Context free grammar

[Cutler4Life]

I learned about this subject when taking computer sciences a while back. Simply, it uses parse trees to explore the different structures of sentence composition. For example, many sentences can have different meanings depending on where the emphasis is placed. It broke sentence structure (S: NP: VP: N: adj PP: a P, etc.) It is very interesting, but I have only been exposed to this in computer science.

          S

          |

          /|\

        S - S

        /      \

      /|\    /|\

      S * S  S / S

    /    |  |      \

    /|\  x /|\  / | \

  ( S )  S * S ( S )

    /      |  |    \ 

  /|\    z  y  /|\

  S + S          S + S

  |  |          |  |

  x  y          x  x

Has anyone else seen this before?

[Kristi]

We had to draw similar trees in linguistics classes. In English syntax to be more precise. You basically have to break down a sentence into individual elements.

[linguaholic]

yes, I had to do this as well in linguistics.

This is also very crucial in computational linguistics...as the computer programs need to parse every sentence..so of course all the "blocks" need to be split in specific classes like nouns, verbs, noun phrases, verb phrases and so on. Of course there are always different possibilities how to parse a sentence / text. However, in computational linguistics you often already have predefined tag sets that you can use to parse texts. In Europe we use a lot the STTS ( Stuttgart-Tübinger-Tagset).

[BWL]

There are two languages that I know of that can be easily parsed into a syntactic tree; Japanese and Tagalog.

Japanese syntax is structured around the verb which occupies the last slot in a sentence (not counting the sentence final particles like "yo", "zo", "ze", "ka", :kana", etc. A typical sentence looks like this (I removed the honorifics and sentence particles to simplify things a bit)

Otoko ga kodomo ni hon o ageta. "The man gave a book to the child".

Otoko    ga        kodomo                ni                    hon              o                              ageta

(MAN)  (actor)  (CHILD)    (receiver or goal)      (BOOK)  (object of action)              (GAVE)

The little words or particles (ga, ni, o) that occur after nouns in a Japanese sentence are the syntactic markers, they are the strings that connect the nouns to the head or top of the syntactic tree, which happens to be the verb "ageta" (GAVE).

Tagalog has a similar structure but in reverse, with the verb occurring at the head in front and particles before the nouns!:

Binigay ng lalaki ang libro sa bata. (The man gave the book to a boy)

Binigay          ng            lalaki            ang              libro                sa                    bata

(GAVE)            (actor)        (MAN)        (object)        (BOOK)            (goal)                (CHILD)

Tagalog is much more complex because the verbs, particles and sentence structure change depending on the focus, whether the actor or the object or the goal is definite or indefinite (having either "the" or "a")

Binigay ng lalaki ang libro sa bata (The man gave the book to a boy)

Bumigay ng libro ang lalaki sa bata (The man gave a book to a boy)

Binigyan ng libro ng lalaki ang bata (A man gave a book to the boy)

Tagalog and other Philippine languages are very complex, very different from Chinese or European languages that most people here are familiar with. But yes, they do fit into a syntactic tree structure.

[Elienc]

I never took any computational linguistics classes, but in Dutch Grammar, we had to analyse sentences in similar fashion, I think each language has it's own system for doing it. The Dutch system seems very different from Japanese as you describe here. £I think it helps a great deal in understanding the grammar system of a language. In Latin you have to do it all the time, only to simply understand it/translate it. We never did it context-free though.

[mareebaybay]

I will definitely have to check this out soon.