mu'ei [ ROI ] possible worlds ROI: "p (= bridi) in PA possible worlds in which q (= sumti tagged by mu'ei)": (cp. {munje, mu'e-}): (proposed by AndR)

Some canonical examples:

  • mi no mu'ei citka = "I couldn't/can't eat; it is impossible that I eat; in no possible worlds do I eat"
    • What's wrong with mi na ka'e citka?
    • Standardly, that means "I am not innately capable of eating", not "It is impossible that I eat". Note too that the fact that mi no mu'ei citka means "I couldn't eat" does not in itself mean there's anything wrong with mi na ka'e citka. --And
  • mi su'o mu'ei citka = "I could/can eat; it is possible that I eat; there are possible worlds wherein I eat"
    • What's wrong with mi ka'e citka (mi nu'o citka or mi pu'i citka depending on circumstances).
    • Standardly, that means "I am innately capable of eating". --And
    • I think mu'ei is a useful addition to the language, but also I think ka'e has to be equivalent to su'omu'ei. It cannot refer to an innate capability of the x1 sumti. If anything, it would be the innate capability of all members of the relationship to enter into that relationship: mi ka'e citka le va plise would mean as much that I am innately capable of eating that apple as that that apple is innately capable of being eaten by me. Is that different from saying that the relationship citka is possible between me and the apple? Are there examples where one is true and the other false? --xorxes
    • Your point is a good one, but in English there is a difference between (i) "I might swim", "It may happen that I swim" and (ii) "I am capable of swimming". It's a tricky difference to pin down, and has been the topic of much research, but it's a real distinction. And Woldy and the mahoste seem to me to be at some pains to be clear that it is meaning (ii) that is intended. However, I'm not sure that it is a distinction that needs to be preserved in cmavo. In a context where the distinction is crucial, kakne and cumki can make it. --And
  • ro nanmu ro mu'ei prenu = "In every possible world, every man is a person; men are necessarily people"
    • What's wrong with leza'i nanmu cu nibli leza'i prenu or even ro nanmu cu prenu?
    • re nanmu cu prenu does not specify whether the claim applies to all worlds or just This World. Same problem, plus others, with your other sentence. --And
  • ro mu'ei <LE> du'u le trene cu spofu kei mi (jai) lercu "If the train breaks down (or: had broken down), I would be late; in every possible world in which the train breaks down, I am late"
    • What's wrong with lenu da'i le trene cu spofu binxo cu rinka leza'i mi jai lerci
    • Among other problems, I believe that is a misuse of da'i (for reasons adumbrated on Lojban List September 2002, but not added to the Wiki), and it doesn't make clear that it is in those worlds that the train breaks down that I am late. That's setting aside any disputes about the definition of rinka. --And
  • su'o mu'ei <LE> du'u le trene cu spofu kei mi (jai) lercu "If the train breaks down (or: had broken down), I could be late (= it could happen that I am late); in some possible world in which the train breaks down, I am late"
    • What's wrong with either lenu da'i le trene cu spfou binxo cu ka'e rinka leza'i mi jai lerci or ganai le trene cu spofu binxo gi cumki fa leza'i tu'a mi lerci, etc.
    • For the first sentence, see remarks above. For the second one, try reading around the wiki pages devoted to "if" and conditionals. They address problems with using ganaigi. --And

[NB "<LE>" — fill in with appropriate gadri; I prefer tu'o or lo'e]

(Arguably, mu'ei is to ROI as cu'o is to MOI, so long as "probability" is understood as some kind of objective quasistatistical phenomenon rather than epistemic concerning our degree of knowledge, certainty and confidence.)


Now that I've figured out what this means, I strongly support the idea of this cmavo.

So I'll attempt to use it:

mi ricfu ro mu'ei ledu'u mi ponse lo megdo rupnu
I am-rich (in) every possible-world-containing the fact-that I have a million dollars.
If I had a million dollars, I'd be rich. (da poi megdo rupnu ponse zo'u da ricfu .i mi se xanri ledu'u mi me da)

So romu'ei means "would", in a way that expresses that you're certain the possible world would be like that. so'amu'ei would be a slightly weaker claim ("most likely would"), down to so'umu'ei ("might possibly") or nomu'ei ("wouldn't").

  • ... and su'o mu'ei means "could". And

This fills a tremendous gap in Lojban, and handles possible worlds in a better way than Loglan. I will make sure to use this when I get the opportunity.

And also suggested ba'oi to talk about the likelihood of something happening in the future under certain conditions, but I don't think this is as necessary. Since there is nothing counterfactual about talking about the future, you could use cu'o and deal with it with one of the many methods under if; I also don't see anything which would be wrong with using mu'eiba. rab.spir

  • mu'ei is not only for counterfactuals. cu'o is in MOI, not ROI. The many methods under if either are flawed or won't work in all circumstances. --And

  • I'm sorry I haven't yet found the time to write up a proper exposition of mu'ei and ba'oi. mu'ei is intended as the optimal way of rendering "if p then q could/would be the case". ba'oi is a variety of mu'ei that restricts possible worlds to those that are identical with the actual world up to the present point in time but not necessarily beyond it. Whoever it was that figured out what mu'ei is got it right. And
    • If the world you're describing is identical with the actual world up to the present point in time, what's counterfactual about it? And if the only reason to use it instead of another if device is the (rather nice) way PA is built into it, why couldn't that be mu'eiba? --rab.spir
      • I don't think mu'ei ba broda entails that the worlds differ from the current/actual world only in the future. Hence the need for ba'oi. --And

Feel I'd like this approach very much: {.i xagmau romu'ei'pu' lenu mi'o na broda} "We shouldn't have ..." etc. --.aulun.

Note that PA ROI PU parses as PA ROI KU PU KU.

What about using "da'i"? As in, "mi ricfu ro roi le du'u da'i mi ponse lo megdo rupnu". That's just something you can use until mu'ei becomes accepted. tinkit

  • Is that the right meaning for roi as sumti tcita? If so, your sentence works, but — problematically — "mi to'e ricfu ro roi le du'u da'i mi ponse lo megdo rupnu" would be no more or less true in any universe in which at no time do you have a million rupnu. So it doesn't really work. --And

  • If I keep using da'i when I mean mu'ei then it would not help mu'ei get accepted at all. My objection to da'i is that, unlike many other attitudinals, there's no way to say the same thing explicitly without attitudinals. (Well, you could do something with vi le munje poi... but that sucks.) I suggest that da'i could still be used for vague counterfactuals, but mu'ei would be more precise.

I think that you're complicating the issue by defining mu'ei in terms of another event. Better to have it just "bridi occurs in n possible worlds." Then "If I had a million dollars, I would be rich" could be romu'eiku ganai da'i mi rupnu megdo ponse gi mi ricfu. da'i could be used to indicate that the particular sub-bridi is not true in this world. — Adam

  • Um, that defeats the entire purpose, considering that you have to introduce "" and "da'i" to make up for it. Besides, what's too complicated? It works the same way as roi. Since PA+ROI acts like a modal, it needs to have an object; is there some other object you think it should have which would be more useful? --rab.spir
    • Fewer moving parts. You can still express the same things, but each part only has one meaning. You also have romu'eiku go ... gi ..., etc. da'i already indicates whether the sub-bridi is true is this world, and the explanation given at the top of the page doesn't have any way to indicate whether the bridi is true in this particular world. Just because something is a tense, doesn't mean that it has an obvious meaning when tagging a sumti (e.g. ta'e and ki).
      • mu'ei doesn't describe what is true in this particular world, and neither does ca describe what was true last Thursday. (Unless "this world" or "last Thursday" is the sumti, of course.) If you want to say that something is also true in this world, put in a je ca'a.
      • You seem to have come across a precise definition for da'i - could you tell me what it is? Funny thing that da'i would have one specific meaning pinned down, and not ui or a'o.
        • da'i, .ui, and .a'o all have precise meanings, whether or not they are clearly representable in English. da'i means "not in this world but in an imaginary one". From what I can tell, that's all mu'ei does, except that mu'ei adds the silly notion of being able to count off these imaginary worlds. Only mathematical truth is arguably "ro mu'ei"; any other application of "ro mu'ei" just reveals a lack of imagination. One argument against mu'ei is that it imports this questionable metaphysical baggage, whereas the simple idea of hypotheticality introduced by "da'i" is relatively uncontroversial. --xod
          • That would be a good reason not to use mu'ei at all. Adam (if that's who I'm still replying to) wants to use mu'ei, da'i, and simultaneously.
          • Phew, I was afraid no one would bring up the questionable metaphysics aspect of this thing. "Count of possible worlds" is almost sci-fi silliness.
          • "sci-fi silliness" — as found in most books on advanced logic, and more usually known as "philosophy".
      • ta'e with a sumti would be "habitually during..."
      • Finally, romu'eiku works already. If you want to use it that way, go ahead! Just don't try to stop us from using it with a sumti.
      • rab.spir

and there any instances where pamu'ei broda is meaningful?

I've been reading up on the development of Loglan, and their most recent system for counterfactuals is very similar to mu'ei. Convergent evolution, I'd say.

  • I didn't notice that in my cursory skimming of Loglan reference materials, but mu'ei is simply a Lojbanization of the most reputable way of handling conditionals in logic/semantics. I imagine that if, prebaseline, any Lojbanist had seriously thought about conditionals & then read up on them, they would have come up with mu'ei. --And

Exactly what "problem" is this supposed to solve? There's already
sufficent (and far more precise) methods for saying the different
sorts of ifs, and da'i (along with certain of the attitudinals,
such as ianai) already allow you to make false statements while
being up-front about it. The "I could ..." usages mentioned above
are already covered by existing cmavo (da'i, ka'e, nu'o, pu'i ...).
Just because something doesn't look exactly like natural languages
doesn't mean it isn't sufficent. -mi'e .djorden.

  • The cmavo you mention here are of no use — da'i is for nonassertive illocutionary force and the others have to do with capability. There are no cmavo that can do what mu'ei does. It would probably be possible to do it with a lujvo, but that is true for 99% of all the cmavo. (99% = hyperbole; I haven't done an actual count.) Most of the cmavo are there to make life easier, not because they are indispensable. --And

So I now recant all my objections to this. I still think some of
your example sentences (which are really all I was going on) can
be said using ka'e, though I now agree that ganai does not work.

The reason I rejected this at first is likely obvious: I had no
idea wtf it really meant (and I was completely misreading 'possible
worlds')---so now that I know what this means, to prevent the same
problem for other people I wrote up a thing on this which should
put it into terms that are hopefully simple enough, at
possible worlds and mu'ei. --Jordan DeLong

ki'e djordyn mi'e And

Created by admin. Last Modification: Monday 22 of September, 2003 19:52:52 GMT by admin.