Interesting features in your conlang

For the discussion of language mechanics, grammar, vocabulary, trends, and other such linguistic topics, in english and other languages.

Moderators: gmalivuk, Moderators General, Prelates

tesseraktik
Posts: 218
Joined: Fri Nov 06, 2009 9:14 pm UTC
Location: Stockholm, Sweden

Re: Interesting features in your conlang

Postby tesseraktik » Wed Sep 22, 2010 12:23 pm UTC

Iulus Cofield wrote:I'm not sure of the value of a language that needs a second language to function. Why not just make your language have additional "empty" pronouns and the vocabulary to define those pronouns in itself?
I'm not sure what you mean by additional "empty" pronouns, but the value of such a language, in my opinion, would be that it could act as an instrument for evaluating words (admittedly very slowly, and not 100% conclusively). The idea is precisely that for the most part, you would express "more complex" concepts ["more complex" in the sense that they do not have their own words in this language] by a combination of already standard words using rules similar to those of, say, Lojban. However, if a speaker feels that he/she can't accurately or comfortably represent the concepts he/she wants to exress using only constructions made using the standard vocabulary, he/she has the option to introduce a new word by declaring it at the beginning of the piece. After ten-thousand years have passed and 100 pages have been written in the language, scholars might check what words people have chosen to declare rather than construct, and use this information, well, not really as experimental evidence of anything, but perhaps as an indicator of... ...something.

On a less lofty level, I believe that individual speakers can benefit from learning "incomplete" languages, because it forces them to reflect on the words they use every day.
Do we need a word for "Hello!"? What is it we mean to express with such words?
Is "love" a fundamental concept? What does a person mean when he or she tells another "I love you."? When one tells one's spouse one loves him/her, and he/she replies "I love you, too.", are they expressing the same emotion? ...or would they, if they had lacked a word for "love", have used different constructions to express that which we simply label as "love"?

As you can tell, the goals of this language are not all that well-defined, but the idea is basically to study language by creating a linuistic challenge and see how people choose to overcome it, similar to the way we study intelligence by creating puzzle-solving challenges for various animals and people and seeing how they solve [or fail to solve] them.
ni'o mi nelci le zirpu sovmabrnornitorinku
Spoiler:
++$_ wrote:What's a "degree"?

EDIT: I looked it up on Wikipedia. Apparently it's some ancient Babylonian unit for angles :/

User avatar
Roĝer
Posts: 445
Joined: Sun Oct 05, 2008 9:36 pm UTC
Location: Many worlds, but mostly Copenhagen.
Contact:

Re: Interesting features in your conlang

Postby Roĝer » Wed Sep 22, 2010 2:59 pm UTC

But the need for certain concepts to have a word will strongly depend on your linguistic background. If your native language, and every other language you know, has a word for love, then you will think it is not possible to not have one. An experiment like this will need monolingual native speakers to have any real value.
Ik ben niet koppig, ik heb gewoon gelijk.

tesseraktik
Posts: 218
Joined: Fri Nov 06, 2009 9:14 pm UTC
Location: Stockholm, Sweden

Re: Interesting features in your conlang

Postby tesseraktik » Wed Sep 22, 2010 8:47 pm UTC

Roĝer wrote:But the need for certain concepts to have a word will strongly depend on your linguistic background. If your native language, and every other language you know, has a word for love, then you will think it is not possible to not have one. An experiment like this will need monolingual native speakers to have any real value.
Not necessarily; given a large enough base, one could still gain some interesting answers if one had data on the languages known by those people. Furthermore, many of the absent words would be ones that are present in very many languages (such as "Hello" or "love", maybe even the word "like"), so that a majority of speakers would be used to having a standard expression for them.

I also disagree with the statement that if every language you know has a word for love, you'll assume it's impossible to make do without one; conlang fans are often faced with such difficulties, and they continue to press on. It may not be easy, but it wouldn't be much of a challenge if were.

Really, because the speaker base would be bound to be much to small to make the linguistic experiment very interesting, I think the personal challenge would be the main reason to pick up such a language; not because you want to contribute to some study, but because you wish to see for yourself which words you are reliant on, if you can become free of that reliance, and if not, ask yourself "Why not?".
ni'o mi nelci le zirpu sovmabrnornitorinku
Spoiler:
++$_ wrote:What's a "degree"?

EDIT: I looked it up on Wikipedia. Apparently it's some ancient Babylonian unit for angles :/

Makri
Posts: 654
Joined: Sat Oct 03, 2009 4:57 pm UTC

Re: Interesting features in your conlang

Postby Makri » Wed Sep 22, 2010 9:00 pm UTC

Why not? Because the thing is not so easy do circumscribe as to make anybody understand you. What's interesting about this? Well, in principle, it's related to the human mind's capacity for metonymy, metaphor and analogy, but that merits a more systematic study anyway, not this weird kind of... game.
¬□(∀♀(∃♂(♀❤♂)⟷∃♂(♂❤♀)))

tesseraktik
Posts: 218
Joined: Fri Nov 06, 2009 9:14 pm UTC
Location: Stockholm, Sweden

Re: Interesting features in your conlang

Postby tesseraktik » Wed Sep 22, 2010 9:08 pm UTC

Makri wrote:Why not? Because the thing is not so easy do circumscribe as to make anybody understand you. What's interesting about this? Well, in principle, it's related to the human mind's capacity for metonymy, metaphor and analogy, but that merits a more systematic study anyway, not this weird kind of... game.
Certainly, it merits more systematic study, and I'm sure there are more systematic studies going on that I know nothing about, anyway. However, games are still fun, and potentially educational.
ni'o mi nelci le zirpu sovmabrnornitorinku
Spoiler:
++$_ wrote:What's a "degree"?

EDIT: I looked it up on Wikipedia. Apparently it's some ancient Babylonian unit for angles :/

vaguelyhumanoid
Posts: 86
Joined: Fri Jun 18, 2010 4:27 am UTC

Re: Interesting features in your conlang

Postby vaguelyhumanoid » Wed Oct 13, 2010 1:02 am UTC

On my way home from school, I thought about the idea of a language where prepositions are expressed entirely through cases and pronouns are expressed entirely through polypersonal verbs. If you wanted to say "I saw a dog under a tree", you'd say "dog-SUBE tree-OBJ I-heard-it". I was also considering a language with both Semitic-style consonant roots and Georgian-style screeves, as well as maybe particles and/or noun cases or even PIE-style "stem" suffixes, but then I realized it would just be second-rate Ithkuil.
I still swear that I didn't know that "šiv" is the Slavic stem for "live". Really, "živat" meaning "to live" is a total coincidence, would I lie to you?
Spoiler:
tesseraktik wrote: of course you need to gornax your frifftop to a proper taibou (which, as the construction of this tempered tutatu suggests, consists of two bed.pans joined by a haiku), or else angry zubat are going to flork off your penis.'

User avatar
Iulus Cofield
WINNING
Posts: 2917
Joined: Wed Apr 07, 2010 9:31 am UTC

Re: Interesting features in your conlang

Postby Iulus Cofield » Fri Mar 04, 2011 7:01 am UTC

I've started work on a language. It's for Discordianism so it's going to be intentionally complex and I hope to make everything relate to the number five in some arbitrary way.
I'm going to try to not have any prepositions at all and hopefully I can swing 55 cases. Has any language ever used a separate case for a predicate noun? I'm farming Wikipedia and Greek Grammar Beyond the Basics for as many types of grammatical functions for nouns as I can.

There are also no vowels, only stops and fricatives.

User avatar
Roĝer
Posts: 445
Joined: Sun Oct 05, 2008 9:36 pm UTC
Location: Many worlds, but mostly Copenhagen.
Contact:

Re: Interesting features in your conlang

Postby Roĝer » Fri Mar 04, 2011 1:32 pm UTC

Iulus Cofield wrote:Has any language ever used a separate case for a predicate noun?


I believe Polish does this, and some other slavic languages too probably.
Ik ben niet koppig, ik heb gewoon gelijk.

Makri
Posts: 654
Joined: Sat Oct 03, 2009 4:57 pm UTC

Re: Interesting features in your conlang

Postby Makri » Fri Mar 04, 2011 1:47 pm UTC

Russian has nominative and instrumental predicative nouns with a slight semantic difference. I believe Polish is the same. So they don't have a separate case for this, but at least they have non-nominative predicative nouns.
¬□(∀♀(∃♂(♀❤♂)⟷∃♂(♂❤♀)))

Derek
Posts: 2179
Joined: Wed Aug 18, 2010 4:15 am UTC

Re: Interesting features in your conlang

Postby Derek » Sat Mar 05, 2011 4:36 pm UTC

Iulus Cofield wrote:I've started work on a language. It's for Discordianism so it's going to be intentionally complex and I hope to make everything relate to the number five in some arbitrary way.
I'm going to try to not have any prepositions at all and hopefully I can swing 55 cases. Has any language ever used a separate case for a predicate noun? I'm farming Wikipedia and Greek Grammar Beyond the Basics for as many types of grammatical functions for nouns as I can.

There are also no vowels, only stops and fricatives.

Wouldn't 25 = 5*5 be more "related" to the number 5? And a bit easier to pull off :P

You should somehow make it have five numbers. Singular, dual, plural, I don't know what else, maybe a zero number?

User avatar
Iulus Cofield
WINNING
Posts: 2917
Joined: Wed Apr 07, 2010 9:31 am UTC

Re: Interesting features in your conlang

Postby Iulus Cofield » Sat Mar 05, 2011 9:21 pm UTC

Oooh, a zero number for "no one" is the subject?

I'm thinking 55 will actually be easier than 25 since I'm going for no prepositions. Think of all the different uses of genitive constructions alone.

Derek
Posts: 2179
Joined: Wed Aug 18, 2010 4:15 am UTC

Re: Interesting features in your conlang

Postby Derek » Sun Mar 06, 2011 5:26 am UTC

Iulus Cofield wrote:Oooh, a zero number for "no one" is the subject?

Yeah, like "No pigs fly" -> "Pigs-(zero) fly". I can't think of a good fifth number though.

User avatar
Iulus Cofield
WINNING
Posts: 2917
Joined: Wed Apr 07, 2010 9:31 am UTC

Re: Interesting features in your conlang

Postby Iulus Cofield » Sun Mar 06, 2011 8:02 am UTC

Trial or quadral are the obvious choices. I think I'm going to go with zero, singular, paucal (a few), collective plural (like mass nouns), and distributed plural (for groups composed of individuals). Thanks for the ideas! I think I'm moving in a highly synthetic direction here.

User avatar
Aberacht
Posts: 22
Joined: Tue Jul 20, 2010 6:11 am UTC

Re: Interesting features in your conlang

Postby Aberacht » Sat Mar 12, 2011 4:42 am UTC

Hello all, I have a linguistic question, but since it pertained to my own personal conlang I felt I should better put it here rather than make a new topic.

My vague question is: What defines a syllable? I am currently focusing heavily on phonology, and I'm trying to experiment with strange consonantal clusters.

Is the definition of a syllable something like "all that can be said within a continuous expelling of air" or something technical like that, or does it truly vary between languages? For instance,

/brɛd/ is clearly a monosyllabic word in English. /b/ plus /r/ is a legal consonant combination. However,
/nupv/ is not allowed in English. To English-speakers, saying /v/ immediately after /p/ just sounds quirky and wrong.

However, is there a real, phonological reason why this is illegal, or is this merely a rule for English alone? When I try to pronounce my experimental word /jefm.fjeʃt.ʊd/ (éfmféśtod in standard spelling), to me it seems like /jefm/ might be a possible syllable; itś just likely that since I'm not used to it it doesn't roll off the tongue easily.

However, am I fooling myself? Am I really saying /jef.m/ with a small gap? What defines a syllable?

Thanks for your time.

User avatar
Iulus Cofield
WINNING
Posts: 2917
Joined: Wed Apr 07, 2010 9:31 am UTC

Re: Interesting features in your conlang

Postby Iulus Cofield » Sat Mar 12, 2011 6:19 am UTC

Every syllable has a nucleus, composed of a segment that can be defined as [+syllabic]. There isn't a universal phonetic rule on what makes a segment [+syllabic], but I'll try to give you something to work with.

A nucleus is the most sonorant segment in a group of sounds. Sonorant basically means loud. There is a hierarchy of sonority with vowels being the most sonorant, then glides/semi-vowels, then approximants/liquids, then nasals, then obstruents. This spectrum can also be thought of in terms of least airflow constriction to most airflow constriction, with vowels having almost no constriction and obstruent stops having total constriction. Voiced segments are also more sonorant than unvoiced segments.

Every natural language (AFAIK) allows a vowel to function as a nucleus and considerably fewer languages allow approximants or nasals to. English is among the latter and most phoneticians agree that /m/, /n/, /r/, and /l/ can function as nuclei in English. Obstruents can even function as nuclei, but this is very, very rare. Berber languages are the only natural languages that I've heard of doing this.

Syllables optionally have either an onset, segment(s) preceding the nucleus, or a coda, segment(s) following the nucleus, or both. I've never heard of a language that totally forbids onsets, but many languages have either no codas or highly restricted codas. All languages have some phonological restrictions on what sequences of segments can make up codas or onsets. These restrictions can be pretty arbitrary in terms of what is actually able to be produced by a vocal tract. English, for example, can not have a non-sibilant fricative precede a stop in an onset, so /sp-/ is possible, but /fp-/ is not. Greek, on the other hand, can have fricatives precede stops in onsets, so both /fp-/ and /sp-/ are entirely possible.

Generally, the sonority of segments in a syllable will go from least sonorant to most sonorant (the nucleus) to least sonorant. /trans/ is an example of this in English. If you treat fricatives and stops together as obstruents, almost all syllables in the world follow this pattern. It's less common if you treat fricatives as slightly more sonorant than stops, /stops/ being a good example of a word that goes very-low sonority --> slightly less sonority --> really sonorant --> barely sonorant --> slightly more sonorant. Russian and Persian have some words, such as /qæbl/ ("before"), which breaks this pattern more strongly.

Defining a syllable as "all that can be said within a continuous expelling of air" is by no means the accepted one. We don't really stop expelling air for any substantial period of time in speech until we're done talking or pausing to think and in those utterances we're going to have a number of segments with equal or near-equal sonority and we're certainly going to have segments that are defined as phonologically [+syllabic] by the rules of the language we're speaking.

I can go into how to break up a series of sounds into syllables in detail if you want, but it usually comes down to giving each [+syllabic] segment as large of an onset as the language will allow.

For an example of unusual syllabic structure, the language I'm working on is pretty good. There are only stops (S) and fricatives (F), so the whole thing has very, very low sonority. To keep it from being straight up insane, I've made all syllables either F or SF, that is, I've restricted the maximum onset to a single stop and forbidden codas. This alone really isn't enough to make it easy though. If there were two F syllables in a row and one is voiced and the other is unvoiced, then there would be a very uneven sonority in the [+syllabic] criteria. So I've made all of my F's phonemically voiceless, only becoming [+voice] if they follow a voiced stop. If there is a voiced SF syllable preceding an F syllable, an epenthetic glottal stop breaks them up, making the phonemically F syllable into a phonetic SF syllable. The sonority is still wonky, but there are never two nuclei adjacent to each other with different levels of sonority.

The short answer to your question is, yes, /jefm/ is a possible syllable, but an unlikely one, mostly because /m/ is substantially more sonorant than /f/, but it is still possible.

User avatar
Aberacht
Posts: 22
Joined: Tue Jul 20, 2010 6:11 am UTC

Re: Interesting features in your conlang

Postby Aberacht » Sat Mar 12, 2011 6:31 am UTC

Iulus Cofield wrote:...

Thank you very much, Iulus! Your response was very informative and enlightening. I had known a bit about sonority and the other concepts you elaborated on, but I had not known how they interacted with one another in the ways you've described. Thanks to you, I can begin to rewrite the phonotactic rules of my language appropriately. Again, many thanks!

User avatar
Iulus Cofield
WINNING
Posts: 2917
Joined: Wed Apr 07, 2010 9:31 am UTC

Re: Interesting features in your conlang

Postby Iulus Cofield » Sat Mar 12, 2011 6:38 am UTC

No problem. Phonology is one of the few things I love more than ponies.

User avatar
Aberacht
Posts: 22
Joined: Tue Jul 20, 2010 6:11 am UTC

Re: Interesting features in your conlang

Postby Aberacht » Sat Mar 12, 2011 7:12 am UTC

Eek, you were right about language tending to minimize the coda. My "experiment" with consonantal cluster codas has led to something which even after ten repetitions I still can't say smoothly:

/ɲəpt.ren/ /jefp.fjeʃt.ʊd/ /əmt.çeh/ /ɟjɑs/ /ehvd/

At least it has been harmonized by sonority; previously, the phrase was filled with difficult to pronounce clusters of improperly ordered sonority. Now I will look into onset clusters rather than coda ones...

Edit: Another side note, to contribute to the thread idea of "interesting features." My language contains four vowels, e, ʊ, ə, and ɑ - however, each vowel also has a palatalized complement: je, jʊ, jə, and jɑ. The /j/ sound is forbidden from appearing anywhere else outside of these vowel combinations.

User avatar
Roĝer
Posts: 445
Joined: Sun Oct 05, 2008 9:36 pm UTC
Location: Many worlds, but mostly Copenhagen.
Contact:

Re: Interesting features in your conlang

Postby Roĝer » Sat Mar 12, 2011 10:00 am UTC

Iulus Cofield wrote:For an example of unusual syllabic structure, the language I'm working on is pretty good. There are only stops (S) and fricatives (F), so the whole thing has very, very low sonority. To keep it from being straight up insane, I've made all syllables either F or SF, that is, I've restricted the maximum onset to a single stop and forbidden codas. This alone really isn't enough to make it easy though. If there were two F syllables in a row and one is voiced and the other is unvoiced, then there would be a very uneven sonority in the [+syllabic] criteria. So I've made all of my F's phonemically voiceless, only becoming [+voice] if they follow a voiced stop. If there is a voiced SF syllable preceding an F syllable, an epenthetic glottal stop breaks them up, making the phonemically F syllable into a phonetic SF syllable. The sonority is still wonky, but there are never two nuclei adjacent to each other with different levels of sonority.


You are constructing a language without vowels? I'm pretty sure that's not found in any natural language.
Ik ben niet koppig, ik heb gewoon gelijk.

Makri
Posts: 654
Joined: Sat Oct 03, 2009 4:57 pm UTC

Re: Interesting features in your conlang

Postby Makri » Sat Mar 12, 2011 10:06 am UTC

Another side note, to contribute to the thread idea of "interesting features." My language contains four vowels, e, ʊ, ə, and ɑ - however, each vowel also has a palatalized complement: je, jʊ, jə, and jɑ. The /j/ sound is forbidden from appearing anywhere else outside of these vowel combinations.


Restricting /j/ to prevocalic contexts (because this is effectively what you do) doesn't strike me as very special... What's more remarkable is the total absence of /i/.
¬□(∀♀(∃♂(♀❤♂)⟷∃♂(♂❤♀)))

User avatar
Aberacht
Posts: 22
Joined: Tue Jul 20, 2010 6:11 am UTC

Re: Interesting features in your conlang

Postby Aberacht » Sat Mar 12, 2011 6:55 pm UTC

Makri wrote:
Another side note, to contribute to the thread idea of "interesting features." My language contains four vowels, e, ʊ, ə, and ɑ - however, each vowel also has a palatalized complement: je, jʊ, jə, and jɑ. The /j/ sound is forbidden from appearing anywhere else outside of these vowel combinations.


Restricting /j/ to prevocalic contexts (because this is effectively what you do) doesn't strike me as very special... What's more remarkable is the total absence of /i/.

While /i/ is really common, I felt like it isn't necessary to have it in every language.

User avatar
Iulus Cofield
WINNING
Posts: 2917
Joined: Wed Apr 07, 2010 9:31 am UTC

Re: Interesting features in your conlang

Postby Iulus Cofield » Sat Mar 12, 2011 8:46 pm UTC

Roĝer wrote:You are constructing a language without vowels? I'm pretty sure that's not found in any natural language.


It isn't as far as I'm aware. I feel like a mad scientist. IT'S PRONOUNCEABLE! PRONOUNCEABLE!

User avatar
jaap
Posts: 2087
Joined: Fri Jul 06, 2007 7:06 am UTC
Contact:

Re: Interesting features in your conlang

Postby jaap » Sat Mar 12, 2011 9:00 pm UTC

Iulus Cofield wrote:
Roĝer wrote:You are constructing a language without vowels? I'm pretty sure that's not found in any natural language.
It isn't as far as I'm aware. I feel like a mad scientist. IT'S PRONOUNCEABLE! PRONOUNCEABLE!

I bet a speaker of this language would be good at beatboxing.

Sheikh al-Majaneen
Name Checks Out On Time, Tips Chambermaid
Posts: 1075
Joined: Fri Jan 01, 2010 5:17 am UTC

Re: Interesting features in your conlang

Postby Sheikh al-Majaneen » Wed Mar 16, 2011 7:35 am UTC

jaap wrote:
Iulus Cofield wrote:
Roĝer wrote:You are constructing a language without vowels? I'm pretty sure that's not found in any natural language.
It isn't as far as I'm aware. I feel like a mad scientist. IT'S PRONOUNCEABLE! PRONOUNCEABLE!

I bet a speaker of this language would be good at beatboxing.

The beat, it speaks to me!

User avatar
eSOANEM
:D
Posts: 3652
Joined: Sun Apr 12, 2009 9:39 pm UTC
Location: Grantabrycge

Re: Interesting features in your conlang

Postby eSOANEM » Fri Jan 27, 2012 5:27 pm UTC

I know this thread's been dead a while, but I thought it better to necro than start a new thread.

I've recently been working on a medieval(ish) fantasy conworld with an array of conlangs (most of which belong to the same family and so are derived from one proto-language) to go with it. In one of the conlangs outside this main family, I decided to have a play with tense and aspect so the first thing I did was add distant past and future tenses and throw out the present.

So far, so ordinary.

Where I think it gets interesting is that then I decided to give each verb two tenses, specifying the time range when the action starts with a prefix and the time range when the action ends with a suffix. For perfective forms, both tenses are the same but, this formation allows for different degrees of imperfectiveness because the verb could expand in only one direction, in both directions and by varying amounts.

As an example, the different tense forms of the verb "ó" (to be) are:

kófony - was in the ancient past
kódan - had always been (was from the ancient past until the past/now)
kótum - will stop having always been (was from the ancient past until the future/now)
kófo - always has and always will be (is/was from ancient past until distant future)
thanódan - was/is
thanótum - is being
thanófo - will always be (from past/now until the distant future)
funótum - is/will be
funófo - will be being (from future/now until distant future)
banófo - will be in the distant future
my pronouns are they

Magnanimous wrote:(fuck the macrons)

Logomachist
Posts: 21
Joined: Mon Mar 08, 2010 8:28 pm UTC

Re: Interesting features in your conlang

Postby Logomachist » Fri Feb 03, 2012 2:55 am UTC

http://forums.xkcd.com/viewtopic.php?f=25&t=22017#p656648 wrote:There are whole language families in the Amazon, so I've heard [citation needed] where it's obligatory to mark the origin of the information you're stating.


Anyone know more about this? I had an idea to do this in an original language of mine but I cut it out because I thought it would be too unwieldy.

vaguelyhumanoid wrote:On my way home from school, I thought about the idea of a language where prepositions are expressed entirely through cases and pronouns are expressed entirely through polypersonal verbs. If you wanted to say "I saw a dog under a tree", you'd say "dog-SUBE tree-OBJ I-heard-it".


Why "I heard it" and not "I saw it"?

User avatar
fizzgig
Posts: 218
Joined: Tue May 18, 2010 10:35 am UTC
Location: Canberra, Australia

Re: Interesting features in your conlang

Postby fizzgig » Fri Feb 03, 2012 4:20 am UTC

The Wikipedia page on Evidentiality might be what you're looking for

Fire Brns
Posts: 1114
Joined: Thu Oct 20, 2011 2:25 pm UTC

Re: Interesting features in your conlang

Postby Fire Brns » Mon Feb 06, 2012 4:40 pm UTC

Verb tenses for want and will. "I want to drive" vs. "I will drive". Where "want" isn't necessarily desire but intention.
Pfhorrest wrote:As someone who is not easily offended, I don't really mind anything in this conversation.
Mighty Jalapeno wrote:It was the Renaissance. Everyone was Italian.

Makri
Posts: 654
Joined: Sat Oct 03, 2009 4:57 pm UTC

Re: Interesting features in your conlang

Postby Makri » Mon Feb 06, 2012 8:48 pm UTC

I don't know what this has to do with tenses for want and will, but English has a future that somehow manages to convey intention: "I'm going to drive."
¬□(∀♀(∃♂(♀❤♂)⟷∃♂(♂❤♀)))

User avatar
eSOANEM
:D
Posts: 3652
Joined: Sun Apr 12, 2009 9:39 pm UTC
Location: Grantabrycge

Re: Interesting features in your conlang

Postby eSOANEM » Mon Feb 06, 2012 10:59 pm UTC

Also, "I'll want to go to the shops tomorrow" seems perfectly natural to me so I think "want" as a modal, at least in my dialect, takes tenses freely.
my pronouns are they

Magnanimous wrote:(fuck the macrons)

Makri
Posts: 654
Joined: Sat Oct 03, 2009 4:57 pm UTC

Re: Interesting features in your conlang

Postby Makri » Tue Feb 07, 2012 10:13 am UTC

Ah, by the way, a small comment on the excessive tense-aspect system you posted above:

You must be aware that "is being" is, in a way, an English speciality. The verb "be" doesn't form progressives in any other language I'm aware of, and of course in the progressive, it doesn't mean "be" but "behave". Other stative verbs don't form progressives even in English. So I think in order to illustrate a tense/aspect system with a progressive, you should chose an eventive verb like "come", and not "be".

Also, the lack of distinction between past and present in the face of such a weird aspect system is, to say the least, disturbing. :mrgreen:
¬□(∀♀(∃♂(♀❤♂)⟷∃♂(♂❤♀)))

User avatar
eSOANEM
:D
Posts: 3652
Joined: Sun Apr 12, 2009 9:39 pm UTC
Location: Grantabrycge

Re: Interesting features in your conlang

Postby eSOANEM » Tue Feb 07, 2012 4:11 pm UTC

Yeah, I was struggling to with what verb to use. I think I'm better at coming up with interesting grammars than I am generating vocab, so at the moment I only have three verbs (the conlang is very much a work in progress), "to be", "to be called" and "to inherit" and I thought it would be even stranger to use "to be called" or "to inherit", ideally I'd have used something like "to dwell" which fits nicely semantically with various degrees of progressive/perfictiveness.

As for the lack of a present tense, the main reason for that is I thought that a person cannot really comment on what is, only on what was recently, will be shortly or what was recently and will still be shortly and so I figured it would be fun to try a language without it. The other language I'm working on atm (which is/will be the root of most of the others) doesn't have tense (the idea is that, in this conworld, this language was the language of the gods and so when would seem an odd concept to have) and only has aspect (although its daughters will probably innovate tense either from aspects or through metaphorical prepositional phrases).
my pronouns are they

Magnanimous wrote:(fuck the macrons)

Makri
Posts: 654
Joined: Sat Oct 03, 2009 4:57 pm UTC

Re: Interesting features in your conlang

Postby Makri » Wed Feb 08, 2012 5:02 am UTC

Well, you have a present progressive, so why no simple present? Stative verbs in simple present are just like eventive verbs in present progressive. Also, what about habituals and geneics?
¬□(∀♀(∃♂(♀❤♂)⟷∃♂(♂❤♀)))

User avatar
eSOANEM
:D
Posts: 3652
Joined: Sun Apr 12, 2009 9:39 pm UTC
Location: Grantabrycge

Re: Interesting features in your conlang

Postby eSOANEM » Wed Feb 08, 2012 7:28 am UTC

Makri wrote:Well, you have a present progressive, so why no simple present? Stative verbs in simple present are just like eventive verbs in present progressive. Also, what about habituals and geneics?


Because of this weird tense/aspect system, the varying degrees of progressive are lines between our four points on the timeline (ancient, past, future, distant future) meaning that all the 1st order progressives (which will be the ones between two adjacent tenses) have their middle outside of one of the other tenses however this doesn't make the past-future progressive a present progressive.

An explanation:

Person A tells me they'll be going to café blah for lunch at 12:30
At 12:30, person B asks me where person A is
I say "they're eating lunch at café blah" using a present progressive

In this tense/aspect system, because they didn't start the eating in the past, I'd respond with:

"they will eat lunch at café blah"

All languages have tenses which cover every point on a timeline, it's just a question of how big a range each tense covers, in this conlang, there isn't a present, but the past and future cover everything right up to it.

I guess this gives the tense/aspect system a slight hint of evidentiality. If I've been told something will happen at a certain time; when it gets to that time, I have no evidence that it was happening so I'd use the future whereas if someone was doing something and said they'd be doing it for an hour more and, half an hour later I'm asked what they're doing, I have evidence it was happening and that it will still be happening so would use the past-future tenspectTM.

Which verbs are stative and which eventive depends on the language so it seems reasonable to me for this conlang to only use one (doesn't matter which) such that the tenspect system works.

As for habituals, there isn't any morphological way of expressing them and I don't know whether I'll do it ultimately with some modal (like "to be used to") or an adverb. And I'm not sure what you mean be geneics so I'm afraid I can't answer.
my pronouns are they

Magnanimous wrote:(fuck the macrons)

Makri
Posts: 654
Joined: Sat Oct 03, 2009 4:57 pm UTC

Re: Interesting features in your conlang

Postby Makri » Wed Feb 08, 2012 12:26 pm UTC

Person A tells me they'll be going to café blah for lunch at 12:30
At 12:30, person B asks me where person A is
I say "they're eating lunch at café blah" using a present progressive

In this tense/aspect system, because they didn't start the eating in the past, I'd respond with:

"they will eat lunch at café blah"


I see two ways to interpret that. First, with an evidential future, which exists in English, but would require a progressive in this case: "they will be eating lunch at café blah right now" This is perfectly straightforward, and an evidential future exists in quite a few languages. But using evidential futures for every present tense makes no sense (and will make every child learning the language turn your future morpheme into a present tense morpheme).

The other idea I had is a perfect, but turned into the future. The analogue of "they have been eating", extending into the future instead of the past, but including the present. So it means that they are eating now and will be eating for some time from now on. I'm not sure how this would be distinguishable from an ordinary non-past tense.
¬□(∀♀(∃♂(♀❤♂)⟷∃♂(♂❤♀)))

User avatar
eSOANEM
:D
Posts: 3652
Joined: Sun Apr 12, 2009 9:39 pm UTC
Location: Grantabrycge

Re: Interesting features in your conlang

Postby eSOANEM » Wed Feb 08, 2012 5:38 pm UTC

Makri wrote:
Person A tells me they'll be going to café blah for lunch at 12:30
At 12:30, person B asks me where person A is
I say "they're eating lunch at café blah" using a present progressive

In this tense/aspect system, because they didn't start the eating in the past, I'd respond with:

"they will eat lunch at café blah"


I see two ways to interpret that. First, with an evidential future, which exists in English, but would require a progressive in this case: "they will be eating lunch at café blah right now" This is perfectly straightforward, and an evidential future exists in quite a few languages. But using evidential futures for every present tense makes no sense (and will make every child learning the language turn your future morpheme into a present tense morpheme).

The other idea I had is a perfect, but turned into the future. The analogue of "they have been eating", extending into the future instead of the past, but including the present. So it means that they are eating now and will be eating for some time from now on. I'm not sure how this would be distinguishable from an ordinary non-past tense.


My thinking behind this is based, almost entirely, on the idea that two two tenses of the verb mark the onset and end of the action with any other thinking coming from the idea that a given construction would be too complicated/require too much thinking and so simplifying in a way which seems "native" to me. There is no particular idea in my head about whether a given tenspect is perfect or evidential and when I described my usage as evidential, that was a quick analysis of where my attempt at thinking natively had left me.

So, the reason I said "they will eat lunch at café blah" instead of "they're eating lunch at café blah" was mainly based on the fact that, from the point of view of the speaker, the action may or may not have begun, but will end in the future, clarifying the statement as "they are eating/will eat lunch at café blah" would be a more precise translation but, as it requires two verbs and conjunctions seems ugly and un-native to me.

This leaves me in the position of having to choose between "they past-eat-future lunch at café blah" or "they future-eat-future lunch at café blah" (roughly "they are eating lunch at café blah" vs. "they will eat lunch at café blah") and decided that a confidence based distinction (whereby if the speaker is confident "they" have started eating they'd use the past-future tenspect and if they are not confident they'd use the future-future tenspect). This example was possibly a bad one because the speaker was told they'd start eating lunch at 12:30 which is also when the speaker is asked where the people are; in this case, the speaker isn't confident either way, a few minutes earlier or later and they would be, but as it is they're not so I decided to use an evidential distinction as the final decider.

So, if person A is the speaker, person B the eater and person C is looking for B we have the following situations:

B: A, I will be eating lunch at 12:30
*B walks off*
*C arrives at 12:00*
C: A, do you know where B is?
A: B will eat lunch at café blah (A is confident B has not started eating lunch)

B: A, I will be eating lunch at 12:30
*B walks off*
*at 12:00 A walks through the café and sees B about to eat his lunch*
*C arrives at 12:05*
C: A, do you know where B is?
A: B past-eat-future lunch at café blah (even though B told A he wouldn't be eating until 12:30, A saw B eating so is confident that B started eating in the past but that he has not finished)

B: A, I will be eating lunch at 12:30
*B walks off*
*C arrives at 12:30*
C: A, do you know where B is?
A: B will eat lunch at café blah (A isn't confident B has started eating neither is A confident B hasn't so A bases their statement on the evidence at their disposal that B will be eating)

B: A, I will be eating lunch at 12:30
*B walks off*
*C arrives at 12:40*
C: A, do you know where B is?
A: B past-eat-future lunch at café blah (A is confident B has started eating lunch)

B: A, I will be eating lunch at 12:30
*B walks off*
*at 12:35 A walks through the café and notices that B isn't there eating their lunch*
*C arrives at 12:40*
C: A, do you know where B is?
A: B will eat lunch at café blah (even though B told A he'd be eating at 12:30, A saw that B wasn't eating their lunch but is confident B didn't eat their lunch early so, based on the evidence at A's disposal decides that B has yet to start eating)

Hopefully that should clarify my thought process a little. It's almost entirely based on the speaker's confidence on whether or not the action has already started as to whether or not the past is included and their confidence on whether or not the action has already ended which determines whether or not the future is included (for "present" tense constructions).
my pronouns are they

Magnanimous wrote:(fuck the macrons)

Makri
Posts: 654
Joined: Sat Oct 03, 2009 4:57 pm UTC

Re: Interesting features in your conlang

Postby Makri » Thu Feb 09, 2012 7:26 am UTC

Ah, now I finally see what's going on there! I don't think it has anything to do with confidence, nor with evidentiality. In essense, your past-...-future just is the present tense.

"B past-eat-future lunch at café blah" just means "Be is eating lunch at café blah". (In case you think that the past-...-future will be combined with "since" and that makes it more like a present perfect progressive - not necessarily. Many languages use the adverb "since" with the simple present tense.)

So if I understand correctly:

k- "starts in the distant past"
than- "starts in the non-distant past"
fun- "starts in the non-distant future"
ban- "starts in the distant future"

-fony "ends in the distant past"
-dan "ends in the non-distanct past"
-tum "ends in the non-distant future"
-fo "ends in the distant future"

Now, of course, that's not a possible system in a natural language, but I see the logic now. And in fact, it has no progressive. That was an artefact of the translation.

The obvious question is now: What do you do with temporal adverbs? What time do they associate with? (I suppose the obvious - iconic - solution would be having adverb positions before and after the verb and associating them accordingly.)
¬□(∀♀(∃♂(♀❤♂)⟷∃♂(♂❤♀)))

User avatar
eSOANEM
:D
Posts: 3652
Joined: Sun Apr 12, 2009 9:39 pm UTC
Location: Grantabrycge

Re: Interesting features in your conlang

Postby eSOANEM » Thu Feb 09, 2012 7:12 pm UTC

Makri wrote:Ah, now I finally see what's going on there! I don't think it has anything to do with confidence, nor with evidentiality. In essense, your past-...-future just is the present tense.

"B past-eat-future lunch at café blah" just means "Be is eating lunch at café blah". (In case you think that the past-...-future will be combined with "since" and that makes it more like a present perfect progressive - not necessarily. Many languages use the adverb "since" with the simple present tense.)

So if I understand correctly:

k- "starts in the distant past"
than- "starts in the non-distant past"
fun- "starts in the non-distant future"
ban- "starts in the distant future"

-fony "ends in the distant past"
-dan "ends in the non-distanct past"
-tum "ends in the non-distant future"
-fo "ends in the distant future"

Now, of course, that's not a possible system in a natural language, but I see the logic now. And in fact, it has no progressive. That was an artefact of the translation.

The obvious question is now: What do you do with temporal adverbs? What time do they associate with? (I suppose the obvious - iconic - solution would be having adverb positions before and after the verb and associating them accordingly.)


1. That's probably a better way of putting it. In my head it's a little more nuanced, but the past-future tenspect = present is certainly very close to what's in there.

2. Translating past-past/past-future/future-future tenspects into English is a lot easier than going the other way round but, as a rule of thumb, present -> past-future is probably fairly accurate.

3. Almost, there are a few irregularities because the verb doesn't have an onset or coda (and is the same vowel as in some of the affixes so k- is actually ko- but os merge into a single o (there are no long vowels) and -fony and -fo are actually -ony and -o but the root acquires a "f" to stop vowel merging (although this seems a little unnecessary and I may drop this).

4. I'm not sure why it's not possible. Morphologically it's no more complicated than an agluttinating language with both tense and aspect and it has an internal logic. As it stands so far, it's far more rigid than any natural language would be but I would not expect it to be used to its full extent in many cases at all and would expect (and so would try to emulate) a simplified system with many of the combinations possible in the "full" system only used when precision is important.

5. By temporal adverbs I presume you mean things like "yesterday", "last year", "later" etc. in which case I have not currently decided. The language is OSV and left branching so adverbs appear before the verb, the main question I guess is whether or not the adverb would shift the temporal focus (so the "past" part of the tenspect would become the "before" part etc. e.g. He will be eating his lunch at 12:30 -> lunch he 12:30 past-eat-future) or not (he will be eating his lunch at 12:30 -> lunch he 12:30 future-eat-future). Shifting the temporal focus seems more interesting and would allow for more specific specification of tenspectbut would be in danger of making it a bit of a kitchen sink language.

I'm pretty sure I don't want temporal adjectives to take the place of the tenspect affixes though because, in order to try and be consistent with the normal affixes I'd need different prefix and suffix forms which would be a pain.
my pronouns are they

Magnanimous wrote:(fuck the macrons)

Makri
Posts: 654
Joined: Sat Oct 03, 2009 4:57 pm UTC

Re: Interesting features in your conlang

Postby Makri » Thu Feb 09, 2012 10:55 pm UTC

I'm not sure why it's not possible.


Because it's a language universal that there is one tense node in the syntax and one time argument in the semantics. That time is always set in a relation to the runtime of an event, whereas your system never makes reference to runtimes, but only to starting and end points.
And of course, every natural language has a semantic present. If it were to be acquired by a child, your system would collapse in certain ways.

He will be eating his lunch at 12:30 -> lunch he 12:30 past-eat-future


How is "he will be eating his lunch" translated by "he past-eat-future"? And the question with the 12:30 is: does it mean that he started at 12:30, or that he will finish at 12:30? There is no way you could express that at 12:30, he will be eating - because you make reference only to beginning and end.
¬□(∀♀(∃♂(♀❤♂)⟷∃♂(♂❤♀)))

User avatar
eSOANEM
:D
Posts: 3652
Joined: Sun Apr 12, 2009 9:39 pm UTC
Location: Grantabrycge

Re: Interesting features in your conlang

Postby eSOANEM » Fri Feb 10, 2012 7:44 am UTC

Makri wrote:
I'm not sure why it's not possible.


Because it's a language universal that there is one tense node in the syntax and one time argument in the semantics. That time is always set in a relation to the runtime of an event, whereas your system never makes reference to runtimes, but only to starting and end points.
And of course, every natural language has a semantic present. If it were to be acquired by a child, your system would collapse in certain ways.

He will be eating his lunch at 12:30 -> lunch he 12:30 past-eat-future


How is "he will be eating his lunch" translated by "he past-eat-future"? And the question with the 12:30 is: does it mean that he started at 12:30, or that he will finish at 12:30? There is no way you could express that at 12:30, he will be eating - because you make reference only to beginning and end.


1. I didn't realise that was the case. Still, without getting Sapir-Whorfian that doesn't necessarily mean that such a system could not be acquired, particularly not, as this is for a conworld, by members of a different species. As for semantic presents, this conlang would include them, just not distinguish entirely between them and a couple of other tenses (in a similar situation to Finnish).

2. That was an idea that, if a temporal adverb was used, it would shift the temporal focus to that time so that the past/future divide becomes a before/after divide around the time specified by the adverb. I'm not sure whether to use that system or just to let adverbs narrow down when the tense refers to exactly though.
my pronouns are they

Magnanimous wrote:(fuck the macrons)


Return to “Language/Linguistics”

Who is online

Users browsing this forum: No registered users and 8 guests