[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
--- In jboske@yahoogroups.com, "Jorge Llambías" <jjllambias@...> wrote:
>
> On 10/6/06, John E. Clifford <clifford-j@...> wrote:
>
> > You can, if you want
> > say that there is nothing but dogs in a dog breed, but that does not
> > make the breed a dog or even dog. When it is counted as one it has
> > moved to another lattice or some such transformation.
>
> I would say that the Golden Retriever is a dog. Wouldn't you?
Generally not, but I admit that in English we can say "The Golden
Retriever is a good sporting dog." and the like. This is -- in terms
of Lojban certainly a bad manner of speaking: it means either that
golden retrievers are good sporting dogs ore that the golden retriever
is a good breed within the sporting dog class. As I keep point out,
The Golden Retriever does not have four legs and a tail, though the
most salient things that fall it do.
> If there is a context in which the Golden Retriever counts as one dog,
> then as far as I'm concerned that's all that is required for the
> Golden Retriever to be one node of the metalinguistic dog lattice.
No, in the lattice I laid out, what counts for GR to be in the lattice
is just that it is made up of dogs. The lattice is aboutr theworld
(domain, whatever) not just about language.
> If you don't accept that there can be a context where the Golden
> Retriever can count as one dog, then our understanding of what can
> count as one dog is simply too incompatible to have a meaningful
> discussion, and perhaps we should instead discuss some other
> example where we do have an agreement.
I think language is often deceptive, one way or another. The golden
retriever is a dog distributively but not personally. Unless we mark
that difference we are going to get into trouble in Lojban, though
English seems less concerned -- the context decides.
> Would you agree that the
> Union Jack can count as one flag in some context, and that in some
> other context there might be many Union Jacks such that each counts
> as one flag?
Unlike "dog," "flag," as a quasi linguistic term, allows for a range
of similar things and the abstracted similarlity (or the accumulation
of all in that range) to carry the same name. I think Lojban makes a
distinction, but I am not sure -- the case is less clear than the dog one.
> Or would you agree that "my going to the market" can
> count as one event in some context (maybe it's my favourite activity)
> and that in some other context there may be many different "my going
> to the market"'s (maybe a different one happened each day this week),
> such that each counts as one event?
The situation with events is less clear than flags, but there are some
reasons for going with the notion that both can be called by the same
name. Afain, I think the same distinctions as in the dog case are
probably useful.
> Consider the relationship between the one dog that is the Golden
> Retriever in one context and the many dogs that are Golden Retrievers
> in some other context, the relationship between the one flag that is the
> Union Jack in one context and the many flags that are Union Jacks in
> some other context, the relationship between tha one event that is my
> going to the market in one context, and the many events that are my
> goings to the market in some other context. Do you see a commonality
> in those relationships?
Of course, they are all the same lattice pattern. This does not mean
that the words involved need to be patterned in the same way (or,
indeed, that they should be). Lojban has the wherewithal -- and the
inclination -- to be careful here and make the distinction (even if it
does not actually do so in the case of flags). I encourage that care.
I don't suppose that horrible things are going to happen if we don't
do it that way -- except of course that we gwet metaphysical
incomprehensibilia like Mr. Dog from failure to mark significant level
differences.
> That metalinguistic relationship (metalinguistic
> because it is a relationship between things that count as
> one dog/flag/event in different context, not in the same context) is
> the relationship I would like to call "subsume" but the label is not
> very relevant. It is not McKays "among" relationship, because it holds
> between each of the lower things (whichis one broda in some context)
> and one upper thing (which is one broda in some other context), not
> betweeen each of many things and the many things together, which are
> all brodas in the same context, as "among" does.
Here finally is the basic difference, I think. The higher nodes in my
lattice just are pluralities of lower nodes, ultimately of normal
individuals. We can take nodes higher up the ladder and treat them as
individuals and so as in still higher nodes. But the relation
involved at every level (down to regular individuals, below which the
physical part-whole relation comes into play)is the mereological
primitive -- or McKay's "among" if you prefer. I have yet no idea what
your nodes are nor what the subsumption relation is (other than that
it has the familiar geometry). Obvious;ly, on my lattice, every upper
node is dog distributively (and only so)though most would never be
referred to in that way but merely as {lo gerku} or "a bunch of dogs."