krilltish (krilltish) wrote in lojban,
krilltish
krilltish
lojban

Another Few Questions

 Two sets of questions in two days, not bad (I believe).  Can someone help me out with the following:


1.  Is "and/or" a single word in Lojban?  I guess it would be more like "and/xor", really.

2.  Does "and/(x)or/any combination of the following with the previous"  exist as a (shorter) word/phrase on Lojban?  It would be used something like:  The ball can be red, polka dotted, and/(x)or/any combination of the following with the previous flat.  That would mean that the ball can be red or red and polka dotted, or red and flat, or red and flat but not polka dotted, etc.  As opposed to it is either red, polka dotted, and flat or none of the previous (for "and") or it can only be red, flat, or polka dotted but not multiple of those options.  I probably won't really understand Lojban conjunctions (connectives?), so go easy on me.  :)

3.  In English I see a slight need for two/more "and"'s.  If I said "Joe, Jack, and Jill and John and Jerry are going to the park" it sounds as if they are going as a group, even if it might mean the actions are done seperately.  Both are technically the same, but seem slightly different to me.  One could use commas, but that can be distracting and difficult to read.  I write using "et" for each sub-group.  So that it is "Joe, Jack, et Jill and John et Jerry are going to the park" or "JJJ are going as one group and JJ are going as another".  How would this be expressed in Lojban?

4.  In Lojban, is there a difference between "can" (to be physically able to) and "may" ("to be allowed to")?  I would not think so, one just fills in the last place for the word (I have a hard time with the terminology, would it be gismu?) that means, roughly, "to be able".  I am too lazy to find that word right now, but I know there is one.  What would happen if that last place was not filled in.  Obviously it is slight ambiguous, although I do not believe too many would object.  How would it be interpreted "can" or "may"?


Thanks again, and probably soon to be "like normal".
  • Post a new comment

    Error

    default userpic
  • 25 comments
1. Is "and/or" a single word in Lojban?

Yes.

Depending on what you're joining, consider {.a}, {ja}, {gi'a}, {gu'a}, or {ga ... gi}. They're all "inclusive or", so "A or B or both".

I guess it would be more like "and/xor", really.

I'm not sure what that is supposed to mean; "and" means it has to be both, "xor" means it can't be both.

"and" is one of {.e je gi'e gu'e ge...gi}; "xor" is one of {.onai jonai gi'onai gu'onai gonai...gi go...ginai} (I *think* either of the last two will work).

2. Does "and/(x)or/any combination of the following with the previous" exist as a (shorter) word/phrase on Lojban?

For starters, logical connectives in Lojban can only connect two items, so you can't directly say "red, green, or blue" but rather you have to say "red or green or blue", which turns into "(red or green) or blue" unless you specifically group it differently, so each "or" has only two operands.

But if you connect everything with inclusive-or, then I think you'll have the meaning you're looking for; "A .a B .a C" is "A and/or B and/or C" and is true for any of A, B, C, A+B, B+C, A+C, A+B+C.

As opposed to it is either red, polka dotted, and flat or none of the previous (for "and")

Yep, that's what you'd get with "and".

or it can only be red, flat, or polka dotted but not multiple of those options.

I imagine that would be what you'd get with "xor", but I confess I don't understand the implications of connecting more than two things with "xor"; I read somewhere that things can get complicated or hard to specify rigorously and never bothered learning the details.

3. In English I see a slight need for two/more "and"'s. If I said "Joe, Jack, and Jill and John and Jerry are going to the park" it sounds as if they are going as a group, even if it might mean the actions are done seperately. Both are technically the same, but seem slightly different to me. One could use commas, but that can be distracting and difficult to read. I write using "et" for each sub-group. So that it is "Joe, Jack, et Jill and John et Jerry are going to the park" or "JJJ are going as one group and JJ are going as another". How would this be expressed in Lojban?

I'm not certain, but I think that for "Joe, Jack, and Jill are going to the park" you'd want {joi} (they're doing it together, as a mass) while for "JJJ are going to the park, and also JJ are going to the park", you'd want one of the logical connectives in -e (you're just asserting both statements).

So possibly something like {la djos joi la djak. joi la djil. .e la djan. joi la djeris. cu klama lo panka}. Or with forethought connectives, {ge la djos joi la djak joi la djil gi la djan joi la djeris klama lo panka}.

(Beware that if you're joining sumti formed from gadri+brivla with {joi}, you'll probably need to use {ku}, which is not elidable in that case. For example, "The father and the mother (together) went to the park" would be {lo patfu ku joi lo mamta cu klama lo panka}, with {ku joi} rather than merely {joi}. This is, I believe, because {joi} can join both sumti and brivla, and the official grammar only has one-token lookahead, so when it sees {lo patfu joi...} it can't tell whether you're going to say {lo patfu joi lo mamta} "the father and the mother (together)" or {lo patfu joi mamta} "the father-and-mother-together".)

4. In Lojban, is there a difference between "can" (to be physically able to) and "may" ("to be allowed to")?

I think so, but I don't know the details.

(One word for "can" is {ka'e}, but the mapping is not 1:1.)
So possibly something like {la djos joi la djak. joi la djil. .e la djan. joi la djeris. cu klama lo panka}. Or with forethought connectives, {ge la djos joi la djak joi la djil gi la djan joi la djeris klama lo panka}.

Piping those through jbofi'e shows that I messed up the scoping.

I suggest {ge joi gi joi gi la djos. gi la djak. gi la djil. gi joi gi la djan. gi la djeris. cu klama lo panka}, though that's not the model of clarity IMO. (Especially since there are some {gi}s which are part of a {joi gi}, other which match a {joi gi}, and another one which matches a {ge}, and you'd have to remember which one goes with which.)

But it does parse the way I intended it to.
Wow, thanks. That really helped a lot. (I know this sounds, to me at least to be sarcastic-sounding but I do not want it to do so. For future notice, if I am being sarcastic, I will mark as " %%%...%%% ". Another question, is there any attitudinal indicator-like word for sarcasm?) I really love Lojban, even if I do not understand it fully. The side effect is that I now find that I am thinking more logically, therefore I find that I am correcting myself in the middle of a sentence more than normal! [zo'o]

Another question (eventually you guys will grow sick of them, and then, slowly, ever so painfully slowly, callus [zo'o]), how do you mark the following text (and then unmark it/end the marking) to be taken figuratively, in Lojban? For example: [jbo figurative language marker here] The sugar cookie was as large as the moon [end jbo figurative language marker here]. Without the markers, would it not be taken to be that the cookie is really as large as the moon? Obviously, these markers would probably be omitted in writing poetry, but otherwise, I would say that they would be quite useful.

By the way, I did not mean "xor", just another mistake.
how do you mark the following text (and then unmark it/end the marking) to be taken figuratively, in Lojban?

I don't know off-hand, but searching the cmavo list for "figurat", I find {pe'a}: "marks a construct as figurative (non-literal/metaphorical) speech/text".

Since it's in selma'o UI, it applies either to the preceding word, or (if I'm not mistaken), if at the beginning of the bridi, to the entire bridi.

So it hasn't got a terminator, but if you want it to apply to an entire sentence/bridi, putting it at the beginning should work, and otherwise you could apply it to, for example, the selbri to mark that the relationship is to be taken figuratively (but that the arguments in the relationship can be literal, for example).

For example: [jbo figurative language marker here] The sugar cookie was as large as the moon [end jbo figurative language marker here].

My attempt would be {lo sakta titnanba cu dunli pe'a lo lunra lo ka ce'u barda}.

Another question, is there any attitudinal indicator-like word for sarcasm?

I imagine so, though I don't know which attitudinal(s) would be appropriate.
Thank you again.
"pe'a" is a UI cmavo, so while it doesn't have its own terminator, it can cooperate with a pair of brackets: "fu'e/fu'o"

a UI cmavo which attaches to either the "fu'e" or the "fu'o" applies to everything which is inside of the brackets, for instance:

lo sakta titnanba cu fu'e pe'a dunli lo lunra lo ka barda fu'o

The sugar cookie is "equal to the moon in bigness." (metaphorically speaking)

fu'e/fu'o aren't very popularly known, at the moment, but i intend to spread & support them in the future, especially in Cniglic :)
a UI cmavo which attaches to either the "fu'e" or the "fu'o" applies to everything which is inside of the brackets

Ah! Thanks for explaining.

I saw that donkeyhot had mentioned them, but I still didn't know how to use them with UI.

fu'e/fu'o aren't very popularly known, at the moment

I have no problem believing that :)
Have you looked at fu'e / fu'o ? Looks like right thing for that.
No I had not, but now I have and I think it could work. Merci bien.
As pne quite nicely illustrated, yes there is a way to say these. The really great point about Lojban connectives is that are "complete"in the technical sense -- any statement in boolean logic can be reduced to the operators you've got in Lojban. So no matter how complex an expression you've got, there will always be a way to translate it into Lojban.
Why couldn't natural languages have thought of that? It is such a great idea. But, then, Lojban would not be so great.
one just fills in the last place for the word (I have a hard time with the terminology, would it be gismu?) that means, roughly, "to be able".

And to answer that question, I think the word you're looking for is "brivla" -- a category of words that can stand at the core of a bridi, which includes gismu but also fu'ivla and even, I think, a couple of cmavo (such as {go'i} and {du}).

can/may

Anonymous

February 21 2008, 18:37:13 UTC 9 years ago

Others have answered your other questions, but I thought I'd answer you about the can/may distinction. There is a difference, and actually a few different possibilities. {kakne} is roughly "can" ("x1 is able to do/be/capable of doing/being x2 (event/state) under conditions x3 (event/state)"). So that's about actual ability. "may" is a little more difficult, but I'd use something with {curmi} (which means "x1 (agent) lets/permits/allows x2 (event) under conditions x3"), so something like {mi jai se curmi fai lo nu cliva} "I'm allowed to leave", or, a little more transparently, {curmi lo nu mi cliva} "It is allowed that I leave". Neither specifies who allows it, but that's a strength I see in the lojban version (making it explicit that you're not specifying who's granting permission). Finally there's {ka'e}, which is more about innate capabilities, and really not what you want for this sort of thing.
Wow, thanks. I agree, I too think that is just another strength in the Lojban system. I did not even think about the innate abilities kind of thing, but that (possibly for that reason) is even more interesting/awesome/helpful. Thank you.
There are also two more specific relatives of "ka'e" which you may find interesting. With "ka'e" you are saying that the relationship expressed is a possible one, that it "can" be true, but it's vague whether it's ever actually happened. The cmavo "pu'i" marks something which can happen and actually has happened, demostrated potential, "can and has," while the cmavo "nu'o" marks something which is potentially possible but has not actually occured, unrealized potential, "can but has not."

[ta'o] Another way to say "I'm allowed to leave" would be: ".i mi cliva da'i .e'adai" -- I leave, hypothetically, which someone gives permission for.
Lojban's connectives can indeed be confusing. I think that the best way to learn them is to come to them from the inside, instead of trying to translate the ideas you already have into Lojbanic terms. English concepts and phrases often translate very weirdly into Lojbanic connectives, which will give you the impression that the connectives are even stranger and more difficult than they really are! Once you understand the connectives in a Lojbanic way, it will be easier to understand how they relate to other concepts you're more used to.

In the Lojbanic way of thinking, there are four most-basic logical connections, symbolized by the vowels A E O and U. There are various cmavo for various levels of connection, but they all follow this same vowel-scheme, for instance you make a bridi-tail connection by adding "gi'" before the vowel, gi'a gi'e etc, and you make a sentence connection by adding ".ij" before the vowel, .ija .ije etc. Each vowel represents a logical relationship between two propositions.

E is probably the easiest to understand: With E, both propositions are true. To say it from another angle, if either one of the things being joined is false, the whole thing is false; only if both are true is the combined proposition true. "mi klama lo zarci gi'e citka lo plise" -- I go to the market and eat an apple. If I don't eat an apple, the proposition is false, even if I go to the market. If I don't go to the market, the proposition is false, even if I eat an apple. Only if I do both is it true. So E is very simple to use for a sort of "and" which means "both of these things are true."

U is very simple to use and understand as well, but sees little use because the things which it easily expresses are things which are difficult & obscure in English-- another sort of argument for thinking from a Lojbanic perspective. :) When you use U, it means that the first proposition is true, regardless of whether the second proposition is true or false. It's sometimes translated as "whether-or-not." For instance: "mi citka lo badna .u lo plise" -- I eat a banana, whether or not I eat an apple. (See how English isn't as good at this?) As long as I eat a banana, the sentence is true, regardless of whether I eat an apple or I don't.

O is conceptually equally simple, on some levels. It may be confusing how it's often translated as "if", because English "if" does many other things as well. What O means is that either both propositions are true, or neither of them are. It could equally be the case that they're both true, or both false, as long as it's not the case that one of them is true and the other isn't. For instance: "mi .o do klama lo zarci" -- Either we both go to the market, or neither of us do.

The final vowel, A, is true if either or both of the propositions it combines are true. The only time that it's false is if both of the propositions it attaches are false. It's sometimes translated into English as "and/or." For instance: "mi ba cadzu lo srasu .a lo bisli" -- I will walk on the grass and/or the ice. I might walk on one or the other, or I might walk on both, but I definitely will walk on at least one of them.

Those are the heart of Lojban's logical connection system. Let them marinate in your mind. Believe in them. That's my advice to you about Lojban's connectives: Get comfortable with those four relationships, and then come back later to put more decorations and complexities.
livejournal told me my comment was too long, so here's the rest:


For instance perhaps the most common complexity you'll encounter is the negation of the first proposition with A: "ganai mi klama lo zarci gi mi citka lo plise" -- I don't go to the market and/or I do eat an apple. Huh?!?! The reason this is common is because it matches another common meaning of the English word "if": If the first proposition is true, then the second must be as well. You may be able to think this through at this point and understand it-- according to the "A" truth function it could be the case that A. I don't go to the market (true) and I do eat an apple (true), B. I don't go to the market (true) but I don't eat an apple either (false), or C. I do go to the market (false), but I do eat an apple (true), and there's just one case that's specifically denied, D. I do go to the market (false), and I also don't eat an apple (false), meaning that in all cases where I go to the market, I must also eat an apple, and the whole proposition amounts to IF X THEN Y, phew-- but that probably seems like a load of crazy bullshit. I know it did to me at first. :) It's only once you've got a solid feeling of what the deal is with the basic truth functions that you can easily see how they can be transformed into more complex relationships through negation and combination.
O is probably also more often encountered in the negative form (and I believe that not-A O B is the same as A O not-B), which is xor: either one is true or the other, but not both.
For instance perhaps the most common complexity you'll encounter is the negation of the first proposition with A: "ganai mi klama lo zarci gi mi citka lo plise" -- I don't go to the market and/or I do eat an apple. Huh?!?! The reason this is common is because it matches another common meaning of the English word "if": If the first proposition is true, then the second must be as well.

Fun fact: at one point I wanted this sort of thing in program code: condition B must hold if condition A also holds (but if not-A, then whether B or not is irrelevant).

I wasn't sure how to do this "elegantly", until I remembered Lojban's idiomatic use of "ganai A gi B" for "if A then B", so I turned it into if (!A || B) { ...} (that is, "not-A or B", which is precisely "ganai A gi B" or, equivalently, "A na.a/naja/.inaja/nagi'a B").
[le gapla'i ja purla'i pinka cu jai cinri fai mi .i]

I tried to say that I found the immediately-above/imediately-prior comment to be interesting. I am not really comfortable with abstractions, so I raised/rose? the abstraction.

Sorry that this is late, but I just thought of something. Please, do not think that I did not read your comment- I did. Well, the thing that I was pondering is, what kind of connective (bridi-tail, sumti, bridi, tanru, etc.) would one use while programming in Lojban. If I am not mistaken, and I might be, does one not only give commands? I do not really work with computers in that way, so please try to not get to technical.

I also do not really understand what the difference between, for example, [na.a] and [.anai] is. Could you, and/or someone else, fill me in?
Well, the thing that I was pondering is, what kind of connective (bridi-tail, sumti, bridi, tanru, etc.) would one use while programming in Lojban.

How long is a piece of string?

I think it'd depend on the programming language.

If it were particularly well-suited to Lojban, I think there would be a place for all of those connectives, depending on what, precisely, you're trying to achieve. At a minimum, I think that sumti connectives and bridi-tail connectives would both be useful.

The traditional connectives in C probably map best to the bridi-tail connectives, but Perl, for example, also lets you use them as sumti connectives, in a way. But a really Lojbanic programming language would let you use all, I imagine.

If I am not mistaken, and I might be, does one not only give commands?

I believe that's what's called an imperative programming language. And while those are popular, they're not the only kind there are.

For example, in logic programming languages, you tell the program about rules and then ask it to draw conclusions from them; or in SQL, you tell the database what the data you want to retrieve looks like and let the database figure out how to actually do the retrieving; so both aren't really commands.

I also do not really understand what the difference between, for example, [na.a] and [.anai] is.

[na.a] is "not-A and/or B" (= ganai A gi B, TFTT); [.anai] is "A and/or not-B" (= ga A ginai B, TTFT).

In "if-then" terms, "A na.a B" is "if A, then B" or "B if A" ("if A is true, then B must also be true; if A is false, then B may or may not be true"), while "A .anai B" is "B only if A" ("if B is true, then A must also be true; if B is false, then A may or may not be true").

("B if and only if A" is "A .o B": "if A is true, then B must also be true, and vice versa", or inversely: "if A is false, then B must also be false, and vice versa".)
Oh, I see. So, basically, the placement of na/nai reverses the order of A and B? (Kind of a question)

Thanks for the programming explanation.
So, basically, the placement of na/nai reverses the order of A and B?

If you look at them as logical implications, then yes.

I prefer to think of it as choosing which of the two is negated. ("not-A and/or B" vs. "A and/or not-B".)

Compare also "A na.e B" and "A .enai B" -- they mean "not A, but B" and "A, but not B" respectively.

And the two versions are (I believe) functionally identical for "A na.o B" and "A .onai B" -- both "not-A if-and-only-if B" and "A if-and-only-if not-B" are both "A xor B".
Oh, I see. That makes a lot more sense now.
You are correct. That is absolutely bloody mad. What the heck is that supposed to mean? I remotely understand; but at the same time, I feel as if someone is bludgeoning me over the side of the head with a club. I am happy that this is a separate comment, the other one was actually comprehensible and it would be a shame to have spoiled it in such a way... :)

lol. Well, thanks for trying.