Re: Science fiction's elves

Wayne Throop wrote:
: "David Tate" <dtate@xxxxxxx>
: I'm not sure how you can have magic acting on its own in DWIM, since
: you need an 'i' to be 'M'ing. But I think your point is valid, to the
: extent that Wayne's DWIM needs to be modified to "do what I mean when I
: say...", for values of 'say' that might not be said out loud.

Well sure, but that's what DWIM means in the first place.
The "when I say X" is so strongly implied that I'm mildly
bemused anybody couldn't hear it.

I don't think it's that clear. In fact, isn't this exactly the
difference between magic and psi that you were suggesting? Psi is DWIM
without the "...when I say X" part. As I understand what you've said
earlier, this means that psi (if it existed) *cannot* be magic, because
it lacks the symbolic reference part. So psi, if it existed, would
have to be mechanism -- which makes psi *extremely* implausible, given
that it ends up as DWIM.

Also, not everyone believes that only language can express meaning. Do
dogs have any understanding of the world that would count as 'meaning'?
Can they mentally assert their desires in distinguishable ways? If
either of those is true, then it would seem that dogs could potentially
do DWIM, but not the "...when I say X" part. You could have a psi dog,
but not a mage dog.

: But the nature of language makes "do what I say (words only)"
: impossible as a magic scheme, because the vast majority of what we say
: is not in the words themselves.

Yes, exactly so. But in particular, typical magical examples leave out
many orders of magnitude more than the actual content of what is said.
And the more that gets built implicitly into the technology, so that the
more you leave out of what you say, it becomes indistinguishable from
magic.

If I understand you correctly, you're talking about gadgets that have
built in the 'pragmatics' of the situation into the gadget, so that to
an untutored observer it *looks* like DWIM-via-symbolism. (I just had
a flash of Jacquard-loom-as-golem; fun idea.)

: You can get around some of that if your magic system requires that
: spells be uttered in True Speech (or equivalent), which is of course
: wholly unambiguous. But even then, I don't think you can avoid
: pragmatics (if I have the right term), the parts of language that are
: understood by the speakers rather than explicit in the utterances.

Right. "Poof, you're a frog" (or a dragon, or whatever) simply
don't contain enough information to describe what actually occurs,
whether the language is the "True Speech" or not.

There's an information-theory limit on this that is eluding me at the
moment, that might have to do with the objective difference between
mechanism and magic; I'll have to think more about it.

:: J Moreno
:: And here you go and do just what Clarke is warning against -- you
:: look at something, label it magic, and then stop thinking about it.

Actually, that's precisely the opposite of what I've been saying. I'm
saying that if there's nothing more there to think about below the level
of symbolism (ie, if it's taken as undecomposable), then you've got
magic. And again, this is the opposite of labeling it magic first, and
then not looking.

Hmm. Is the atomic nature of the symbolism required? I'm thinking
here about the language/script in "Story of Your Life" (or, to a lesser
extent, _So You Want to Be a Wizard_), which seems to be agglutinative.
I suppose at some level each 'morpheme' of the language has to be
unanalyzable in terms of mechanism, which is enough.

One way to look at it is that technology is bottom-up, and
magic is top-down. Technology is mechanism, and magic is meaning.
They meet in the middle, and become indistinguishable.

If they meet at all. ;-)

The remaining bit of distinguishment is that in magical descriptions
of the world (and simplifying a bit) symbolism is normally taken to be
objective, and undecomposable.

Is that necessary? Could symbolism be subjective, but tied to the
semantic framework of the operative intelligence? (Oh, cool, you could
have the mage equivalent of Sapir-Whorf, wherein the language and
thought patterns of the mage affect the nature of that mage's magic...)

You "direct by symbols". But direct *how*? In magic, the how is a
given, it's the basic nature of the world.

Or the subjective symbolism of the mind doing the magic. (An insane
magician might be very powerful, and very dangerous...)

In technology, the how is a mechanism. The basic nature of mechanism
is many times simpler than that of symbolism.

Is it asymmetric? Could you say rather[1] that the nature of mechanism
is extremely complicated (or possibly indescribable) if you try to
describe it symbolically, whereas symbolism is extremely complicated
(or possibly indescribable) if you try to describe it
mechanistically...?

David Tate

[1] In the sense of "...and have a coherent system"

.

Relevant Pages

  • Re: Science fictions elves
    ... extent that Wayne's DWIM needs to be modified to "do what I mean when I ... And the more that gets built implicitly into the technology, ... pragmatics, the parts of language that are ... of the world symbolism is normally taken to be ...
    (rec.arts.sf.written)
  • Re: Science fictions elves
    ... Psi is DWIM without the "...when I say X" part. ... I think that's a reasonable rephrasing of my intended meaning. ... :: of the world symbolism is normally taken to ...
    (rec.arts.sf.written)
  • Re: Is this why any discussion of PELLEAS... is pointless?
    ... this Belgian playwright, poet, essayist an beekkeeper has faded from the ... Funny...but his first plays were written for performances by marionettes fearing ... and that only populations familiar with the language ... I think that stagecraft was the vehicle for much of his 'symbolism' -- ...
    (rec.music.opera)
  • Re: Science fictions elves
    ... Wayne already replied to this, but I'll add a couple of comments... ... extent that Wayne's DWIM needs to be modified to "do what I mean when I ... whether the language is the "True Speech" or not. ...
    (rec.arts.sf.written)
  • Re: A definition of two (2) please?
    ... How would you formulate 1), ... whereby logic is itself a language. ... apple x and apple y ... While avoiding natural language by using symbolism can shorten the time ...
    (sci.logic)