Mathy person interested in concept, unsure where to begin.

[nxt...@gmail.com]

Hello, I believe here is a good place to get help on this matter.

As the title states, I have a brain that is very good at math, computer science, and the like.

However!

I've tried learning a foreign language twice now, and Esperanto twice too.

I've got problems with memorization, as well as other disabilities affecting language processing.

(also, horrible spelling)

I've constructed a conlang, and I am in the process of constructing another, both with decisively non-English grammar structures.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Now, with that personal details out of the way, The thing which I ask is bellow.

What is the best way to learn Lojban from a math/logic perspective? Is there any references that I should have on hand?

Thanks in advance,

Romaji ####

[John E Clifford]

Sounds to me like you are a Lojbanologist rather than a Lojbbanist.  Barring someone actually writing a good text book (on the record, a low probability event) learning Lojban has always been a massive memory project with lots of writing, making mistakes, arguing about corrections and then going back at it again.  Learning *about* Lojban, however, is pretty straightforward and can be done quite a bit by reading CLL and various bits from these lists. Have at it!

As for your own conlangs, I would take them to the Conlangers group on Facebook, which is friendly and helpful.

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+un...@googlegroups.com.
To post to this group, send email to loj...@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.

[Romaji ####]

I was looking for a kind of "quick reference" guide to get me started on Lojiban, with stuff like a list of key words and suff/pre fixes, and basic grammar rules, preferably in a logic based environment.

So, basically, I'd be writing the textbook?

On your second point, I don't use FB.

[TR NS]

On Tuesday, September 23, 2014 8:02:54 PM UTC-4, Romaji #### wrote:

I was looking for a kind of "quick reference" guide to get me started on Lojiban, with stuff like a list of key words and suff/pre fixes, and basic grammar rules, preferably in a logic based environment.

So, basically, I'd be writing the textbook?

You can just read it: http://www.amazon.com/Complete-Lojban-Language-John-Cowan/dp/0966028309

[Romaji ####]

Uh...

I've just started.

I'm not willing to pay anything yet.

What about some online, community stuff?

Or is that all going to not be what I want?

Does anyone else here think they even unambiguously know what I want?

[Jorge Llambías]

On Tue, Sep 23, 2014 at 9:17 PM, Romaji #### <nxt...@gmail.com> wrote:

Uh...

I've just started.

I'm not willing to pay anything yet.

What about some online, community stuff?

Or is that all going to not be what I want?

Does anyone else here think they even unambiguously know what I want?

CLL is online, you don't need to buy the book unless you want to. 

You may want to start here, it has the link to the online CLL and some other useful material: http://www.lojban.org/tiki/Learning

mu'o mi'e xorxes

[Romaji ####]

I'll start here.

And I will find out what you just said there

--
You received this message because you are subscribed to a topic in the Google Groups "lojban" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lojban/f9L_hAbBcwc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lojban+un...@googlegroups.com.

[Gleki Arxokuna]

2014-09-24 4:17 GMT+04:00 Romaji #### <nxt...@gmail.com>:

Uh...

I've just started.

I'm not willing to pay anything yet.

What about some online, community stuff?

Or is that all going to not be what I want?

Does anyone else here think they even unambiguously know what I want?

CLL (the reference grammar) is free.

all the links are here http://mw.lojban.org/index.php?title=Learning/en 

Secondary links (unsorted) http://mw.lojban.org/index.php?title=Learning_materials._Secondary_links

(which has some charts of lojban grammar)

However, I doubt Lojban has ever been described from a logician's mathematician's viewpoint no matter how strange it might seem to you.

On Tuesday, September 23, 2014 8:14:09 PM UTC-4, TR NS wrote:

On Tuesday, September 23, 2014 8:02:54 PM UTC-4, Romaji #### wrote:

I was looking for a kind of "quick reference" guide to get me started on Lojiban, with stuff like a list of key words and suff/pre fixes, and basic grammar rules, preferably in a logic based environment.

So, basically, I'd be writing the textbook?

You can just read it: http://www.amazon.com/Complete-Lojban-Language-John-Cowan/dp/0966028309

--

[Romaji ####]

Never?

Interesting.

I'll attempt to remedy that problem.

--
You received this message because you are subscribed to a topic in the Google Groups "lojban" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lojban/f9L_hAbBcwc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lojban+un...@googlegroups.com.

[Gleki Arxokuna]

I'm not an expert in math or logic or programming. I'm even surprised to hear that Lojban shares some properties with programming languages (try converting any simplest program to and from lojban and vice versa using a described algorithm for an programming language).

http://mw.lojban.org/index.php?title=Expansion_of_logical_connections

http://jbovlaste.lojban.org/dict/ifle

Also in {i lo nu da broda cu co'e lo nu da brode} the two {da} are different local variables but in

{i da zo'u lo nu da broda cu co'e lo nu da brode} it is one global variable whereas in 

{i da zo'u lo nu da zo'u da broda cu co'e lo nu da zo'u da brode} they are again different variables.

Correct me someone if im wrong.

Lambda calculus and {ce'u} also comes to mind.

Those were some thoughts on what to include into a future textbook.

[John E Clifford]

The most clearly missing item in this area is the big one, a program for converting any Lojban sentence to a formula of an appropriate logic formulary.  Or, conversely (or perhaps reciprocally) a set of transformations for converting formulae of logic into grammatical sentences of Lojban.  Uniqueness is desirable in both directions but essential in the sentence to formula program.  This is the proof of the "logical language" claim.  (For now, things about the language which make this task harder -- or things that would make it easier if they were in the language -- should be noted but not acted on.)

[Jorge Llambías]

On Wed, Sep 24, 2014 at 10:58 AM, Gleki Arxokuna <gleki.is...@gmail.com> wrote:

Also in {i lo nu da broda cu co'e lo nu da brode} the two {da} are different local variables but in

{i da zo'u lo nu da broda cu co'e lo nu da brode} it is one global variable whereas in 

{i da zo'u lo nu da zo'u da broda cu co'e lo nu da zo'u da brode} they are again different variables.

Correct me someone if im wrong.

I would say you're correct about the second one, possibly correct about the first one, and probably wrong about the last one.

If you were to write the quantifiers explicitly instead of relying on not always clearly stated rules for figuring out where they are. it would be easier. 

[Jorge Llambías]

On Wed, Sep 24, 2014 at 11:03 AM, 'John E Clifford' via lojban <loj...@googlegroups.com> wrote:

The most clearly missing item in this area is the big one, a program for converting any Lojban sentence to a formula of an appropriate logic formulary.  Or, conversely (or perhaps reciprocally) a set of transformations for converting formulae of logic into grammatical sentences of Lojban.

The second part, converting logical formuae into gramatical sentences of Lojban is actually rather trivial. The Lojban to logic is non-trivial because Lojban allows many shortcuts and non-logical elements that are not always well defined, but you are never required to use those when going in the logic to Lojban direction..

[John E Clifford]

Logic to Lojban.  Poorly put.  Yes, we can always actually speak a formula in Lojban.  The trick is get a scheme whereby any Lojban sentence can be shown to be derived from a formula without loss of meaning.  If Lojban really does contain non-logical elements (I'm not sure what that means, but ...) then the claim falls apart.  If the shortcuts are not rule governed, then the claim falls apart.  So the hope is that both of those suggestions are false and that the two projects proposed here are in fact two sides of the same coin.  Since the formula to language move is just (theoretically) standard linguistics, it ought to be relatively easy and then the trip back be pretty straightforward (by forcing if nothing else).  The suspicion that underlies most recommendations to change Lojban or replace it is that, in the fairly blind attempt to achieve monoparsing without a suitable grammar, many complications were inserted that actually interfere with the desired result, rather than just being unduly (it seems from the outside) complex ways of getting there.

[Jorge Llambías]

On Wed, Sep 24, 2014 at 11:53 AM, 'John E Clifford' via lojban <loj...@googlegroups.com> wrote:

 If Lojban really does contain non-logical elements (I'm not sure what that means, but ...)

I was thinking of attitudinals, for example. Or vocatives.

 

 Since the formula to language move is just (theoretically) standard linguistics, it ought to be relatively easy and then the trip back be pretty straightforward (by forcing if nothing else).  

Yes, all of that is pretty easy. 

[John E Clifford]

Both attitudinals and vocatives (and Gricean operators and evidentials and just about anything else I can think of) have places in the extended logics that comprise appropriate bases, so those are not problems.

It is pretty easy -- in theory.  Working out the practical details is the bitch.  But I suspect the real reason it has not been done is that no one has until recently been very explicit about what needs to be done, the whole having been expressed in vague generalities rather than (slightly) more specific programs.  Score a point or two for the radical revisionists.

[Jorge Llambías]

On Wed, Sep 24, 2014 at 1:01 PM, 'John E Clifford' via lojban <loj...@googlegroups.com> wrote:

Both attitudinals and vocatives (and Gricean operators and evidentials and just about anything else I can think of) have places in the extended logics that comprise appropriate bases, so those are not problems.

If those extended logics use the same formalism of FOPL, then their formulas can be directly and automatically translated into Lojban (but without using Lojban's "shortcuts", of course). If they each use a different particular syntax, then they may not have an automatic translation into Lojban, or at least each case would have to be analysed separately to see how it can be accommodated. I don't think it's reasonable to expect Lojban to automatically translate every formalism ever used though. 

 

It is pretty easy -- in theory.  Working out the practical details is the bitch.  But I suspect the real reason it has not been done is that no one has until recently been very explicit about what needs to be done, the whole having been expressed in vague generalities rather than (slightly) more specific programs.  Score a point or two for the radical revisionists.

A lot of work needs to be done to explain how Lojban's "shortcuts", and its "bells and whistles", translate into something that can be called "logical". If that's what you're saying, I agree. What I thought you were saying, but perhaps you were not, is that you had doubts that Lojban would have any trouble expressing FOPL. That part is trivial.

[Romaji ####]

Ok, all that are here.

Where did the current discussion come from?

Because I believe I am totally lost.

--

You received this message because you are subscribed to a topic in the Google Groups "lojban" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lojban/f9L_hAbBcwc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lojban+un...@googlegroups.com.

[John E Clifford]

Sorry; you wandered in at the end of the latest week of a 60-year discussion.  I was supposed to write up a summary 20 years ago or so, but did not and so much has changed since then (again; there have been countless changes dues to changes in Linguistics -- Chomsky, Montague, and the Neos -- and Logic -- Montague again, Kripke, and countless others from the 60s on) that it would be time for a new summary anyhow.  Maybe your interest and an attempt to answer question you have or I think you should have will get me back on track (I have started but got bogged down in details before I got the broad survey done). 

--

You received this message because you are subscribed to the Google Groups "lojban" group.

To unsubscribe from this group and stop receiving emails from it, send an email to lojban+un...@googlegroups.com.

[John E Clifford]

No, the worry is the opposite: whether every sentence of Lojban has a unique (up to equivalence, say) representation in logic which can be automatically derived.  Your project is indeed trivial until you throw in all the qualifiers: colloquial, compact, ergonomic (I have no idea what that means for a sentence, but it seems popular these days), unambiguous, etc.  Then we clearly need a full set (whatever tat means -- enough for all the cases we know of or can think of, I suppose) of transformations, not just the "read it as written" (with a few easy additons) version that is trivial. By way of connecting all this up, the formula derived from a sentence is the same (up to equivalence, of course -- but maybe not even with that condition) as the formula from which the sentence is derived.  That is the test of the logicality of the language.  

[Jorge Llambías]

On Wed, Sep 24, 2014 at 3:41 PM, 'John E Clifford' via lojban <loj...@googlegroups.com> wrote:

No, the worry is the opposite: whether every sentence of Lojban has a unique (up to equivalence, say) representation in logic which can be automatically derived.  

Right. As I said: 

Logic -> Lojban    trivial

Lojban -> Logic   very much non-trivial

 

Your project is indeed trivial until you throw in all the qualifiers: colloquial, compact, ergonomic (I have no idea what that means for a sentence, but it seems popular these days), unambiguous, etc.

Except for "unambiguous", I agree. The automatic translation from FOPL to Lojban is unambiguous, but usually not colloquial/compact/ergonomic.

 Then we clearly need a full set (whatever tat means -- enough for all the cases we know of or can think of, I suppose) of transformations, not just the "read it as written" (with a few easy additons) version that is trivial.

Right, that's very much a requisite for the Lojban -> Logic direction, and necessary also to achieve colloquial/compact/ergonomic in the Logic -> Lojban part.

 

By way of connecting all this up, the formula derived from a sentence is the same (up to equivalence, of course -- but maybe not even with that condition) as the formula from which the sentence is derived.  That is the test of the logicality of the language.  

The formula derived from the sentence derived from a formula is that same formula. That part is trivial. The problem are the sentences not derived from any formula (or not yet known to be derived from any formula).

[John E. Clifford]

Well, in theory in general, every sentence is derived from a formula.  The question is whether we can automatically derive that formula from the sentence. The best (only?) way to test this is to see whether the formula team 2 derives from a sentence supplied by team 1 is the same formula that team 1 used to derive the sentence in the first place.  A weaker test is to see whether the automatically derived formula corresponds to the intuitive reading of the sentence as provided by proficient speakers who re logically sophisticated, etc.

Sent from my iPad

[Jorge Llambías]

On Wed, Sep 24, 2014 at 4:16 PM, 'John E. Clifford' via lojban <loj...@googlegroups.com> wrote:

Well, in theory in general, every sentence is derived from a formula.  The question is whether we can automatically derive that formula from the sentence.

With today's Lojban, in general we cannot. (For a restricted set of Lojban sentences, we can.)

 

The best (only?) way to test this is to see whether the formula team 2 derives from a sentence supplied by team 1 is the same formula that team 1 used to derive the sentence in the first place.

I don't think that would prove much. If team 1 and team 2 are minimally competent, that test would be passed with flying colors. Indeed that test could even be passed by two automatic converters, since conversion from Logic to Lojban can be done automatically, and the reverse can also be done automatically when starting from the restricted set of Lojban that the first part would generate.

The true test would be to start with a general Lojban sentence, then convert it to a formula (that's the hard part), and then see how well the automatically generated sentence from that formula matches the original sentence.

 

 A weaker test is to see whether the automatically derived formula corresponds to the intuitive reading of the sentence as provided by proficient speakers who re logically sophisticated, etc.

There is currently no automatically derived FOPL formula for Lojban sentences in general. (Only for a restricted subset of Lojban there is.)

[John E. Clifford]

We are at a bit of cross purposes here.  Team 1 is using the full potential of a modern theoretical grammar, one that would derive every sentence of a language from some formula, not just sentences of some set trivially matching the structure of the logic.  For most languages and, indeed, for most sentences in those languages, several non-equivalent formulae may give rise to the same sentence (most languages are syntactically ambiguous).  Lojban is planned to avoid this: a given sentence can come from only one formula (up to equivalence - speaking of which, of course, equivalent sentences in Lojban derive from the same formula or equivalent ones). This means that every logically significant feature of the formula must be represented somehow in the sentence and. If that representation is shortcut somehow, that shortcut must be marked to allow a unique reconstruction.  The logic > Lojban process and the Lojban > logic are of course distinct but presumably developed together very closely, as is the surface grammar available for ordinary use (PEG at present). So the tests proposed are not trivial.

Sent from my iPad

[Jorge Llambías]

On Wed, Sep 24, 2014 at 5:07 PM, 'John E. Clifford' via lojban <loj...@googlegroups.com> wrote:

We are at a bit of cross purposes here.  Team 1 is using the full potential of a modern theoretical grammar, one that would derive every sentence of a language from some formula, not just sentences of some set trivially matching the structure of the logic.  For most languages and, indeed, for most sentences in those languages, several non-equivalent formulae may give rise to the same sentence (most languages are syntactically ambiguous).  Lojban is planned to avoid this: a given sentence can come from only one formula (up to equivalence - speaking of which, of course, equivalent sentences in Lojban derive from the same formula or equivalent ones). This means that every logically significant feature of the formula must be represented somehow in the sentence and. If that representation is shortcut somehow, that shortcut must be marked to allow a unique reconstruction.  The logic > Lojban process and the Lojban > logic are of course distinct but presumably developed together very closely, as is the surface grammar available for ordinary use (PEG at present). So the tests proposed are not trivial.

Perhaps a concrete example might help. By "formula" I mean something from FOPL like "Ex: P(x) & ~ Q(x)"

Team 1 receives the formula and puts it into (horribly clumsy but faithful and unambiguous) Lojban thus:  

"su'o da zo'u ge broda fa da gi na ku zo'u brode fa da". 

Team 2 gets the Lojban and reconstructs unambiguously the original formula: "Ex: P(x) & ~ Q(x)"

The two teams must of course have a dictionary that says that "P" is "broda" and "Q" is "brode", and it could happen that Team 2 comes up with non-essential differences in the formula, such as: "Ey: P(y) & ~ Q(y)".

Any FOPL formula that team 1 receives can be subjected to this automatic process. You obviously have something different in mind, but I can't work out what it is..

[John E Clifford]

Well, given that formula there isn't a lot to do but maybe team 1 coming up with 'da ge broda gi nai brode' or 'da na ku ge ganai broda gi brode'

[Jorge Llambías]

On Wed, Sep 24, 2014 at 7:43 PM, 'John E Clifford' via lojban <loj...@googlegroups.com> wrote:

Well, given that formula there isn't a lot to do but maybe team 1 coming up with 'da ge broda gi nai brode' or 'da na ku ge ganai broda gi brode'

That would be team 1 making extra Lojban -> Lojban moves that are not always well defined, so risking messing up. If team 1 sticks with vanilla FOPL -> Lojban without trying any additional fancy Lojban -> Lojban shortcuts, then they can be assured that team 2 will be able to do the Lojban -> FOPL without any trouble.

That's not to say that we don't need to define all the Lojban -> Lojban shortcuts, of course we do, but they are not required in order to be able to express any FOPL in Lojban. They are required to go the other way, and express any Lojban in FOPL

[John E. Clifford]

As the mystics say, the way up is the way down. We won't get the way From Lojban to logic right if we haven't worked out the other direction, because only then can we be sure that Lojban has incorporated all the clues needed.  In the examples here, some of the shifts are very secondary, but deprenexification is surely essential (and dangerous, as you say, though this case is Ok apparently) and so is this sort of collapse.  The Principia to Polish shift and the logical moves are just dressing.

Sent from my iPad

--

[Jorge Llambías]

On Wed, Sep 24, 2014 at 8:16 PM, 'John E. Clifford' via lojban <loj...@googlegroups.com> wrote:

As the mystics say, the way up is the way down. We won't get the way From Lojban to logic right if we haven't worked out the other direction, because only then can we be sure that Lojban has incorporated all the clues needed.

For me there's just no way one can get the FOPL to Lojban wrong, if one knows the very simple and trivial rules that apply.

 

 In the examples here, some of the shifts are very secondary, but deprenexification is surely essential

Deprenexification is needed for getting concise/ergonomic forms, but not to get the logic right in the first place.

 

(and dangerous, as you say, though this case is Ok apparently) and so is this sort of collapse.  The Principia to Polish shift and the logical moves are just dressing.

The move is fine in this simple case. Getting the rule precisely right for the general case needs some consideration of several issues, but it's still one of the easy ones. But this is already part of Lojban->Lojban moves, not part of the FOPL->Lojban step,  

[John E. Clifford]

Well, exactly when which rules apply is probably an important issue.  We know that pronoun rules are late and deprenexification is fairly early, but just what comes where is going to depend on best ways to set limits.  The current grammar is probably a target for the deeper layers, but, as it stands, it seems to have lost some essential pieces, so that it looks like the same sentence may come from more than one nonequivalent formula even though it has only one parse.  The easiest cases seem to come from deprenexification, where scope is not specified (can't find my list at the moment).

Sent from my iPad

[Bob LeChevalier, President and Founder - LLG]

On 9/24/2014 9:58 AM, Gleki Arxokuna wrote:
> I'm not an expert in math or logic or programming. I'm even surprised to
> hear that Lojban shares some properties with programming languages (try
> converting any simplest program to and from lojban and vice versa using
> a described algorithm for an programming language).

Nick Nicholas more or less devised a way of mapping Lojban to Prolog
back in 1993.

http://www.lojban.org/files/papers/lojban_parser_paper
http://www.lojban.org/files/software/analyser


lojbab