[Corpora-List] Lakoff on Fillmore

[Yorick Wilks]

John
It is sad though expected: he was the nicest of men, not common among academics, and a massive intellectual figure.
My line always was that he was the only linguist (except possibly Halliday) whose influence could be clearly seen in NLP--or at the least what he was 
doing closely paralleled what was going on in NLP--though I suppose we must now include Lakoff for his influence on recent iARPA support for metaphor computation
(though I happen personally to think that influence wholly misguided, but thats another matter). One trouble in the long relationship of
linguistics and NLP--as you know as well as me----is that linguists tend to think ideas only count when they have thought them. Though CJF himself had no arrogance or partisanship at all and was much closer to Wittgenstein's view that it didnt matter too much who had thought  what when (I paraphrase).

In the case of frames it is obvious that Minsky was pushing from MIT a view of frame representation by 1972 or so, with direct consequences for meaning representation, and following as he noted Bartlett's work of the 30s, long before any linguist used the word for anything other than a simple syntactic substitution context (as in the work of Zelig Harris and others of the "anthropological" school) a matter that had nothing to do with what came to be called "frame semantics". Lakoff's account you cite is just wrong in many of its emphases as when he wrote:

 "Chuck's classic example involved the semantic field buy-sell-goods-price-cost. The common mental structure defining such words is based on the commercial event scenario: Person 1 has possessions and wants to exchange them for money. Person 2 has the money and wants to exchange it for such a possession. There is mutual exchange. Person 1 is called a seller; Person 2 is called a buyer; the possession exchanged is called the goods; and the money is called the price. Those named the basic "semantic roles" -- the conceptual elements of the frame."

This quote refers to 1974-75 but Roger Schank has been using precisely this scenario to make precisely these claims about meaning structure at Stanford (30 miles away!) since 1969. Schank and others at Stanford  I cannot mention were independently using what were plainly case notions in meaning representation since 1965 or so. It is unclear whether those owed anything to knowledge of CJFs work or not--I suspect not--it was simply same idea/different places, as so often in such matters---one (CJFs) was tightly tied to a critique of the Chomskyan linguistic  paradigm of the time and the other (NLP) trend not: it was was just generally anti-Chomsyan and pro-semantics and computation. When Lakoff writes:

"we were regularly visited in my living room by three friends who drove over from Palo Alto -- Terry Winograd, Danny Bobrow, and Don Norman. They wanted to find out what they could about the details of frame semantics since they were working on a knowledge representation language for computer science, which eventually developed into KL-ONE -- a classic frame-based knowledge representation language in computer science. It was because of Chuck that it came to be "frame-based."

it is just as sign of how selective Berkeley's knowledge of Stanford was at the time--had Lakoff got in his car and driven back the other way he would have seen structures and claims that would have been very familiar. But, as always, it is impossible for him to believe that things COME FROM computing and NLP--in his universe AI/NLP people drive over and get the good stuff at his place.
Yorick




On 21 Feb 2014, at 21:26, John F Sowa wrote:

> It's sad to hear that Fillmore died, and the following web page
> has a good review of his work:
> 
>   http://linguistics.berkeley.edu/charles-j-fillmore-1929-2014
> 
> Re Lakoff's summary:  the only points I disagree with are George's
> constant claims that "cognitive science is a young field" in which
> all the major discoveries occurred in the past 40 years.
> 
> Minsky also had something to say about frames, and his 1974 article
> cited a lot of earlier work:
> 
>   http://web.media.mit.edu/~minsky/papers/Frames/frames.html
> 
> I cited earlier work in my article on semantic networks:
> 
>   http://www.jfsowa.com/pubs/semnet.htm
> 
> I have a very high regard for Fillmore's work, but it's important
> to put all the developments in perspective.  When you shorten the
> perspective to 40 years, it makes progress look much faster than
> it really is.  I believe we have a long way to go before the
> "singularity" when we can download our tired old brains into
> the computer.
> 
> John
> 
> _______________________________________________
> UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
> Corpora mailing list
> Corpora at uib.no
> http://mailman.uib.no/listinfo/corpora

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/corpora/attachments/20140222/00c979a6/attachment.htm>
-------------- next part --------------
_______________________________________________
UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora