Re: XML 1.0 is simple. was: RE: almost four years ago....

["W. E. Perry" <wperry@fiduciary.com>]

Now this is an interesting Saturday morning. I haven't seen some of these
examples trotted out since I taught Greek rhetoric 25 years ago. One of the most
interesting bits of this discussion is how closely it corresponds to the points
Al Snell is making in the parallel 'four years ago . . .' thread. Historically,
and to the extent that we understand it as a purely rhetorical device,
synecdoche was intended to shape the *perception* of the whole by the
characteristics of the part, or, very occasionally, vice versa. That
interpretation is a reasonable inference from Isocrates' explanation in the
earliest discussion which we have of this rhetorical device. That usage would
seem to correspond to what I read as Simon's suspicion that some would like to
shape the perception of the whole of their 'greater XML' by the perceived
virtues of a much smaller core (or perhaps shape the perception of their
particular parts of XML by the virtues of what was once a much pithier whole).

Unfortunately for our discussion (and remarkably parallel to it), Aristotle, the
father of object-oriented thinking, weighed in on this rhetorical device early
on, and his interpretation of the transference which Len's examples illustrate
has been the accepted orthodoxy ever since. It is significant that Aristotle did
not use the word 'metonymy', which as simple Greek would have meant no more than
'calling something by a different name', and which would have restricted his
point about transference to purely lexical relationships. Aristotle used (in
fact, coined in this sense) the much broader term 'metaphor', our understanding
of which in this usage is fundamentally dependent upon the Aristotelian taxonomy
of genus and species. That is, metaphor in the Aristotelian sense--which
subsumes synecdoche, previously understood as a purely rhetorical device--is
specifically about the inheritance of properties by species from genus; about
the structural similarity of a genus and its species, and the extent to which
the differences can be described by the presence or absence of attributes; and
about the relationships of taxonomic siblings based on a fundamental structural
identity.

So now we verbose types are left to defend the purely rhetorical against the
all-embracing taxonomic explanation. The word 'rhetoric' is rooted in the term
for a public performer of poetry. Most simply understood, rhetorical devices are
about the actual mechanics of performance--the syntax, if you will, of the
instance. Clearly there are those of us who find it useful to understand the
processing of an XML instance in just those same simple terms. And, like
Aristotle, there are others who see first nodes nd properties. The debate is not
a new one.

Respectfully,

Walter Perry