Linguistics

See also: [
LITERATURE INDEX] BUT: [Language] (Philo DIR) [Languages] (i swear i was sobre when i set these files up!! :(

Linguistics

On this page: {Definitions, etc} {Susumu Kuno's article (extract only) {Briefing on Attention, Intentions, & Structure of Discourse} {The Usual Suspects} Definitions

Definitions


Briefing on Attention, Intentions, & Structure of Discourse

From {
Kuno's ref}, text Pp.3&ff. by Barbara J. Gorsz (SRI, Menlo, Calif) & Candace L. Sidner (BBN Labs, Cambridge). P.3

2.The Basic Theory

BEGIN BLOCK QUOTE Discourse structure is a coposite of 3 inter-acting constinutiends: A linguistic structure, An intentional structure An attention state. THese 3 consitutents of discourse sturcre deal with different aspects of the utterences in a dicourse. Utterance -- the actual saying or *writing* [tips towel to Ms. G and Ms. S] of particualr sequences of phrase and clauses -- are the lingusit structure's basic elements. Intentions -- of a particular sort and a small number of relationships between them proovide the basic elements of intentional sturcutre.. Attetnional state -- contains information about the objects, propertires, relations, and discourse intetions that are most salient at any given point. [in time, i assume?] It is an abstraction of of the focus of attention of the discourse participlants; it serves to summarise the information from previous utteraances crucuisal for processing subsequent ones [Note 1], thus obviating the need for keeping a complete history of the discourse. (again, continue with NOTE-1) END BLOCK QUOTE NOTES (This section only) [1] Hmmm. This looks like a STATE thingie to me. We could use a bingo card for part of it if we are to "measure" or at least "lay out" the POINT of the discourse (eg, speech to inform, convince, etc). As to the EXACT state then we have the problem (once again) of PRONOUNS and all of that; ie, "in-process" references to things which changes over time. And since we are NOT numbering our pronouns, we will occasionally be REQUIRED to re-identify what we mean. For example, if several of the objects in the discourse are of the same gender, we will have to clarify what we mean (after a while) by "he" or "she" or "it", etc. Best example is ["Eats, Shoots and Leaves"] *** This goes back to "logging" the information. Which is of course refered to in "2010" viz the Russian scientis's comment "But, the 9000 series uses holographic memory, so simple erasure would not work", and the somewhat naive reply is "I used a tape worm; it hunts down specific memories". This problem can not easily be solved (if it can be at all). The problem is that with reality-based structures, you simply end up with GAPS in the record, as well pointers to non-existent things. In the case of time-travel (eg, the FOLD in "Babylon-Squared" -- see below), the problem creates (potentially) paradoxes that can not be "un-done" (similar to the "Veils of Azlaroc" by SaberHaggen ??sp??. Thus, at best, the meaning becomes complicated and possibly impossible to follow. Even if we allow for Dadaist or Hermetical things, the problem simply devolves to either very serious silliness or at the very least muddled communications that is *completely* dismissed by much of the audience (listeners/viewers/readers/players). The FOLD problem (briefly) {Back to the TEXT} Next: Susumu Kuno's article. {Back to the TOP of this page}

Susumu Kuno's Article

Extracted from "Language and Aritificial Intellegence", ed. by M. Nagao, (Proceedings of an International Symposium on Language and Artificial Intelligence held in Kyoto, 1986.03.16 thru 1986.03.21), full text is pages 87-111. Extracts as noted. P.87. NOTE: Kuno's use of SUB-SCRIPTS has been changed here to the form '# for simplified entry. (We appologise for the in-conveninece) Also paragraph and spelling have been altered to be consistent with current practice.

Anaphora and Discourse Principles

1. Discourse principles. Every major syntactic pattern of a language requires two kinds of characteristics - a forumulation of syntactic conditions that need to be full-filled for the pattern to be used, as well as a forumulation of semantic and discourse -- based on conditions that need to be fulll-filled at the same time. The pronoun-antecedent relationship is no exception to this requirement. In order to understand how pronounss work in a language, elucidating the semantic discourse conditions that control them is as important as explainging the structural relations that must hold between the pronoun and the antecedent. Unfortunately, the former taks has not received enough attention. ... (1) Predictability Requirement for Pronouns.
A pronounc can proceed a co-referential NP only if the speaker can legitimately [Note 1] assume that the hearer can determine its referent from the *preceeding* context. [KUNO 1972] REFERENCES (this section only) [KUNO 1972] "Functional Sentence Perspective - A case study from Japanese and English", Linguistic Inquiry 3 Pp.269-320. {Back to the TEXT} NOTES (this section only) [1] The act of "legitimisation" begs the question of the knowing the audience, and is *possibly* the most common source of mis-understanding. Also, it seems to me that this would also be a source of humor; esp, if the use of he/him/her/its etc is mis-used for a given object. My most often proffered example is using "she" when refering to "God" (see map). {Back to the TEXT} [] {Back to the TEXT} [] {Back to the TEXT} [] {Back to the TEXT} [] {Back to the TEXT} [] {Back to the TEXT} [] {Back to the TEXT} [] {Back to the TEXT} [] {Back to the TEXT} Next: . {Back to the TOP of this page}

Next: . {Back to the TOP of this page}

Next: . {Back to the TOP of this page}

Next: . {Back to the TOP of this page}

Next: . {Back to the TOP of this page}

Next: . {Back to the TOP of this page}

Next: . {Back to the TOP of this page}

Next: . {Back to the TOP of this page}

Next: . {Back to the TOP of this page}

Next: . {Back to the TOP of this page}

Next: . {Back to the TOP of this page}

Next: . {Back to the TOP of this page}

Next: . {Back to the TOP of this page}

Next: . {Back to the TOP of this page}