: values are expressed as a sequence of digits in positional notation
*:* [[Lojban]]: ''pa re ci vo mu'' (lit. ‘1 2 3 4 5’) denotes the number 12,345<ref name="lojban-numbers">https://lojban.github.io/cll/18/2/</ref>
* Usage
*; As quantifiers
*: numbers attach to terms and scope over the predicate, signifying how many possible values of the term satisfy the predicate
*:* [[Lojban]]: ''ci da'' (lit. ‘three something’) = ‘there exist exactly three things that…’<ref>https://lojban.github.io/cll/6/6/</ref>. Lojban has an extensive assortment of number grammar particles which allow to construct more elaborate quantifications: for example, ''su'o (pa)'' ‘at least one’ for ∃, ''ro'' ‘all’ for ∀, or even things like ''da'a su'e rau'' ‘all but at most enough’.
 
=== Semantic Foundations ===
 
* System of Logic
*; Predicate Logic
*:* [[Toaq]]<ref>http://toaq.org/#predicates</ref>
 
== References ==
<references />

Latest revision as of 11:16, 10 March 2022

Goals of the Logical Language Modules Project

Build a Library of Logical Language Modular Components and Tools.

These components should aim to:

  • A full logical language can be assembled wholly or partly from existing modules.
    • Innovators can focus on developing a new module rather than a whole language.
  • Enumerate the design space of Logical Languages.
    • Provide alternate modules for different design choices.
  • Describe modules' strengths, weaknesses, and compatibility.
    • Do not apply value judgements or advocate for any particular option.
  • Be systematic and well documented.
    • A method or a choice without documented justification will be lost.
  • Provide software for working with these components.

Related Work

There are a number of existing software tools for constructed languages. Some of these include related input datasets, such as vocabulary.

This project differs from these existing tools, because it is focused on producing logical languages.

Some functions, such as generating word morphologies (absent of meaning), are largely arbitrary decisions and do not need to be done differently for logical languages (although we wouldn't want to prevent a logical language from controlling word morphologies more closely).

Some functions, like providing an explicit mapping to a semantic foundation, are a key focus for logical languages. This will need to be a focus of software tooling specific to logical languages.

Modular Logical Language Architecture

  • Morphology
    • Written
      • Writing System
    • Verbal
      • Phonotactics
        • Phonology
  • Vocabulary
  • Syntax
    • Predicates and Arguments
    • Numbers and Counting
  • Semantic Foundation

Module Design Space and Features

Morphology

  • Self-segmentation strategies
    Continuation marker
    certain feature or features of a syllable is used to determine if it is a continuation of the previous word
    • Toaq uses the flat toneme for this[1]
    Word length marker
    the first syllable of each word determines how long the word is
    Initial consonant cluster
    Sentinel value
    programming concept: special predetermined value as a signal for termination
    • Lojban: name words (cmevla) are booked with pauses[4][5]; foreign quotes triggered with zoi/la'o start and end with any chosen one-word terminator[6]
  • Relation between Written and Verbal
    Audio-Visual Isomorphism

Written

  • Writing System
    Alphabet
    Syllabary
    Logography
  • Structure
    Linear text
    Cognitive Maps

Verbal

Gestures

Vocabulary

Provides set of non-core/non-syntactic/non-structural words with defined meanings Not attached to specific phonological forms Different vocabulary module instances could incorporate different philosophies, such as world view for constructing composites - have a word for 'tooth', or use a compound like 'mouth-stone'?

  • Vocabulary for Opposites
    One word per dimension, use negation
    Smaller, more atomic vocabulary.
    • happy vs unhappy
    • Lojban: gleki vs tolgleki
    One word per dimension direction
    • happy vs sad
    • Lojban: gleki vs badri
    (Note: it may be debatable that happiness and sadness are opposites, for this example, assume they are)
  • Vocabulary for Intensities
    Multiple words for different intensities
    • happy vs elated vs ecstatic
    Modifier for different intensities
    • happy vs very happy vs extremely happy
    (Note: you may need to ignore some connotations for these examples to fit better)


Syntax

Predicates and their arguments

  • Sentence functions
    Predicate–argument
    predicates, together with arguments, form predications; both must be present (at least implicitly)
    • see(man, cat)
    • present in most logical languages
    Predicate–variable
    predicates only accept variable labels as terms; these, in turn, can be bound by quantifier expressions and restricted with subsequent predicate clauses
    Predicate
    no arguments are exposed; predicates are implicitly connected with quantified variables
    • man₁ = see₁; see₂ = cat₁
    • Eberban
  • Ad-hoc predicate composition
    Serial predicates
    the embedding of one predicate's structure inside another
    Compound metaphors
    predicate apposition as modification of one predicate's meaning by the other's
    • Lojban has a highly developed appositional tanru grammar

Numbers and counting

  • Representation
    Numerals as a part of speech
    numbers constitute a separate grammatical class
    Numbers as predicates
    numbers are cardinality predicates (‘x₁ is three in number’ etc.)
  • Composition
    Concatenation
    values are expressed as a sequence of digits in positional notation
    • Lojban: pa re ci vo mu (lit. ‘1 2 3 4 5’) denotes the number 12,345[13]
  • Usage
    As quantifiers
    numbers attach to terms and scope over the predicate, signifying how many possible values of the term satisfy the predicate
    • Lojban: ci da (lit. ‘three something’) = ‘there exist exactly three things that…’[15]. Lojban has an extensive assortment of number grammar particles which allow to construct more elaborate quantifications: for example, su'o (pa) ‘at least one’ for ∃, ro ‘all’ for ∀, or even things like da'a su'e rau ‘all but at most enough’.

Semantic Foundations

  • System of Logic
    Predicate Logic

References