lo'e: Squinting

From Lojban
Jump to: navigation, search

(An elaboration of [1] ). I'll be incorporating elaborations and corrections I have received into this document in the next few days. -- nitcion

Some of this stuff is still unresolved, and will have to wait for me to start reading Formal Semantics. I'll be doing meta commentary in italics.

OK, what's the deal with this lo'e? It was a recent debate on jboske, and a hectic one at that, but it looks like And's myopic singulariser won over Jorge's intensional article (though they may converge yet.) Gobbledygook. OK, let me walk through it. (What I'm going to say may not contradict Jorge's lo'e, actually, but we'll defer that debate to jboske.)

When you say "The lion lives in Africa"... no, let's drop that. When you say "The typical American likes baseball" what do you mean?

ro merko cu nelci le kelcrbeisbolo . No, that's claiming every single one does, and that's not true.

so'a merko cu nelci le kelcrbeisbolo . Kinda a true, I suppose, but it's not capturing the notion of it being typical, it being normal, it being a generalisation. Now, we have tense cmavo for that. But let's see what other gadri can do for us.

loi merko cu nelci le kelcrbeisbolo . Actually no. For all the talk of Mr Water and Mr Rabbit that erstwhile Lojban pedagogy has used, masses are not in fact what you need here. A mass simply says that you cannot make the bridi claim of individuals in the group, but only in the whole group. {loi nanmu cu bevri le pipno}: not Andrew, Barry and Chris each carried the piano, but the three of them in concert carried the piano. Now, is that relevant here? Surely not. Andrew can like baseball all on his lonesome, without needing any assistance from Barry or Chris.

The answer to our question, of course, is lo'e merko cu nelci le kelcrbeisbolo : The American (the typical American) likes baseball.

So where does this lo'e merko d00d live? How many kids has she got? Did he cheat on his taxes last year? Do you think she'll go out with me?

Now, these questions are nonsense, right? But why? For any American {ro da poi ke'a merko je prenu}, you should be able to ask how many kids they've got. Why can't I ask that about this lo'e merko d00d?

Because this lo'e merko d00d isn't a person. S/he's a phantom. An abstraction, if you will.

The average American may have 1.3 kids, but no one person alive has a fractional offspring. That's because the average American is a mathematical abstraction out of all the Americans out there --- the childful and the childless. Now, count all the Americans there are. Bob and John and Robin and xod and Jay and Mark and and and. They all have names; this is an extensionally defined set (meaning you can count 'em.)

Now squint. Squint enough, that all the differences between these Americans fade out. What are you left with?

What you're left with is an abstraction. But this abstraction, this phantom, still has some properties. It has the properties that most individual Americans normally has. So you can speak about those properties; you can make propositional claims about this phantom. But not every claim you can make of an individual American can be made of this phantom too.

  • .i xu la bab. rirni re da .i go'i
  • .i xu la djan. rirni re da .i na go'i
  • .i xu lo'e merko cu rirni re da .i na go'i

...

  • .i xu la fred. nakni .i go'i
  • .i xu la salis. nakni .i na go'i
  • .i xu lo'e merko cu nakni .i ba'e na'i go'i

Moreover, you can count individuals; but you cannot count this phantom. When you squint, you see one abstract generalisation. re lo'e merko is meaningless. If the average American earns $50k, can the average American get together with another average American and buy Enron shares? That's meaningless. (There is such a thing as an average couple; but that's another story.)

The details of what you can and cannot claim of this phantom generalisation figure are still hazy; but let's move on to clausal abstractions.

We will come back to this, because the nature of squinting is a major controversy. Is it objective, based on a survey of the population? ("71% of all Americans have stated that they like baseball in a recent poll. Therefore...") Or is it influenced by subjective perspective? ("I am obsessed by Liv Tyler, so I see a little bit of Liv Tyler in The Typical American --- and besides, there's no such thing as a survey of 300 million people").

And we have a second problem, which haunts this: what is the difference between le'e and lo'e It is true of the cultural stereotype of Americans that le'e merko cu nelci le kelcrbeisbolo; does this necessarily transfer to lo'e merko?

Lojban is odd among the languages I know, at least, in that it treats nominalisations --- clausal abstractions --- exactly like any other sumti. In particular, you can count most sumti; they're extensional. Well, you can count nominalisations too.

So far so good? You can speak of {pa cifno} and {re cinfo}; you can also speak of {pa nu cecla} and {re nu cecla}.

What's so surprising about that? We say "one shooting" and "two shootings" in English. We understand them as bounded events, in particular places with particular participants, and distinguishable from each other; so Oswald and Kennedy were involved in one shooting, and Lincoln and Booth in another.

So. You like swimming, ok? How many swimmings do you like?

Here we have a problem. What does {mi nelci lenu mi limna} mean? "I like swimming", you might think. Think again. What would {mi nelci le merko} mean? That you love all Americans on earth? Probably not. Probably you're referring to a specific, context-salient American. One American is distinct from another; you can separate them from each other, and single out the one you like in particular.

What does {lenu mi limna} mean? It doesn't mean 'swimming' in general. No sir. It means a swim. A particular swim, just as {le merko} means a particular American. What distinguishes Americans from each other? Their properties, their names, whatever. What distinguishes particular events from each other? Their times, their places, their arguments. {lenu mi limna la pacifikas de'i li 2002pi'e5pi'e1} is distinct from {lenu mi limna la atlantikas de'i li 2001pi'e3pi'e15}.

"But I ain't talking about swimming in the Pacific on May Day, or in the Atlantic on the Ides of March. I'm talking about swimming in general."

"Nonsense. If you have {le nu limna}, you have {le nu da limna de de'i di}. There are only specific events of swimming --- specific swims; just as there are only specific Americans. There are no such things as generalisations of events; there are only particular events holding between particular participants at particular times and places..."

... unless you squint.

Conjure up in your mind all the swims you've had, real and potential. Squint away their particular details. What are you left with? You're left with a phantom abstraction --- as opposed to a concrete abstraction! --- which involves you, and water, and not much else. Because everything else is details, and you're squinting those away. What you're left with, is swimming.

So. I liked my swim : .i mi nelci lenu mi limna

I like swimming: .i mi nelci lo'enu mi limna

Once you abstract out {lo'e nu limna} from {ro lonu limna}, you'll find there are things you can say about any particular swim, that you just cannot say about swimming in general. Just as it's meaningful to say whether the typical American likes baseball, but not whether the typical American will go out with you Friday week. So:

  • .i do nelci lenu do limna de'i ma .i de'i li 2002pi'e5pi'e1
  • .i do nelci lo'enu do limna de'i ma--- .i na'i su'o da zo'u: mi nelci lo'enu mi limna de'i da

Swimming ain't swims. It's a mooshy glob of swims. That's why And is saying it as lo'enu. For that matter, that's why Jorge and And have said it as tu'o nu.

The whole point of squinting is to see one mooshy glob instead of five hundred sharp focus individuals. If you can still discern two or three, you're not squinting hard enough. And there's not much point in counting when there can be only one thing to count. tu'o is the non-number; it's the refusal to count. So it's been invoked in this cause too.

To which And adds that there's no point refusing to count, as far as Gricean maxims are concerned, unless there's only 0 or 1 member of the class. So tu'o broda amounts to lo pa broda --- and le pa broda and lo'e pa broda. Further, in a population of one, le, lo, lo'e and so on presumably add up to the same thing, since the denotation ends up the same (they all end up referring to the same single entity).

Jordan retorts that this use of tu'o is bogus. My intent here is not to defend it (I think it is politically icky whatever its logical merits), but merely to clarify its motivation, since it has been seen a lot.

One last step. At the last minute, Lojban introduced a distinction between {nu}, stuff that happens in the world, and {du'u}, claims about the world, concepts about what's going on. Languages sometimes distinguish between them, but not as routinely as Lojban does. If something is {nu}, it's not {du'u}; and vice versa. If you want something covering both, you use {su'u}. I doubt most Lojbanists know su'u is even there; and as I said in the lessons, I think they should, because people may well not want to make the nu/du'u distinction.

So, when you know that Fred swims, you know a claim, not an event. And just like events, claims are specific; they have all their arguments filled. So you can know the propositions: {mi djuno ledu'u la fred. limna la pacifikas de'i li 2002pi'e5pi'e1}, and {mi djuno ledu'u la fred. limna la atlantikas de'i li 2001pi'e3pi'e15}, and {mi djuno ledu'u la fred. limna la .indikas de'i li 2001pi'e7pi'e14}. And then, you can squint, and induce a generalisation: {mi djuno lo'edu'u la fred. limna}. {limna ma}? The question is invalid. You're not making a claim about a particular swim, in a particular body of liquid. You're generalising.

This is hugely controversial, and I admit I was probably carried away. Events are defined by all their participants, including place and time: (Pacific, Mayday) and (Atlantic, Ides) define different events. This is not necessarily so for propositions. One way of turning places off is squinting them off: .i na'i lo'enu pa da zo'u: mi limna da But another way is just plain not knowing what goes in there: .i mi na birti ledu'u do limna makau What you're uncertain of is presumably not a squint of a million propositions, but a distinct proposition on its own. And claiming that ducks can walk may be considered a single claim, with a don't-care x2 of cadzu. And doesn't dispute that there is only one claim involved in something like .i mi na birti ledu'u do limna makau; he does dispute that this should automatically mean you use le rather than lo'e. More on this when I've read further.