Comparisons

Language learning: How to speak Toki Pona, translation problems, advice, memory aids, tools and methods to learn Toki Pona and other languages faster
Lingva lernado: Kiel paroli Tokiponon, tradukproblemoj, konsiloj, memoraj helpiloj, iloj kaj metodoj por pli rapide lerni Tokiponon kaj aliajn lingvojn
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Comparisons

Post by janKipo »

The standard way to make a comparison in tp, "A is more x than B”, is with the two-sentence unit ‘A li x mute; B li x lili’. This approach fits nicely with certain aspects of tp: nothing like conjugations or declension, keep all setneces simple, and so on. But it is not pleasant for people from other languages. It is long and, frankly, klunky. It maes true claim using sentences which are often (usually?) false. Since it is not a single sentence, it cannot be combined conveniently with other sentences, in ‘la’ sentences, for example. But it remains the standard because alternative suggestions have not been very much better and have not gained support.

The simplest alternate suggestion is just to use ‘mute’, which does mean “more” according to the glossary, though with no examples. Indeed, in many contexts, things like ‘ni li pona mute’ clearly mean “That’s better”, not merely (or even) “That’s very good”. But moving away from implicit comparison with other situations to explicit comparisons with other things raises the problem of what goes into 'A li x mute ... B’ for ‘...’? Some semiconscious connection with ‘ante’ perhaps or a phonological echo of “than” suggests ’tan’. But ’tawa’ also seems right (’tawa’ always seems right for any preposition) and, to be safe, ‘pi’ surely covers whatever the relation is. Even ‘lon’ finds a place. But none makes a strong case. Indeed, it is not clear that a preposition is what is needed here. There are several other approaches.

One is to leap over comparatives to superlatives: "A is the most x". This is usually treated as “A is more x than anything else”. But, we might take the superlative as basic and then define comparatives with in it. One approach along this line is to take “A is the most x” a ‘A li x ali’ “A is totally, perfectly, x” This is probably hyperbole, but, given the standard approach, it seems minor. It is also open to some clarification: “among Bs” rather than absolutely. And here ‘tan’ seems natural for the group involved. So “A is the most x B” is just ‘A li x ali tan B’. But B can be any class, so in particular, the class whose only members are A and B, and the winner then is the one that is more x: ‘A li x ali tan A en B’. Still a little klunky but shorter and clearer than the standard comparison. Of course, all the steps here are questionable and certainly not standard.

Another path is to take a practical approach: how do we decide that A is xer than B? We put them together (somehow) and compare. Operationally, then, "A next to B looks x”, ‘A li x lukin lon poka {B}’. There are questions about whether this very physical approach can be extended to other sorts of properties (pona, for example) and whether, even for physical one, ‘lukin’ and ‘lon poka’ are the best way of putting it, but, even if they are sometimes metaphorical, they seem to be pretty close to right. This suggestion is relatively new, so there is still a chance for furtehr refinements.

There is also a principle, long held in Linguistics but rarely mentioned (linguists are as good at hiding the good stuff as theologians), that adjectives are all inherently comparatives, that is that “green” is really “greener than” or “green for”, with obvious second terms dropped. (Loglan tried to make this explicit but literalists managed to ruin its usefulness.) So, "A is more x than B” is just “A is x B” somehow realized. The term dropped for just plain “green” is (depending on which form the comparison takes) “the typical/average sort of thing that A is” or “the class o things A is”. The arguments for this notion hinges on the relativity of most adjectives: a small galaxy is a lot larger than a large microbe, say, and the naturalness of specifying not just natural classes, as in comparing two people, but also in throwing in odd-ball comparisons. “He is pleasingly slender -- for an elephant". This all means that we don’t really need a “more” expression. But we do still need a “compared to” and then we are -- for tp -- back to considering ‘tawa’ and ‘tan’ and ‘pi’, with the same uncertainties: ‘A li x ... B’. (We might consider the Loglan solution, just another argument, but that just goes too much against the strict SVO structure of tp.)

toki pona being what it is, someone was bound to suggest that, in comparisons, adjectives become verbs with the second term as object: ‘A li x e B’, “A (out)xes B”. The problem with this is that it adds another layer of ambiguity to a construction which already has two or three. On the other hand, this solution is clearly within tp. It might be that (as the English just above suggest) an extended version of ‘x’ would wor. But that starts the controversy about the right extension: ‘mute’ or ‘namako’ or ’sewi’ or selo’ or ... .

There are probably other suggestions for comparisons, but all will probably fit within these general pictures (please correct this view). For now, then, the official form, with all it faults (further list probably available) remains as the only clear way to do comparisons. Pity.

Nothing about “A is more x than y” (and so “A is more x than B is y”).
SoweliNasa
Posts: 1
Joined: Fri May 04, 2018 3:59 pm

Re: Comparisons

Post by SoweliNasa »

How about X defeats Y?

lipu li anpa e kiwen.
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Comparisons

Post by janKipo »

I have in the, interim come fully around to using ‘mute’ (and ‘lili’) as prepositions; 'mi suli mute sina’ for “I am bigger than you” and so on. There is a fuller exposition of Facebook a week or so ago.
Last edited by janKipo on Tue May 15, 2018 10:05 am, edited 1 time in total.
SoweliTeloNasa
Posts: 11
Joined: Wed May 02, 2018 8:43 am

Re: Comparisons

Post by SoweliTeloNasa »

Using mute/lili as more/less than makes sense to me. It seems in the spirit of being able to use the tp vocabulary in any context that makes sense, and would be similar to other languages. According to a friend at work that speaks Indonesian, a lot of the grammar is similar to tp. And using mute/lili as more/less would be the same as it would be said in Indonesian (I big more you).
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Comparisons

Post by janKipo »

Interesting! Thanks!
janKuka
Posts: 16
Joined: Sat May 19, 2018 3:02 pm

Re: Comparisons

Post by janKuka »

How about connecting the two sentences with la?


A li x mute la, B li x lili.


This way, we're not saying "A is very x; B is a little x" (which might each be false), but rather "If A is very x, then B is a little x," which feels (to me?) like a comparison by its very structure. Plus, we're keeping it all in the same sentence.

Plus, we might then even have a way to tackle those more intricate cases:


A is more x than y : A li x mute la, A li y lili.

A is more x than B is y : A li x mute la, B li y lili.


How does that seem?
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Comparisons

Post by janKipo »

Well, my literal mind goes immediately to “But A is not very x, so what does that make B?” Strictly, then the whole is true, even if B is clearly even more very x than A. I suppose we could go with some subjunctive sense, but even then, if A were very x, why would B only slightly x any more than is actually the case. I know, we are working on an idiom and factual details don’t really count, but still we don’t want to say something blatantly false (which is the original problem, after all).
The double comparisons present further problems (we want to be able to do them, but the probleems are inherent in any attempt). It’s not clear that, barring a deeper analysis, one can say that A is xer than B is y or even that A is xer than it is y. It might be, alas, that being a little blue is a lot bluer than being very cold is cold, sinnce the scales and gradients are likely very different.

If we really want to do this up brown, we have to go to a situation where, for every sort of thing and every property there is (in theory, to be fadged up as needed) an ordering of those things for that property, from not al all to totally. To say that a thing has a property is then to say that it is on that scale at or beyond a certain point (we say”average” for things of that sort but leave it open whether we mean mean or median). “More x” for things of the same sort (a mare’s nest) then mean further to the right in the parade. Comparing two properties, then involves the things whose properties are being compared and their relative positions (standard deviations and all that stuff) in the two scales. I don’t see much in tp that is going to allow any of this in even a very non-compact way. So we need a nice fiction of some sort. I don’t have one for the complex case, but the prepositional ‘mute’ still seems to work for the simple cases of two of the same sorts of things with reference to the same property
.
janKuka
Posts: 16
Joined: Sat May 19, 2018 3:02 pm

Re: Comparisons

Post by janKuka »

You make some excellent points, jan Kipo o, as usual!

What about, then, if we change it up like so:


B li x la, A li x mute.

(or even, for emphasis: B li x la, A li x pi mute kin.)


Would this go some way toward resolving the difficulty in the simple case?

As for the more complex cases, while I certainly agree with all your points, it strikes me that in ordinary speech these things tend to be done on the basis of a casual, usually almost-unexamined intuition—so I'm not sure I see why Toki Pona would be at a particular disadvantage in this regard...

For example (using the traditional structure for now, in order to tackle one issue at a time), I'm not sure I see how "mi wile moku lili; mi wile lape mute" is any more problematic than "I'm sleepier than I'm hungry." Of course, they both hinge on questionable epistemological/definitional grounds, but isn't that perhaps more of a broad linguistic or even psychological challenge than a strictly Toki Pona one?
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Comparisons

Post by janKipo »

The “if” problems with ‘la’ remain. What if A is not x, nor B neither. One may still be more x than the other.
But the whole scientific bit about curves and standard deviations is unrealistic in real language, where we go by spur of the moment intuitions, “feelings”. in many cases (and, of course, often get disagreements as a result).
The “I’m sleepier than I’m hungry” is probably just a way of saying “My desire to sleep is stronger than my desire to eat”, an uncomplicated comparison of the same thing, strength of desire, in two different objects. I’m not sure other, apparently more complcated cases, can be spread out this way, but I am not sure they can’t either. ‘wile lape mi li wawa mute wile moku mi'
janKuka
Posts: 16
Joined: Sat May 19, 2018 3:02 pm

Re: Comparisons

Post by janKuka »

Touché... And yet... And yet...

I can't put my finger on why just yet, but for some reason the fact that using an "if...then" structure leads to some true statements that would be meaningless per formal logic just isn't bothering me so much here... Maybe it's because it seems to be less problematic than the standard solution? I mean, if neither A nor B is big, but A is bigger than B, saying "A is big; B is small" includes one falsehood and one truth. But in the same context to say "If B is big, then A is very big" at least doesn't include any falsehoods. It lacks the thrust to mean anything in a logical context, of course, but somehow to me it still seems clearly to be communicating something in a rhetorical context—and what that something is seems pretty clear to me as well. I have no idea whether my linguistic or cultural backgrounds might be influencing these experienced clarities. But, for whatever reason, to me it just does feel clear, somehow. (In fact, the intended meaning actually seems much CLEARER to me, intuitively, than that of "A is big; B is small" when neither of them is big.)

Regarding the idea of using "mute" as a preposition, I see three objections:

1. It is explicitly making a change to the TP lexicon, whether semantically or syntactically or both. I am not completely opposed to this, to be sure, but I guess I intuitively subscribe to some kind of "razor" principle here that would advocate that whenever two solutions might work, choose the one that doesn't do this.

2. The change it is proposing adds ambiguity into one of the areas of TP in which it feels to me as if ambiguity can be hardest to resolve: identifying when a given word is and is not acting as a preposition. Even if it turns out that in all cases only one reading would be possible with this construction, I still feel a pang at the thought of needing to add one more word to the list of words of which one has to warn newcomers, "When you see this word, first stop and check whether it is functioning as a preposition in that sentence or not." (Even if nothing else, it complicates a language whose biggest draw is its lack of complications.)

3. If I want to be super picky (and, since I'm talking to jan Kipo, I do, bwa ha ha!), I would argue that the preposition solution doesn't actually solve the problem of having to use an untruth to get at a truth. (Sorry!) After all, "mi lape lon tomo mi" can only be true if "mi lape" is true—and so on, I think, for all other TP sentences including a PP. If we take "mute" as a preposition, then it would seem to follow that "wile lape mi li wawa mute wile moku mi" would only be true if "wile lape mi li wawa" were true. If we took this prepositional sense of "mute" to mean "more than" (as it seems to be), then we could see the problem even in the English: "My desire to sleep is strong even more than my desire to eat." It seems to me that if we were to postulate (which we could, of course) that that's simply not how this sense of "mute" works, then it seems to me that we would in essence be saying that it is not actually a preposition, but something else—perhaps just a part of a thoroughly idiomatic construction, such as "X ala X."
Post Reply