the meta-model

This meta-model is (or at least is becoming) my technical interpretation of the NLP Meta-Model.. Richard Bandler made the original up. I hope that he considers Sabrina a sort of paid advertisement for his efforts, and not some kind of copyright infringement. (you can buy Bandler's books if you want!)

Note: my [programming notes] are sketchy at best.. These are just rough ideas I had detailing SOME of what needs to be done. This file is a half-breed.. It's human-readable on its way to becoming compiler readable.. :)

further note: Just because Bandler and Grinder divided things up this way, doesn't mean we have to. Parts of this could go into the parser, parts into the neural net, etc... Perhaps we could pop meta-model questions onto a stack of some kind, and let the AI stuff make guesses or ask questions, depending on how strong it is..

1. deletion
   1. are there verbs in sentence with missing arguments?
      • passive voice
      • i'm (verb - running, etc.)
      • i feel xyz...
      • perhaps special parsing for non E' sentences?
      • -ing words
      • note some sentences have 2 verbs <-like that
   2. adjectives - the bored man said hi. (bored about what?) (said hi to whom?)
   3. comparatives - more xyz, faster/better/smarter - than what?
   4. suprelatives - best most worst - compared to what?
   5. clearly/obviously - to whom? (p.68 of StrucI - "it is" test)
   6. Modal operators of necessity - have to, must, etc (what would happen..?)
   7. Modal ops of possibility - impossible, can't, may not (what prevents?)
2. distortion - nominalizations
   1. "listen to the surface structure presented by the client"
   2. for each word { ! verb/process && (actually describe an event/process?) || (is there some event/process word like it?) }
   3. does it fit: an ongoing __________...?
      [maybe this is a case for neural nets or markov chains.. or just a database of which nouns are nominalizations and which are not... and speaking of that, we can simplify the dictionary if we just don't allow words to be verbs or nouns.. treat them as nominalizations, and clarify them... i wonder if that would work..]
3. generalization
   1. Referential Indices
      1. for each word in sentence{ ! process/verb && (does it pick out a specific person, place, or thing?)}
      2. Responses: who, specifically? what, specifically?
   2. Special case: universal quantifiers
      1. XYZ is always true.
         1. Is this always the case?
            You mean NOBODY can EVER do that AT ALL?
            (and similar, depending on the quantifier)
         2. Have you ever not XYZ?
         3. Could you imagine a situation in which not XYZ is true?
      2. Can't X / X is Impossible
         1. What prevents X?
         2. What would happen if X?
      3. Symmetrical Predicates
         • Person1 always/never action person2
         • symmetrical in that converse has to be true..
         • swap Persons & ask as question:
         • Barney always fights with me --> Do you always fight with barney?
           [if we know a verb is vt, we're halfway to this.. just add an extra flag to it.. symmetrical or not.. fight,argue,talk,walk,dance.. also, regexp: /with/ ]
      4. Non-Symmetrical Predicates
         • converse doesn't logically have to be true, but ask the reverse anyway, because it's often true anyway.. (projection)
         • Barney hates me --> do you hate barney?
           [regexp meta /[vt]/ ]
   3. Complex Generalization -> Equivalance
   4. Incompletely specified verbs
4. Presuppositions, Syntactic environments for:
   1. simple (existence) presups
      [we don't want to talk about ALL of these, as some are too obvious, but it's good to be able to talk about them if necessary ... like if she's teaching meta-model or analyzing text... so these may be low priority (except for in story-telling, as these lead directly to adding in anti-math symbols)]
      1. proper names
         bob kicked me --> there's someone named bob out there..
         [list of baby names... gender.pm gives gender...]
      2. pronouns /\b(he|she|him|her|they|them)\b/
         I kicked him --> there's a guy out there
         [simple pattern match]
      3. definite descriptions
         I kicked the man in the barney suit --> there was a man in a barney suit
         [probably would have to parse out noun phrases..]
      4. generic noun phrases meta: /plural noun/
         Barneys are all alike --> there are barneys.
         [this requires knowing parts of speech/plurality]
      5. some quantifiers /\b(all|every|each|some|many|few|none|any)\b/
         If any of the barneys show up, I'll kick them --> there are barneys.
         [simple pattern match..?]
   2. complex presupps
      1. relative clauses /the [n] (who|which|that) [v]/
         The barney that wore the plaid hat kicked me --> A barney kicked me.
         [matching who|witch|that could cue more in-depth parse..]
      2. subordinate clauses of time /(after|as|before|duringprior|while|when)/
         The bug bit me while i was sleeping --> I was sleeping. (also: The bug bit me. (might not technically be a presupp, but it would be silly to miss this.))
         [regexp & return the clause(s)..]
      3. cleft sentence /it was (.*) (that|which|who)/i
         It was barney who ate my cat --> someone ate my cat.
         [regexp for the voice, +??]
      4. psuedo-cleft sentences /what (.*) is (.*)/
         What barney plans to do is dominate the world -> Barney plans to do something.
         [regexp +??]
      5. stressed sentences /if .* (emphasized noun)/
         If Barney killed MIGHTY MOUSE, the end is near --> Barney killed someone.
         [we could invent AML, the analogue markup language! :)]
      6. complex adjectives /new|old|former|present|previous/
         My new barney suit doesn't fit --> I have or had an old barney suit. (this one seems a little iffy..)
         [regexp]
      7. ordinal numbers /first|second|third...|another/
         When I skinned the third barney, i was in for a surprise. --> I'd already skinned two barneys.
         [regexp... might fire a rule if string matches -th or -nd or -rst or -rd]
      8. comparatives /\b(\w*er|more|less)\b/
         Moe is a better barney than Fred is. --> Fred is a barney.
         [regexp.. this will also fire off in main meta-model]
      9. comparative as /as .* as/
         Moe isn't as purple as Fred --> Fred is purple.
         [regexp]
      10. repetive cue words /too|also|either|again|back/
          I wouldn't tell barney that, either. --> someone has told barney that before.
          [regexp +??]
      11. repetetive verbs and adverbs /re-[v/adv]/
          If barney repeats the magic word, kill him --> barney has said the magic word before.
          [in this example, we'd have to know that repeat is a form of said... but it could also mean did or went through, etc (barney repeated the third grade)]
      12. qualifiers /only|even|except|just/
          Even Barney can learn Perl --> barney can learn Perl.
          [regexp +??]
      13. change of place verbs /come|go|leave|depart|enter|arrive/
          If barney arrived in town, there was probably trouble. --> barney was going to town
          [regexp +??]
      14. change of time verbs and adverbs /begin|end|stop|start|continue/
          If you stop barney-bashing, i'll let you live --> You have been barney bashing.
          [if you end your barney-bashing --> barney bashing has become a noun! ([possesive pro/noun] [v]) is another nominalization pattern..)]
      15. change of state verbs /change|transform|turn into|become/
          If mighty mouse becomes a barney, I'll kill him myself. --> mighty mouse is not now a barney.
          []
      16. factive verbs and adjectives /|odd|aware|know|realize|regret/
          It's odd that barney has lasted this long -> barney has lasted this long. ..
          We all regret that barney survived -> barney survived.
          [regexp +??]
      17. commentary adjectives and adverbs /lucky|fortunate|neat|cool|innocently|obviously|..../
          It's so cool that you wear barney underroos --> you wear barney underroos.
          [perhaps this should be done in the abstractor/parser]
      18. counterfactual conditional clauses (subjuncive)
          If I were a rich man, I wouldn't have to work hard --> I am not a rich man.
          [many people will botch the grammar and say "if I was a rich man", so this should be fuzzy]
      19. contrary-to-expection /should|happen to|decide|after all/
          Should any of your force be caught or killed... --> I doubt you'll be caught or killed.
          [regexp +??]
      20. selectional restrictions
          If barney becomes pregnant -> barney is a woman
          Joe saw the mountains flying over france. -> joe flew over france.
          Joe saw the geese flying over the mountain -> geese flew over the mountain.
          [this is a big one... the last two examples were inspired by the docs at www.cyc.com.. we may have to just let it slide for now.]
      21. questions /who|what|when|where/
          who killed mr body? --> proffessor plumb, in the lounge, with the candlestick
          oh, wait:
          who killed barney? --> someone killed barney.
          [who-->someone, what-->something, when-->sometime where-->somewhere]
      22. negative questions
          Didn't the wise man once say, "eat at moe's?" --> i thought the wise man once said, "eat at moe's".
          [regexp + reconjugating?]
      23. rhetorical questions
          Who cares what barney thinks? --> no one cares what barney thinks.
          [.. and just how do we know if somethings rhetorical? .. she could ask.. :) ]
      24. spurious not
          You don't think you're switching to the dark side? --> you're switching to the dark side.
          b/g's example was: "I wonder if you're not being a little unfair --> i think that you're being unfair"]
          [maybe just remove the not? nahh...]
5. Semantic Well-Formedness
6. Well-Formed in Therapy