"Should" and "would"

54 views
Skip to first unread message

mudri

unread,
Oct 28, 2012, 3:27:08 PM10/28/12
to lojban-b...@googlegroups.com
Is there any way in Lojban to talk about what one should do, should have done or would do in a given event? I've been looking for a while, but can't find anything.

selpa'i

unread,
Oct 28, 2012, 4:05:42 PM10/28/12
to lojban-b...@googlegroups.com
For "should", the word {.ei} is usually used:

.ei do xendo lo cnino
You should be kind to strangers.

.ei do ta pu na lebna
You should not have taken that.

.ei mi ma zukte
What should I do?


For "would", it's slightly more difficult. "Would" usually marks a
phrase as non-factual, which is done by {da'i}:

ju'o da'i do se pluka
You would surely enjoy it.

da'i do se xamgu
It would be good for you.


Sometimes, the "would" appears together with an if-clause, in which case
a {da'i} isn't always needed:

mi va'o lo nu [da'i] mi jipci cu na kuckla lo dargu
If I was a chicken, I wouldn't cross the road.

Here the {da'i} is not strictly needed, because the event tagged by va'o
is not claimed by the speaker. The {da'i} can be used to add emphasis on
the non-factualness of the sentence, however.


There is also a less common way for "would" that uses {mu'ei}:

lo nu sidju lo drata kei mi ro mu'ei pluka
I would always enjoy helping others.

mi no mu'ei lo du'u do mi slabu cu na djica lo nu mi do speni
In no world in which I know you would I not want to marry you.


Hope this helps.

mu'o mi'e la selpa'i

--
doị mèlbi mlenì'u
.i do càtlu ki'u
ma fe la xàmpre ŭu
.i do tìnsa càrmi
gi'e sìrji se tàrmi
.i taị bo pu cìtka lo gràna ku


.

mudri

unread,
Oct 28, 2012, 4:56:55 PM10/28/12
to lojban-b...@googlegroups.com
{.ei} fits sometimes (in the {.eiro'a} sense), but I was primarily thinking of the use of "should have done x" to mean "x would have optimised your position (in life, &c.)". To take "position" literally, you could say whilst reviewing a chess match "you should have taken that bishop with your knight, rather than your pawn". There is no obligation to do this (the erring player could still have won); it is more of a suggestion.

Just before writing that, I noticed that {.e'u} works for some present/future tense "should" instances. "It's a good book; you should read it." can be translated with a simple {.e'u ko broda} or {mi stidi lo nu do broda}. But this is not easy to turn into a "should have". Maybe "I would have suggested..." is a better way to think of it (somehow incorporating the {stidi} solution). So now I'll look at that "would".

{va'o} seems to be a good fit for if... would statements. I've come across the fact that abstractions are not necessarily true before. As for {da'i} I haven't understood before what it means. It's listed as "supposing", but, as a UI, it can't create the structures I'd expect from "supposing". To me, "supposing" is usually used in "supposing that x, y", where x is the supposed fact and y is the main statement. But I suppose (zo'o) that your explanation makes sense.

I don't understand how you've used {mu'ei}. How can we claim things about every alternate history? What if in one of them you had different parents, and inherited a few selfishness genes (assuming that this alternate person could be considered "you", and not a child from that other parent - be'o - who was not born either). I fear I may have misunderstood completely.

selpa'i

unread,
Oct 28, 2012, 6:27:54 PM10/28/12
to lojban-b...@googlegroups.com
Am 28.10.2012 21:56, schrieb mudri:
> {.ei} fits sometimes (in the {.eiro'a} sense), but I was primarily
> thinking of the use of "should have done x" to mean "x would have
> optimised your position (in life, &c.)".

You're probably reading the old definition of {.ei}. Its use has changed
somewhat to make it more useful so that the three examples I gave are
now exactly what you want, that is, they aren't really about obligation.

> To take "position" literally,
> you could say whilst reviewing a chess match "you should have taken that
> bishop with your knight, rather than your pawn". There is no obligation
> to do this (the erring player could still have won); it is more of a
> suggestion.

.ei lo do xirma .e nai lo do sonci pu catra lo va xanto
Your knight and not your pawn should have killed that bishop.

(I happen to have played chess in Lojban a few times :) )

> {va'o} seems to be a good fit for if... would statements. I've come
> across the fact that abstractions are not necessarily true before. As
> for {da'i} I haven't understood before what it means. It's listed as
> "supposing", but, as a UI, it can't create the structures I'd expect
> from "supposing". To me, "supposing" is usually used in "supposing that
> x, y", where x is the supposed fact and y is the main statement. But I
> suppose (zo'o) that your explanation makes sense.

Whatever you mark with {da'i} (and that can be either the entire
sentence or just a part of it depending on where you place it) becomes
counterfactual. It's a supposed situation, just like a phrase that
contains "would". Generally, cmavo definitions can't be taken by
themselves, there is usually quite a bit of history behind them. It
helps to check those past discussions (and their conclusions) and to
compare usage as well.

>
> I don't understand how you've used {mu'ei}. How can we claim things
> about every alternate history? What if in one of them you had different
> parents, and inherited a few selfishness genes (assuming that this
> alternate person could be considered "you", and not a child from that
> other parent - be'o - who was not born either).

PA mu'ei == TAG

The things you listed above are all things we can tag with {mu'ei} as
being part of the alternate world(s) we're talking about. The second
example I gave shows this, let me repeat it and add an interlinear gloss:

mi no mu'ei lo du'u do mi slabu cu na djica lo nu mi do speni
I, in zero worlds in which it is the case that I know you, will not want
to marry you.
(Which more or less means "I would want to marry you no matter what")

This is a strong statement, but since it's about love, I think it's a
fitting means to express one's strong feelings.
It is true that it could include worlds in which you have a different
backstory, the only thing that must be the case is "lo du'u do mi
slabu". You could go for a complete specification of all the
predications that hold in the world you are talking about, but that
would be difficult. Consider the example sentence to use an incomplete
specification. The speaker only felt the need to mention this one
predication.

mudri

unread,
Oct 28, 2012, 7:32:57 PM10/28/12
to lojban-b...@googlegroups.com
You're probably reading the old definition of {.ei}. Its use has changed 
somewhat to make it more useful so that the three examples I gave are
now exactly what you want, that is, they aren't really about obligation.
Where is this new definition? I worry that I'll be misunderstood if everyone reads ".ei UI1 obligation attitudinal: obligation - freedom".

selpa'i

unread,
Oct 28, 2012, 8:29:40 PM10/28/12
to lojban-b...@googlegroups.com
You shouldn't be misunderstood if you use the most common
interpretation, which is that {.ei} means "ought to"/"should".

You can always check the BPFK definitions (which are technically not yet
absolutely finalized, but which are not extremely likely to see any
drastic changes either):

http://www.lojban.org/tiki/BPFK+Section%3A+Irrealis+Attitudinals

Michael Turniansky

unread,
Nov 9, 2012, 1:41:19 PM11/9/12
to lojban-b...@googlegroups.com
  I disagree that da'i is not needed below.  Without it, the sentence reads "Since I am I chicken, there is no road crossing."
          --gejyspa (12 days behind in email. I had brought it down to 9, but have been busy at work.)

selpa'i

unread,
Nov 9, 2012, 1:50:19 PM11/9/12
to lojban-b...@googlegroups.com
Am 09.11.2012 19:41, schrieb Michael Turniansky:
> I disagree that da'i is not needed below. Without it, the sentence
> reads "Since I am I chicken, there is no road crossing."

No, it does not (necessarily) mean that. Abstractions are by default not
asserted.
You can get your meaning explicitly by using da'i nai, but if context is
strong enough, you can get away without that. Still, the default is that
it is not asserted.

mu'o mi'e selpa'i

la gleki

unread,
Nov 10, 2012, 2:00:01 AM11/10/12
to lojban-b...@googlegroups.com
I should add that Lojban follows one interesting principle of language planning called "facultative precision".

In short, 
1. if you remove words from a Lojbanic utterance you slowly lose precision. 
2. And vice versa, of course (by adding more words you prove that Lojban is the most precise language in the world, although, this second aspect of Lojban is unfortunately known much more than the 1st aspect).

So e.g. when you remove {da'i} the sentence becomes more vague but it does not get the opposite or another meaning.

The principle is described in the following article.
No wonder they mention Loglan (however, they criticize it based on their subjective biased feelings).

mudri

unread,
Nov 10, 2012, 5:25:33 AM11/10/12
to lojban-b...@googlegroups.com
What does {da'i} actually do? Is it a bit like {pau}, in that it's helpful but only the {nai} form changes the meaning?

selpa'i

unread,
Nov 10, 2012, 8:49:57 AM11/10/12
to lojban-b...@googlegroups.com
Am 10.11.2012 11:25, schrieb mudri:
> What does {da'i} actually do? Is it a bit like {pau}, in that it's
> helpful but only the {nai} form changes the meaning?

No. Every sentence has an assertive force, or a mood. In simple terms, a
sentence is either asserted, or not asserted.

da'i removes the assertive force, that is, it produces a sentence that
is not claimed to be true by the speaker. There are many attitudinals
that change the mood of a sentence in Lojban; those classified as
"irrealis attitudinals". An example would be .a'o:

.a'o do mi vitke
I hope you visit me.

The speaker does not assert that "you visit me" actually
happens/happened, they are only expressing a hope about it happening.

You could call da'i the base irrealis, the neutral non-assertive
attitudinal, while all the other irrealis markers have some added
semantics involving hope, expectation, desire etc.

On the other hand, a sentence without da'i or an irrealis marker is
understood to be asserted. You can be exlicit about the assertive force
on this end of the spectrum and use ju'a, which is by default
(implicitly) present in any sentence that does not contain an irrealis
marker.

selpa'i

unread,
Nov 10, 2012, 8:52:16 AM11/10/12
to lojban-b...@googlegroups.com
Am 10.11.2012 14:49, schrieb selpa'i:
> On the other hand, a sentence without da'i or an irrealis marker is
> understood to be asserted. You can be exlicit about the assertive force
> on this end of the spectrum and use ju'a, which is by default
> (implicitly) present in any sentence that does not contain an irrealis
> marker.

Or really any other word that changes the basis and/or intention of the
utterance.

la gleki

unread,
Nov 10, 2012, 9:31:10 AM11/10/12
to lojban-b...@googlegroups.com
in my opinion 
da'i explicitly marks the bridi as hypothetical.
da'inai explicitly marks the bridi as actual.

Without both da'i and da'inai the bridi will have no such implications.

Michael Turniansky

unread,
Nov 10, 2012, 11:07:20 PM11/10/12
to lojban-b...@googlegroups.com
  I'm not really ready to concede your contention that "abstractions are by default not asserted"  True, in some cases they can be implications, like "mi terve'u ba lo nu mi catlu le skina" (I will buy after I watch the movie") If I never finish watching the movie (or never even started watching one), I will never go shopping. But in other cases they are quite definitely asserted: "bai lo nu carvi kei mi ca stali lo nenri" (Forced by the rain, I am staying inside).  I am certainly asserting that it is raining, or the sentence is pretty well meaningless.

   Nevertheless, I did one to correct one aspect of my post.  I missed the "mi" at the beginning of the sentence in question, "mi va'o lo nu mi jipci cu na kuckla lo dargu", so my translation should have read, "Since I am a chicken, I will not cross the road".

                --gejyspa

--
You received this message because you are subscribed to the Google Groups "Lojban Beginners" group.
To post to this group, send email to lojban-beginners@googlegroups.com.
To unsubscribe from this group, send email to lojban-beginners+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban-beginners?hl=en.


Devin Prater

unread,
Nov 11, 2012, 1:19:49 AM11/11/12
to lojban-b...@googlegroups.com
[coi rodo]
I am attempting to teach a friend of mine from school how to speak Lojban. Her native language is Russian, but she basically grew up in a Southern English community, so her thought processes, as it seems to me, are a mixture of the two cultures. A few days ago, I was trying to get her to see how the diphthongs in Lojban are in fact logical, and how the C is logical in the way it is pronounced. I think there was something else she wanted me to ask about, but I can't remember what it is. Anyways, she basically wants to know how the c is logically phonetic, (she's gotten way too used to the English c I'm guessing), but she did observe that the tc makes since if the c makes since, but she cannot grasp the need for the c to have the sh sound

la gleki

unread,
Nov 11, 2012, 2:12:32 AM11/11/12
to lojban-b...@googlegroups.com


On Sunday, November 11, 2012 10:20:00 AM UTC+4, la .devin.preitr wrote:
[coi rodo]
I am attempting to teach a friend of mine from school how to speak Lojban. Her native language is Russian

 coi la .devin.preitr  
Short note. For Russians we have a separate group http://vk.com/menli.bangu. Alternatively, you can give them my e-mail gleki.is...@gmail.com All the gismu have been translated to Russian by far (a separate Russian dictionary is http://vlasisku.lojban.org/vlasisku.ru/badna). Cmavo are almost all translated, the Wave lessons will be completed soon (according to what I was told by the translator).

Devin Prater

unread,
Nov 11, 2012, 3:16:13 AM11/11/12
to lojban-b...@googlegroups.com
Ok, I will forward your email to her, if that's alright.

Sent from my iPod with gmail
--
You received this message because you are subscribed to the Google Groups "Lojban Beginners" group.
To view this discussion on the web visit https://groups.google.com/d/msg/lojban-beginners/-/1hO0OZ7UigIJ.
To post to this group, send email to lojban-b...@googlegroups.com.
To unsubscribe from this group, send email to lojban-beginne...@googlegroups.com.

Devin Prater

unread,
Nov 11, 2012, 3:34:45 AM11/11/12
to lojban-b...@googlegroups.com
And a question I have, how do you get your vocabulary? And is there any online/apps that can help with this?


Sent from my iPod with gmail

On 11 Nov 2012, at 01:12, la gleki <gleki.is...@gmail.com> wrote:

--
You received this message because you are subscribed to the Google Groups "Lojban Beginners" group.
To view this discussion on the web visit https://groups.google.com/d/msg/lojban-beginners/-/1hO0OZ7UigIJ.
To post to this group, send email to lojban-b...@googlegroups.com.
To unsubscribe from this group, send email to lojban-beginne...@googlegroups.com.

la gleki

unread,
Nov 11, 2012, 5:58:58 AM11/11/12
to lojban-b...@googlegroups.com


On Sunday, November 11, 2012 12:34:58 PM UTC+4, la .devin.preitr wrote:
And a question I have, how do you get your vocabulary? And is there any online/apps that can help with this?

Do you know about memrise.com? Register there and start learning flashcards. It's free. And not boring. Many people become memrise addicts after a week of constant using it.

For smartphones you can use Anki. It also has Lojban sets.

Here are the links for learning Lojban.
http://dag.github.com/cll/ - is the full textbook
http://vlasisku.lojban.org/ - is the dictionary
http://www.memrise.com/topic/lojban/wordlists/ - is a great place where you can learn lojban words online. Just *register* on that website, try several courses and use the one that you like most
http://lojban.org/jboski - automatic grammar check of what you write in Lojban (which is unique among all languages!) with English glossing (i.e. translation).
http://camxes.lojban.org/ - Even better than the previous item. But without glossing.
https://groups.google.com/forum/#!forum/lojban-beginners - a community this is where you can ask any questions.
http://webchat.freenode.net/?channels=ckule,jbosnu,jbopre,lojban is a chat where you can make friends and ask for help for learning lojban (in #ckule section)

Adam Lopresto

unread,
Nov 13, 2012, 10:04:10 AM11/13/12
to lojban-b...@googlegroups.com
There's nothing particularly "logical" or "illogical" about the choice of 'c' to represent /ʃ/ (the "sh" sound). It's more a matter of 'c' being available (since both 's' and 'k' serve for its main uses), and better than the other letters that didn't get used ('h', 'q', 'w').


On Sun, Nov 11, 2012 at 12:19 AM, Devin Prater <r.d.t....@gmail.com> wrote:
[coi rodo]
I am attempting to teach a friend of mine from school how to speak Lojban. Her native language is Russian, but she basically grew up in a Southern English community, so her thought processes, as it seems to me, are a mixture of the two cultures. A few days ago, I was trying to get her to see how the diphthongs in Lojban are in fact logical, and how the C is logical in the way it is pronounced. I think there was something else she wanted me to ask about, but I can't remember what it is. Anyways, she basically wants to know how the c is logically phonetic, (she's gotten way too used to the English c I'm guessing), but she did observe that the tc makes since if the c makes since, but she cannot grasp the need for the c to have the sh sound
--
You received this message because you are subscribed to the Google Groups "Lojban Beginners" group.
To post to this group, send email to lojban-b...@googlegroups.com.
To unsubscribe from this group, send email to lojban-beginne...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages