0
   

English Defined as a Formal Grammar

 
 
stuh505
 
Reply Sat 7 Aug, 2004 08:25 pm
Hello, I'm looking for a complete formal grammar definition of the English language for a program I'm writing to communicate in English.

http://www.scientificpsychic.com/grammar/enggram1.html

This is the best website I could find, and the definitions I get from here are pretty good as you can see, but incomplete...I would like to find a more complete/official one.


<sentence> =
<simple_sentence> |
<compound_sentence>

<simple_sentence> =
<declarative_sentence> |
<interrogative_sentence> |
<imperative_sentence>

<compound_sentence> =
<simple_sentence> <conjunction> <simple_sentence> |
"Either" <declarative_sentence> "or" <declarative_sentence> |
"Either" <imperative_sentence> "or" <imperative_sentence>

<declarative_sentence> = <subject> <predicate>

<subject> = <simple_subject> | <compound_subject>

<simple_subject> = <noun_phrase> | <nominative_personal_pronoun>

<noun phrase> =
<"the"> <specific proper noun> |
<proper noun> |
<non-personal pronoun> |
<article> [<adverb>* <adjective>] <noun> |
[<adverb>* <adjective>] <noun-plural> |
<proper noun-possessive> [<adverb>* <adjective>] <noun> |
<personal possessive adjective> [<adverb>* <adjective>] <noun> |
<article> <common noun-possessive>
[<adverb>* <adjective>] <noun>

<compound subject> =
<simple subject> ("and" | "or") <simple subject> |
"Either" <simple subject> "or" <simple subject> |
"Neither" <simple subject> "nor" <simple subject>

<predicate> = (<verb> | <verb_phrase>) <complement>

<pronoun> = <demonstrative_pronoun> |
<interrogative_pronoun> |
<relative_pronoun> |
<indefinite_pronoun> |
<personal_pronoun>

<verb> = <V1s> |<V2s> |<V3s> |
<V1p> |<V2p> |<V3p> |
<Vpast> |<linking verb>

<linking verb> = "am" |"are" |"is" | "was"| "were" |
"look" | "looks" | "looked" |
"become" | "became" | "become" | ...

<verb phrase> =
("had" |"have" |"has") <Vpastp> |
("had" |"have" |"has") "been" [<Vpastp> | <Ving>] |
<auxV> "have" <Vpastp> |
<auxV> "have" "been" [<Vpastp> | <Ving>] |
<auxV> "be" [<Vpastp> | <Ving>] |
<auxV> <Vinf> |
"ought" "to" <Vinf> |
"ought" "to" "be" [<Vpastp> | <Ving>] |
"ought" "to" "have" <Vpastp> |
"ought" "to" "have" "been" [<Vpastp> | <Ving>] |
("do" |"does" |"did") [<Vinf>] |
("am" |"are" |"is" |"was" |"were") [<Vpastp> | <Ving>] |
("am" |"are" |"is" |"was" |"were") "being" [<Vpastp>] |
("am" |"are" |"is" |"was" |"were") "going" "to" [<Vinf>]

<complement> =
[[<indirect object>] <object>] |
[<adverb>* <adjective>] |
[<prep phr>*] |
["to" <Vinf> [<object>]] |
[<Ving>]

<indirect object> = <object> =
<simple object> | <compound object>

<simple object> = <noun phrase> |
<objective personal pronoun>

<compound object> =
<simple object> ("and" | "or") <simple object>

<Interrogative Sentence> =
<Declarative Sentence>"?"

<Imperative Sentence> = <verb> <complement>
  • Topic Stats
  • Top Replies
  • Link to this Topic
Type: Discussion • Score: 0 • Views: 3,417 • Replies: 18
No top replies

 
fresco
 
  1  
Reply Sun 8 Aug, 2004 01:39 pm
stuh505

From what I remember of linguistics, I believe "a complete formal grammar" could be an elusive ideal. Even the term "grammar" is subject to debate. I suggest you try Crystal "Cambridge Encyclopedia of Language" for further comment.

NB Even the Romans had problems with what we might think were "agreed rules" for Latin grammar and had to resort to abbreviations on inscriptions in order to avoid committing themselves to a particular word form.
0 Replies
 
Letty
 
  1  
Reply Sun 8 Aug, 2004 02:36 pm
Whatever you do, stuh. Don't try transformational grammar. Rolling Eyes
0 Replies
 
stuh505
 
  1  
Reply Sun 8 Aug, 2004 04:26 pm
Letty,

I have never heard of transformational grammar before, but upon looking it up, I have read that it is simply a means of describing the comprehension of the semantics of grammar...designed by, coincidentally, Noam Chomsky, who incidentally is also the designer of the Chomsky Hierarchy which defines formal grammar's into 4 basic levels of complexity. His nomenclature is still the standard used in computer science today. Anyway, his concept of transformational grammar does not seem to be a very bold assertion, and I certainly agree with it. What prompted you to make the comment?

fresco:

Perhaps I will not be able to find a complete formal grammar for English, but it is undeniable that English can be defined completely as a formal grammar...because no matter how confusing, contorted, and complex English may be...there is no limit to the complexity that can be represented in a formal grammar.

Furthermore, in order to teach the language to my program, I must define English formally as I have started to do above...so it is not really a question of CAN I, but HOW WELL can I. Obviously, I would like it to be perfect.

If anyone knows of any good informative web sites (please do not refer me to books, I haven't the time nor the money) which discuss the formulation of syntax in detail, please let me know. They do not need to define it in terms of the formal syntax, I am capable of writing that myself given that their language is unambiguous.
0 Replies
 
Letty
 
  1  
Reply Sun 8 Aug, 2004 04:39 pm
Because, stuh. Transformational grammar is just smoke and mirrors. Trees of derivation far outweigh the clumsiness of diagramming. Even the most touted of linguists are tempted to reinvent without any particular justification. Chomsky is no exception.
0 Replies
 
stuh505
 
  1  
Reply Sun 8 Aug, 2004 04:51 pm
Letty,

transformational grammar is merely a theory that states that there is more than one grammatical way to say the same thing. one can hardly argue with that. at least, that's the definition I have been reading...

Quote:
0 Replies
 
Letty
 
  1  
Reply Sun 8 Aug, 2004 04:59 pm
stuh, grammar is just so boring, ya know? It doesn't matter which way you analyze it or parse it, it's math in sheep's clothing.
0 Replies
 
stuh505
 
  1  
Reply Sun 8 Aug, 2004 05:23 pm
grammar is of course arbitrary, and in itself boring...but writing a computer program that can understand it...I don't think that's boring!

math and science, on the other hand, can be used to model and predict just about everything in the living world!

and my friend, there is nothing like the satisfaction one gets from accurately modelling real world phenomena. true understanding is only achieved when you're able to recreate the event. and that kind of understanding...is like...going from being a user to a creator.

Maybe I'm just different...

maybe I'm the only one who, to choose a bullet, instead of looking at charts and listening to experts, attempts to solve never-before solved differential equations in order to compute a function for bullet velocity

maybe I'm the only one who, when trying to program lifecycles into organisms, looks for an equation to determine how much oxygen to the brain is necessary for survival and determine precisely how much blood loss will result in unconsciousness

maybe I'm the only one who, when trying to write a program to communicate in English, attempts to define the whole damn language

maybe I'm the only one who attempts to recreate the big bang in computer simulations

maybe I'm the only one...
0 Replies
 
Letty
 
  1  
Reply Sun 8 Aug, 2004 05:27 pm
No, honey. You are NOT the only one. You translate your life into production, I translate mine into poetry with an occasional bit of fluff thrown in. Fresco is busy trying to educate his charges, and in between, he takes a holiday. <smile>
0 Replies
 
fresco
 
  1  
Reply Mon 9 Aug, 2004 01:35 am
stuh

A key issue in "grammar" is the status of "sentence". Halliday's grammar for example takes "discourse" as the level of analysis with regard to the fact that we rarely "speak" in sentences. The significance of the a priori unit of analysis is relative to the task in hand. It may be that your own objective puts artificial contraints on the formal system used.

(Hi Letty...the Cornish ferryman sends his regards !)
0 Replies
 
stuh505
 
  1  
Reply Mon 9 Aug, 2004 11:11 am
fresco,

My intention is not to make a "perfect" program, that is capable of understanding every sentence or sentence fragment.

I certainly cannot understand every thing that people say to me, I don't expect the computer program to do so either!

But I CAN program in all of the technically correct sentence structures...as well as many of the technically incorrect slang sentence structures.
0 Replies
 
Letty
 
  1  
Reply Mon 9 Aug, 2004 11:34 am
stuh, there is also an approach to grammar called Structural Grammar. It is quite simple as opposed to the other two. It purports that all English sentences are composed of sentence patterns.

Subject-Verb

Subject-Verb-complement.

The rest are simply expansions of these two basic structures.

Although I know little of computer programming, I would think that his would be easy and informative at the same time.

(fresco, just glad your ferryman wasn't the one of the river Styx)
0 Replies
 
stuh505
 
  1  
Reply Mon 9 Aug, 2004 02:41 pm
Letty,

the format you have described is really no different than the format I am using.

<declarative_sentence> = <subject> <predicate>
<predicate> = (<verb> | <verb_phrase>) <complement>

as you can see, the form which you have proposed is just a small subset of what i already have

one thing i would like to do is classify the individual words into groups so that it is more versatile. for instance, by replacing where it says "the" with <article> so that any article could work there

the, a, an, my, her, his = article ( i think)
and, or = ?
either, neither = ?
0 Replies
 
Letty
 
  1  
Reply Mon 9 Aug, 2004 04:04 pm
ok, stuh. Let's see if I can recall:

a,an,the--definite articles
my, his, her--possessive pronouns functioning as articles.

and, or, (connectors) conjunctions connecting independent clauses or nouns/pronouns.

not only, but also. correlative conjunctions (same function)
either or, neither nor...same as not only, but also.

because, introduces a dependent clause. (subordinate conjunction?)

As I recall in transformational grammar...one of the was called a pre-article. The problem there was agreement.
0 Replies
 
fresco
 
  1  
Reply Tue 10 Aug, 2004 01:41 am
stuh

A couple of issues I remember which may have some bearing.

Transformational grammar (et al) indicates the importance of context in resolving categorizations, hence the move to "deep structure".

Consider e.g.

John is easy to understand/John is eager to understand,

Can easy/eager be differentially categorized from "the surface" alone? ....or alternatively is "John" functioning as "object" in the first and "subject" in the second ?

It is sometimes worth considering a context specific a subset of "the language" to highlight the issues...e.g. ATC (Air Traffic Control). It is significant that little attempt has been made to automate verbal interface in even this relatively closed contextual system.
0 Replies
 
stuh505
 
  1  
Reply Tue 10 Aug, 2004 11:29 am
seems to work from the same context to me:

John.EasyToUnderstand = true
John.EagerToUnderstand = true

if there were two kinds of words that in this situation didn't work, I would just make two different kinds of sentences and specify which kind of word goes in each. i dont see how this is a problem

are you suggestign that, since "little attempt" in your opinion has been made to do this...I should not attempt?
0 Replies
 
fresco
 
  1  
Reply Tue 10 Aug, 2004 01:36 pm
No...attempt by all means...but your concepts of "English" and "official" seem to imply that there is some authority in these areas. In fact there are many different varieties of "English" both for native and non-native speakers according to social context. These are loosely defined with many "rules" merely a matter of convention.
To take a simple example The US sentence ..."Did he wash the car yet"... is ungrammatical in UK English (at present !)

In the extreme, I believe the problem for learners of languages like Japanese is exacerbated by the lack of appreciation of social nuances which completely govern appropriate word choice.
In short "grammar" may always be situational to some degree.
0 Replies
 
stuh505
 
  1  
Reply Tue 10 Aug, 2004 02:48 pm
fresco,

ah yes, languages do evolve. but language for any particular time period and region does have recognized rules which can be expressed formally. when I say English, I refer to present day American English.
0 Replies
 
zml
 
  1  
Reply Wed 15 Sep, 2010 05:33 am
@stuh505,
Hi, Stuh505..

My name is Zaw Min Lwin. I live in Myanmar. Nice to meet you!
I am also interesting in the topic you post. If you understand and get complete rules of the English Sentence structure, please may i know that information to me. I am also searching the complete CFG productions rules for english sentence structure. I am writing a grammar checker for english sentence and generate parset tree that represent the input english sentence.
My email is:
[email protected]
0 Replies
 
 

Related Topics

deal - Question by WBYeats
Let pupils abandon spelling rules, says academic - Discussion by Robert Gentel
Please, I need help. - Question by imsak
Is this sentence grammatically correct? - Question by Sydney-Strock
"come from" - Question by mcook
concentrated - Question by WBYeats
 
  1. Forums
  2. » English Defined as a Formal Grammar
Copyright © 2024 MadLab, LLC :: Terms of Service :: Privacy Policy :: Page generated in 0.03 seconds on 04/28/2024 at 12:05:43