LISTSERV mailing list manager LISTSERV 16.5

Help for CONLANG Archives


CONLANG Archives

CONLANG Archives


CONLANG@LISTSERV.BROWN.EDU


View:

Message:

[

First

|

Previous

|

Next

|

Last

]

By Topic:

[

First

|

Previous

|

Next

|

Last

]

By Author:

[

First

|

Previous

|

Next

|

Last

]

Font:

Proportional Font

LISTSERV Archives

LISTSERV Archives

CONLANG Home

CONLANG Home

CONLANG  August 2015, Week 1

CONLANG August 2015, Week 1

Subject:

Re: Simpler than Hodor

From:

Jason Cullen <[log in to unmask]>

Reply-To:

Constructed Languages List <[log in to unmask]>

Date:

Mon, 3 Aug 2015 15:29:20 -0500

Content-Type:

text/plain

Parts/Attachments:

Parts/Attachments

text/plain (176 lines)

Questions:

Why bother having 'topic' in your system? In your example, "Boy nominative
topic, hill on accusative topic, house accusative topic, see-topic" it
seems it would be easier to just say clause-topic, which, of course, isn't
a topic.
Why do both the preposition and its complement get marked 'accusative'?

On Mon, Aug 3, 2015 at 9:34 AM, Patrik Austin <[log in to unmask]>
wrote:

> Here's a very simple language that takes a minute to learn but a lifetime
> to explore. I'm using an English dictionary, but in principle any lexicon
> would do, ignoring any given grammatical indications of word class, valency
> etc. (using the semantic description only).
>
>
>
> The grammar has the production rule A ‒> aAb only where *a* is a set of
> prepositions and *b* is a set of nouns. It's basically one command line, or
> the possible full expansion of S ‒> aSb | ab | <empty>
>
>
>
> Here's a practical implementation:
>
>
>
> (1) Each noun is marked with the suffix -]; for example: dog].
>
> (2) Each word with the semantic formula “in the x of y” (i.e. preposition)
> is marked with the prefix [-; for example: [with.
>
> (3) Each noun is preceded by a preposition, e.g. [with dog] - “in the
> 'with' of dog”.
>
>
>
> Example sentence:
>
>
>
> [topic [nom boy] [acc [on hill] house] see]
>
> - The boy sees a house which is on the hill.
>
>
>
> I abbreviated nominative to nom and accusative to acc. As you can see, the
> square bracket affixes don't just tell us the word class but also the
> organisation of the sentence. The sentence ends as soon as there are an
> equal number of open and closed brackets.
>
>
>
> Here's a complex interrogative sentence:
>
>
>
> [query [state negative] [topic [nom boy] [acc [on hill] house] [because
> [nom it] dark] see] whether]
>
> - Does the boy not see the house which is on the hill because it is dark?
>
>
>
> This actually suggests that the house is dark. Sub indexes from the
> dictionary - it1, it2, it3 etc. - may be used to eliminate lexical
> ambiguity. Determiners such as 'a' and 'the' etc. can also be added in the
> right place. A most practical way is to use them as noun markers, but let's
> keep it simple.
>
>
>
> This language can express everything there are words or symbols available
> for; if you like, you can use it with pictograms or sign language. It has
> the simplest possible deterministic context-free grammar; a binary grammar
> and the simplest possible nonlinear grammar. This means that it is
> completely unambiguous as well as the simplest language that generates a
> conventional parse tree, and it could also be used as a mathematical or
> programming language. To me the language is particularly important because
> I'm using it to produce sentences in the fundamental language (see more
> below).
>
>
>
> Though it may seem weird at first, as long as it's used verb-final, it
> portrays only characteristics that are already familiar from natural
> languages. The O+S+V word order is however free.
>
>
>
> How is this different from AllNoun? Note that the square brackets I'm using
> are POS tags, not operators, and can be replaced with e.g. the article 'a'
> or 'the' for nouns only or the suffixes -PREP and -NOUN for each. As
> Frathwiki has it: "Contrary to its name, AllNoun does not in fact contain
> all nouns - - AllNoun uses four operators as part of its grammar." And
> LangMaker: "the creator admits that the conlang is no longer supported by
> him & that it has 'major failings.' It does not appear that the grammar is
> complete."
>
>
>
> However it is possible to construct an unambiguous language with all nouns
> with the regular grammar A → a. You can see how it works by copying the
> sentence [topic [nom boy] [acc [on hill] house] see] and pasting it to an
> online parse tree generator, such as http://ironcreek.net/phpsyntaxtree/
>
>
>
> The branches can be read one by one bottom-up with the semantic rule that
> each word modifies the one after it, or likewise that the latter one
> relates to the previous one. It can also be understood as an agglutinative
> language where the lines represent hyphens within clustered words. The DCFG
> sentence gives us four linear sentences that read:
>
>
>
> (1) boy-nom-topic
>
> (2) hill-on-acc-topic
>
> (3) house-acc-topic
>
> (4) see-topic
>
>
>
> (nominative = the form of a [word] when it is the subject[.])
>
> (accusative = the form of a [word] that shows that it is the direct
> object[.])
>
>
>
> Note that 'see' can be replaced with 'seeing' and 'on' can be replaced with
> 'top'.
>
>
>
> Let's double check the semantics:
>
>
>
> (1) There is a boy-nominative (the nominative is related to the boy) and a
> nominative-topic (the topic is related to the nominative)
>
> (2) There is a hill-on (the 'on' is related to the hill) and an
> on-accusative (the accusative is related to the 'on') …
>
> (3) There is a house-accusative (the house is related to the accusative)
> and an accusative-topic (the topic is related to the accusative)
>
> (4) There is a see-topic (the topic is related to the 'see')
>
>
>
> If we want to add punctuation, we can read it as a full sentence: "Boy
> nominative topic, hill on accusative topic, house accusative topic,
> see-topic." However I prefer to illustrate the full sentence as a mind-map
> possibly with the topic in the centre, but it also might make sense to use
> a different parse method that could for instance place the action in the
> centre.
>
>
>
> I changed my mind about uploading a linguistics article this time, but I
> think there will be one available sometime in September-October.
>



-- 
Jason Cullen
MA Applied English Linguistics

Top of Message | Previous Page | Permalink

Advanced Options


Options

Error during command authentication.

Error - unable to initiate communication with LISTSERV (errno=111). The server is probably not started.

Log In

Log In

Get Password

Get Password


Subscribe or Unsubscribe

Subscribe or Unsubscribe


Archives

ATOM RSS1 RSS2



LISTSERV.BROWN.EDU

CataList Email List Search Powered by the LISTSERV Email List Manager