TECH: Lujvo Paper (Part 3)

[Nick Nicholas <nsn@MULLIAN.EE.MU.OZ.AU>]

2.2. Eliminating irrelevant places as lujvo definition

By using lujvo type classification, we have concentrated on eliminating
redundant lujvo places. It is also important to eliminate places
irrelevant to the *definition* of the relation desired; this is a skill
lujvo-makers need to learn. Leaving in or omitting places can make a
big difference to the concept the lujvo ends up specifying.

A good example of this is the distinction
between a (financial) treasury and an (electric) battery. Both have
been expressed in Lojban as lujvo involving the gismu {sorcu}, store.
In Lojban, {sorcu} has the following place structure:

x1 is a store/deposit/supply/reserve of materials x2 in containment
x3 at location x4.

The concept of a treasury is defined in Lojban by the sumti in its lujvo.
Obviously the (financial) reserve itself is important to the definition.
Possibly the materials involved (types of money) and the reserve's location
also matter in defining the concept. This can be debated, and will be
considered in more detail below. But it is hard to see how the physical
containment of the money matters to the concept of a treasury. The walls of
a treasury may be changed from brick to reinforced concrete, but it remains
identifiably the same entity. To leave in the "containment" place, we would
have to argue that the containment matters to the essence of what a treasury
is. But this is not the case with treasuries as we know them; we usually
don't care what contains the money, as long as we know that there is a lot
of money in some given place. Thus we don't include containment in the final
place structure of {dinsro}: we do not say:

... is a store/reserve of money... in containment...

The need to eliminate irrelevant information is even clearer in defining the
place structure of {dicysro}. If we consider a battery as a store of
electricty, its physical location is immaterial. A battery does not become
a different battery if it is moved to the North Pole, or if it is inserted
into or taken out of a radio. The same electricity is stored in the battery,
in the same containment, giving the same battery. If you change the container
of the battery, on the other hand, keeping it in the same location, people
will consider it a different battery. So when you want to specify a battery
uniquely, you need to refer to its container, but not to its location, which
is not fixed. Leaving the "location" place in would imply that physical
location *is* important to the Lojban definition of a battery, that location
makes a battery something distinct from other batteries. Since we don't
believe this to be the case, the place is omitted.

In omitting places, it may be helpful to use the experimental cmavo
{xo'e}, of selma'o KOhA. xo'e "turns off" the place of the bridi that
it fills in: it makes it unimportant to the definition of the bridi.
Consider {se citka}. All places of any bridi are considered to be
implicitly filled by *some* value, so when we say {le snuji cu se citka},
we do not merely say that the sandwich is food, but that it is being
eaten (or will be eaten, or may be eaten, or can be eaten, or is fit
to be eaten) by *somebody*: {le snuji cu se citka ba'e zo'e}. But it
is possible to consider food independently of the entity eating it.
Much food goes uneaten, after all, and some is never intended to be eaten.
The claim {da se citka}, inherently identical to {da se citka zo'e},
is not true of such food, but it is food nonetheless; the eater of the food
is not part of the *definition* of food. We can say that the display sandwich
at the deli {cu se citka xo'e}, is eaten --- but don't consider the eater
at all: the eater doesn't "enter into the equation". But by doing so, we've
effectively recovered the place structure of {cidja}, which does not have a
sumti place for an eater. (It does have a place for who the food is fit for,
but that isn't the same thing).

We can do the same for beverage: {se pinxa} implies something actually
drunk by someone; {se pinxe be xo'e} is closer to "beverage" as we
know it, in that stuff poured down the drain doesn't cease being a
beverage, though it does cease being a {se pinxe} (noone's drinking it:
{le se pinxe na ca'a se pinxe}).

(An important question, requiring more investigation, is whether such {xo'e}
can be assumed in the place structure of lujvo made of just a SE cmavo and
a gismu. Can a {selpinxe} denote a beverage, as distinct to a {se pinxe}?
This would mean that the lujvo (selpinxe) and gismu (se pinxe) form of a word
can take on different and unpredictable meanings. Our position in compiling
lujvo place structures has been conservative: we have not yet considered
this distinction as valid, but we are aware that it may turn up in future
research, if the language heads in that direction. For that reason, lujvo
like {selpinxe} have not been listed separately.)

A battery, then, can be described using this mechanism as a {sorcu befo
xo'e}, and a treasury as a {sorcu befi xo'e}.

There is another important issue: how much information should be
contained in a lujvo which could just as conveniently be given by a
gismu. Let us reconsider {gekyzda}. For the moment, we have accepted
the place structure of this bridi as: x1 is the doghouse of x2. We have
accepted that the breed of dog is not information that is important to
the definition of the doghouse, but to the definition of the dog itself:
rather than say {la monrePOS. gekyzda la spat. la sanktbernard.}, we
choose to say {la monrePOS. gekyzda la spat. noi gerku la sanktbernard.}

But what information do we gain in saying {la monrePOS. gekyzda la spat.},
that isn't already given by {la monrePOS. zdani la spat. noi gerku},
especially if we already know Spot to be a dog. According to this school of
thought, which we'll term Lean Lujvo (si'o toltiljvo), we often make lujvo
to blindly match the lexicon of our native languages, where a gismu by itself
(given the appropriate context) is sufficient. For example, we can talk of a
waiter as a djabe'ipre, a food-carrying-person. But Lean Lujvo would prefer,
once it's obvious we are talking about a worker in a restaurant, that we refer
to her as a {bevri}, instead of blindly translating "waiter" by {djabe'ipre},
and giving redundant information. This affects place structures significantly:
in this school of thought, {la monrePOS. zdani la spat.} says as much as {la
monrePOS. gekyzda la spat.}, if we know Spot to be a dog (information which
can always be filled in by context). So if we want to relate {la monrePOS.}
and {la spat}, it would seem pointless to use {gekyzda} as the bridi, when
{zdani} will do the job just as well. But one could ask whether this means
we need the {gerku} place of {gekyzda} at all: perhaps only the x1 place need
be left in.

This issue becomes pervasive with {be}-lujvo. Typically, the x2 of the tertanru
is described by the seltanru. One can argue in some cases, that the x2 is so
well described by it, that it needn't appear in the final place structure
at all, that to do so would be redundant. We then have to consider, whether
that x2 is to be a generic or a specific entity. Take {gekyzda}: if we
interpret the place structure as: x1 is-a-house-for-a-dog for a dog (x2),
we could be redundant, because we already know x2 is a dog from the seltanru.
If we make (x2) a specific dog, then our predicate has the same place
structure as {zdani} itself. We need then to show it actually conveys
significantly more information, nonetheless, than {zdani}, and thus deserves
to exist.

Body parts provide much better illustration for the Lean Lujvo case. Take
the lujvo for skull, sedbo'u. There are four places to be considered: the
bone (b1), the system of which the bone is a part (b2), the head (s1), and
the system of which the head is part (s2). b2 can either be a person or a
body part. If a person, then s2=b2 (since s2 is obviously a person); if a
part, the s1=b2. In either case, though, do we really need a separate
place for the head? x1 is the skull of the head x2 of person x3? Obviously
not: we know that the bone is a head-bone, so we don't need a place in
the place structure telling us so. The final structure can only be x1 (b1)
is the skull of x2 (s2) --- which is the place structure you'd want anyway.

Or take {jbogerna}, Lojban grammar. {gerna} has three places: grammar g1,
language g2, and text g3. But we already know the language to be Lojban:
it would be ludicrous to have place structure: x1 is the Lojban grammar
of language x2, which we already know to be Lojban, for text x3. But things
are not as clear-cut for {gicygerna}: it is reasonable to claim there are
different Englishes (dialectical variants), and place ge2 might be kept
to denote the particular form of English specified by the grammar. We
will probably decide, though, that such a nicety is best served by a form
like {nort,mbria. zei gicygerna}, which will still have just two sumti
(unless you start concentrating on idiolects of Northumbrian for your ge2,
which is possible, but ultimately not profitable). {gicygerna} itself, in the
interest of compact place structures, will probably also have just two
arguments: grammar and text.

This issue has far-reaching implications as to what precision we wish to
invest our lujvo with. No categorical verdict would be helpful at this stage
of development of the language. To help keep place structures as small and
manageable as possible, we have cautiously supported Lean Lujvo in our
proposed place structure list, but the judgement typically needs to be made
independently for each individual lujvo, and there are many other factors
to be considered.


Momenton senpretende paseman mi retenis kaj # [Victor Sadler, _Memkritiko_ 90]
   kultis kvazaux                           &  (NICK NICHOLAS. Melbourne.
      senhorlogxan elizeon                  #   Australia. IRC: nicxjo.
         (Dume:                             &   nsn@munagin.ee.mu.oz.au .)