Conlanging Features you Avoid

A forum for all topics related to constructed languages
User avatar
qwed117
mongolian
mongolian
Posts: 3964
Joined: 20 Nov 2014 02:27

Re: Conlanging Features you Avoid

Post by qwed117 »

Adarain wrote:
Keenir wrote:
Adarain wrote:Oh yea, triconsonantal roots I've been avoiding so far. They're such a Semitic thing, imo any language that has them should be an a posteriori one (unless someone can point me to an unrelated language [both according to tree and wave model] that also has something similar, in which case I'll take it all back)
if I may ask, what are tree and wave?

times like this, I wonder if triconsonantal roots are an artifact of analysis and-or the script.
Basically, the tree model explains how German and English share many similar things because they are on the same branch of the language tree. The wave model explains why Romanian has many shared features with other Balkan languages, because they're near each other they may affect each other.

So basically, can someone show me a natural language which is neither closely related to the semitic languages, nor has been in heavy contact with one, that features consonantal roots.
Ablaut's pretty similar. I'm making a language that has triconsonantal roots, but uses them in a manner far different from Semitic languages.
Spoiler:
My minicity is Zyphrazia and Novland
What is made of man will crumble away.

The SqwedgePad
User avatar
eldin raigmore
korean
korean
Posts: 6051
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Conlanging Features you Avoid

Post by eldin raigmore »

Adarain wrote:Oh yea, triconsonantal roots I've been avoiding so far. They're such a Semitic thing, imo any language that has them should be an a posteriori one (unless someone can point me to an unrelated language [both according to tree and wave model] that also has something similar, in which case I'll take it all back)
IIRC Milewski (the famous Americanist) wrote about some Native American language that he thought had a triconsonantal root system.
If he was right, that could neither have been a familial shared-feature, nor an areal shared-feature.
It was many years ago, so, maybe his notion has been rejected since then.
OTOH maybe he never said that; maybe I misremembered what he said.

But if you're serious about finding a natlang 3Cons that's not Afro-Asiatic, you could start with Milewski.

BTW 3Cons natlangs are, for the most part, a subset of Afro-Asiatic languages that's a superset of Semitic languages. So even without Milewski's "Amerind"(? not sure Milewski would have used this term) exception(s), there are some non-Semitic natural triconsonantal-root-system languages. Granted, though, they're "genetically" the next thing to Semitic.
User avatar
qwed117
mongolian
mongolian
Posts: 3964
Joined: 20 Nov 2014 02:27

Re: Conlanging Features you Avoid

Post by qwed117 »

eldin raigmore wrote:
Adarain wrote:Oh yea, triconsonantal roots I've been avoiding so far. They're such a Semitic thing, imo any language that has them should be an a posteriori one (unless someone can point me to an unrelated language [both according to tree and wave model] that also has something similar, in which case I'll take it all back)
IIRC Milewski (the famous Americanist) wrote about some Native American language that he thought had a triconsonantal root system.
If he was right, that could neither have been a familial shared-feature, nor an areal shared-feature.
It was many years ago, so, maybe his notion has been rejected since then.
OTOH maybe he never said that; maybe I misremembered what he said.

But if you're serious about finding a natlang 3Cons that's not Afro-Asiatic, you could start with Milewski.

BTW 3Cons natlangs are, for the most part, a subset of Afro-Asiatic languages that's a superset of Semitic languages. So even without Milewski's "Amerind"(? not sure Milewski would have used this term) exception(s), there are some non-Semitic natural triconsonantal-root-system languages. Granted, though, they're "genetically" the next thing to Semitic.
Yeah, it appears as if vowel harmony and ablaut occur in Valley Yokuts, which is what google tells me Milewski thought had triconsonantal features.
Sm ppl rgy tht nglsh hs sm dgr f cnsnntl rts n sm wrds, bts qut crtn tht tht thnkng s ncrrct. Fr xmpl cld nd clld r ndstngushble wtht vwls, bt mn smthng vry dffrnt. Thr prnncitin s vry dffrnt t.

Anyways, many pronunciations have probably merged. Arabic strikes me as mesosynthetic.
Spoiler:
My minicity is Zyphrazia and Novland
What is made of man will crumble away.

The SqwedgePad
User avatar
eldin raigmore
korean
korean
Posts: 6051
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Conlanging Features you Avoid

Post by eldin raigmore »

qwed117 wrote: Yeah, it appears as if vowel harmony and ablaut occur in Valley Yokuts, which is what google tells me Milewski thought had triconsonantal features.
Thanks.

(3Cons strike me as "runaway ablaut".)
User avatar
Ahzoh
mongolian
mongolian
Posts: 4127
Joined: 20 Oct 2013 02:57
Location: Canada

Re: Conlanging Features you Avoid

Post by Ahzoh »

eldin raigmore wrote:(3Cons strike me as "runaway ablaut".)
That's what I also said in my explanatory post.
Image Śād Warḫālali (Vrkhazhian) [ WIKI | CWS ]
Cellular Automaton
rupestrian
rupestrian
Posts: 11
Joined: 08 Apr 2022 01:33

Re: Conlanging Features you Avoid

Post by Cellular Automaton »

I tend to avoid agglutinative or fusional languages. Most of my languages don't have full voicing contrasts, none have voiced fricatives, and I don't often use vowel systems with atypical rounding (front rounded or back unrounded vowels). As for what I do tend towards, I often find myself making languages with a defined order for the object and verb, but with a subject that can move to the start or end of a sentence. I also have a lot of heterorganic affricates in my languages, especially /t͡f/ (I just like the sound, even if I have to romanize it as <ꜰ> for SCA purposes). Reading this makes it look more like a sprachbund than a conlanger's tendencies... (also sorry for the egregious necropost, I just think this is an interesting thread!)
User avatar
Titus Flavius
sinic
sinic
Posts: 288
Joined: 13 Apr 2021 14:53
Location: Sol III

Re: Conlanging Features you Avoid

Post by Titus Flavius »

I tend to avoid isolating languages. Most of my languages have full voicing contrasts, many have voiced fricatives, and I like vowel systems with atypical rounding (front rounded or back unrounded vowels).
ω - near-close near-back unrounded vowel.
XIPA
:pol: :mrgreen:
:eng: [:)]
:esp: [:S]
:lat: [:'(]
teotlxixtli
cuneiform
cuneiform
Posts: 76
Joined: 05 Jan 2021 04:37

Re: Conlanging Features you Avoid

Post by teotlxixtli »

I’ve never liked implosives, especially the palatal implosive. They just seem like so much more work to pronounce than other consonants. Also coda /r/ (the trill, that is) annoys me personally because /i/ is hard for me to pronounce before it. And the vowel /æ/ feels like nails on a chalkboard to me for some reason

Masculine/feminine gender always seemed pedestrian to me, but I’d include them if the culture that spoke the language had more than two (either a third gender or distinct category for trans folks or something)

Languages that don’t distinguish “hand” and “arm” just seem like they’re taking the piss honestly
Cellular Automaton
rupestrian
rupestrian
Posts: 11
Joined: 08 Apr 2022 01:33

Re: Conlanging Features you Avoid

Post by Cellular Automaton »

Titus Flavius wrote: 07 May 2022 01:03 I tend to avoid isolating languages. Most of my languages have full voicing contrasts, many have voiced fricatives, and I like vowel systems with atypical rounding (front rounded or back unrounded vowels).
Together, we can make either no language or every language.
User avatar
MissTerry
hieroglyphic
hieroglyphic
Posts: 37
Joined: 14 Apr 2022 02:23

Re: Conlanging Features you Avoid

Post by MissTerry »

cybrxkhan wrote: 23 Aug 2012 05:27 Wondering if anybody is phobic towards any conlanging features.
I absolutely hate the sound the letter [r] makes at the end of a word. Like in the Spanish word "Tomar," or the American English word "Radar." All the conlangs I've ever invented since grade school were non-rhotic, where the letter and sound "R" is never the final letter or is never pronounced if it is final.

One of the features of Basha Humrayan is that most of its words end in a vowel, and so the "R" sound will never be a final sound.
User avatar
Lorik
cuneiform
cuneiform
Posts: 87
Joined: 27 Nov 2021 12:30
Location: Brazil

Re: Conlanging Features you Avoid

Post by Lorik »

Off the top of my head, I tend to avoid:
- Any sounds that I can't easily pronounce, especially glottals;
- Sibilants other than [s] and [z];
- Stress-timed languages (the conlang I'm currently working on is stress-timed, but it's the only stress-timed language I've ever made);
- The use of any letter that isn't in my Brazilian Portuguese keyboard or my Russian keyboard (makes everything a hassle to type);
- Defective writing systems;
Native: :bra: | Fluent: :eng: :fra: | Intermediate: :rus:
Image Лохдан [ɫəɣˈdã] is the conlang I'm currently working on. It's pretty different from its now abandoned predecessor, Lohdan [loʀˈdɑ̃ː].
Salmoneus
MVP
MVP
Posts: 2605
Joined: 19 Sep 2011 19:37

Re: Conlanging Features you Avoid

Post by Salmoneus »

teotlxixtli wrote: 07 May 2022 04:17 Masculine/feminine gender always seemed pedestrian to me, but I’d include them if the culture that spoke the language had more than two (either a third gender or distinct category for trans folks or something)
Just to be clear, because this is a common misconception: cross-linguistically, "gender" in the linguistic sense has nothing to do with "gender" in the sociological sense.

English is extremely strange in having residual gender marking on pronouns for humans, despite having lost gender everywhere else, and because it's only for humans (which can have social gender) and because it's not on nouns generally it's gradually become almost entirely dependent on the social gender of the human concerned (since there's nothing else left for the distinction to be guided by). Because of this, there's been concern about changing the pronouns in the few cases when they're not perceived to match the socal gender. Because it's only on pronouns, however, with no agreement phenomena, linguistic gender in English isn't actually linguistic gender anymore.

This is not how linguistic gender works in languages that actually have it. There, gender is simply a division of nouns into two categories, with agreement between the noun and other things (usually adjectives, pronouns, sometimes verbs). This makes it easier to tell which word goes with which. In Indo-European languages (and many others), "man" and "woman" happen to have different genders, and so those genders are commonly called 'masculine' and 'feminine'. But there is generally no inherent relationship between the social gender of a person and the linguistic gender of the noun used to describe them (as the same person can be referred to by many different nouns).

In Irish, for instance, the word for "girlfriend" (and indeed "girl" in general) is masculine, even though girls and girlfriends are usually socially female. Likewise, in German the word for "girl" is neuter, not feminine.

Similarly, whether a society has a third social gender (although, let's be clear: if you relegate trans people to a third category, then by definition you're not accepting them as trans) is largely unrelated to whether a language spoken by that society has a third linguistic gender.
User avatar
Lorik
cuneiform
cuneiform
Posts: 87
Joined: 27 Nov 2021 12:30
Location: Brazil

Re: Conlanging Features you Avoid

Post by Lorik »

Salmoneus wrote: This is not how linguistic gender works in languages that actually have it. There, gender is simply a division of nouns into two categories, with agreement between the noun and other things (usually adjectives, pronouns, sometimes verbs). This makes it easier to tell which word goes with which. In Indo-European languages (and many others), "man" and "woman" happen to have different genders, and so those genders are commonly called 'masculine' and 'feminine'. But there is generally no inherent relationship between the social gender of a person and the linguistic gender of the noun used to describe them (as the same person can be referred to by many different nouns).
I feel like this only tells half the story. When there is no noun to agree with, other gendered words must agree with the person's social gender. So for example, in Portuguese, the phrase "I'm tired" is said differently depending on whether it's a woman or a man who's saying it:

A woman would say:
Estou cansada.
be.1SG.PRS tired-F
I'm tired.

While a man would say:
Estou cansado.
be.1SG.PRS tired-M
I'm tired.

As you can see, there is no noun for the adjective to agree with, so it has to agree with the speaker's social gender. Another example in Portuguese is that even something as simple as "thank you" will be different depending on the speaker: a woman would say "obrigada", while a man would say "obrigado". This demonstrates that grammatical gender is often directly linked to a person's social gender. Of course, the examples I've given are in Portuguese, but this also applies to the other gendered languages I speak - French and Russian - and many other languages.

That being said, I don't think that languages with feminine/masculine gender are "pedestrian". If I did, I wouldn't make nearly all my conlangs - including the one I'm currently working on - have feminine/masculine gender.
Native: :bra: | Fluent: :eng: :fra: | Intermediate: :rus:
Image Лохдан [ɫəɣˈdã] is the conlang I'm currently working on. It's pretty different from its now abandoned predecessor, Lohdan [loʀˈdɑ̃ː].
Salmoneus
MVP
MVP
Posts: 2605
Joined: 19 Sep 2011 19:37

Re: Conlanging Features you Avoid

Post by Salmoneus »

Yes, I was simplifying somewhat, in two ways.

Firstly, the 'gender' of pronouns doesn't always match the gender system of nouns. It usually does, but not always. Going back to my examples: although cailín is masculine, and takes masculine adjectives, it actually takes the feminine pronoun. Effectively, what we can say is that the noun has a gender, and also the thing itself has a gender, and it's the latter that the pronoun agrees with. Apparently the situation is even more interesting with the German example: there, the pronoun usually (though colloquially sometimes not) agrees with the neuter gender of maedchen when the pronoun and the noun occur in the same sentence, but often does not agree when the pronoun and the noun are in different sentences.

Similarly with your Portuguese examples, the adjectives agree with the gender of the first-person pronoun, which is in turn derived from the gender of the person the pronoun refers to.

Secondly, even nouns that don't appear in the sentence can still have gender. In the Irish example, a really interesting thing happens: cailín, a masculine noun, can effectively act as a pronoun replacing more specific nouns (basically having the meaning 'thing'), if the noun it's replacing is feminine. That is, if you're talking about something feminine, you can just call it a 'girl'. But since the same thing can be referred to by different nouns, which may differ in gender, this can't just be based on some inherent property of the thing itself - there has to be an implied but unstated noun that cailín is semantically (but not grammatically!) agreeing with. So you could in theory have a masculine thing that's potentially referred to by a feminine noun, but isn't because that feminine noun has been replaced by the masculine noun 'cailín', which is then agreed with by a feminine pronoun...

I would actually suggest that this is what's happening in the other examples as well. That when you say 'cailín' or 'maedchen', you are in some sense categorising these under the heading of 'bean' or 'Frau' (woman), and that there's then a sort of 'substitution' process in which anything under a 'feminine' heading can be substituted by the feminine pronoun (even if the noun under that heading is actually masculine), or by a feminine pseudo-pronoun like the pseudo-pronominal use of 'cailín' as 'thing'. Similarly, when a Portuguese person says 'I', they're thinking of themselves, and they put themselves under a heading with a gender, and then use a substitution rule that replaces any gendered word for themselves with 'I', which is not overtly gendered but is grammatically gendered (because adjectives agree with it in gender).



HOWEVER! While clearly in some cultures the 'heading term' that humans are sorted under is now often a social gender noun, so that substitution results in linguistic agreement patterns that appear to be based on social gender, there's no reason this would necessarily happen. To take the end-point example: most cultures have social gender, but most languages do not have gender distinctions in pronouns: having no (i.e. only one) gender in pronouns doesn't preclude having two social genders, and adding a second linguistic gender doesn't make a society more likely to have a second (or third) social gender. So I think it's fair to say that social and linguistic gender aren't inherently related - though it's true that social gender is highly salient in some cultures, and therefore likely to influence how people categorise things, which can have an affect on the language indirectly.

[thought experiment: imagine a society in which social status were extremely important linguistically, and in which the word 'monarch' were feminine. We could imagine similar substitution processes to the above leading to monarchs by default being referred to as 'she'...]
User avatar
Mándinrùh
cuneiform
cuneiform
Posts: 134
Joined: 21 Aug 2016 20:37
Location: New England

Re: Conlanging Features you Avoid

Post by Mándinrùh »

Lorik wrote: 07 May 2022 12:29 - Defective writing systems;
Oh, I love defective writing systems. In Atili, you can have a word pronounced /nɨŋˈɡɛ/ that's spelled the equivalent of NILKUNEE.

On the contrary, I don't like fully predictable writing systems, because outside of languages that only just picked up writing, they don't exist in the real world. And even when Cherokee developed a new writing system from scratch, right out of the gate it didn't make all of the distinctions in Cherokee phonology.
Creator of Image Redentran
Creator of Image Bwángxùd
Creator of Image Atili
My website | My blog
User avatar
Sequor
sinic
sinic
Posts: 357
Joined: 30 Jun 2012 06:13

Re: Conlanging Features you Avoid

Post by Sequor »

Salmoneus wrote: 08 May 2022 00:21[thought experiment: imagine a society in which social status were extremely important linguistically, and in which the word 'monarch' were feminine. We could imagine similar substitution processes to the above leading to monarchs by default being referred to as 'she'...]
This reminds me, I was surprised to learn that some Arabic masculine human nouns end in ــة -a(tun) (the typical feminine singular ending), most famously خليفة xaliifa(tun) 'caliph' (plural: خلفاء xulafaaʔ(u)), and that some others take the ending in their plural pattern e.g.
- أستاذ ʔustaað '(male) professor; grammar teacher' (also simply a title of courtesy for a man) plural أساتذة ʔasaatiða(tun),
- طالب tˤaalib 'seeker, student, scholar' pl. طلبة tˤalaba(tun),
- أخ ʔax 'brother' pl. إخوة ʔixwa(tun),
- دكتور duktuur 'doctor' pl. دكاترة dakaatira(tun)
...and was equally disappointed that this didn't mean such nouns took the pronoun 'she' or other feminine agreement due to their morphology.

Also, I think I've sometimes seen Spanish "su majestad" 'Your Majesty' with feminine agreement, in spite of a masculine referent, due to the noun majestad being feminine!?? I'm not quite sure though.
hīc sunt linguificēs. hēr bēoþ tungemakeras.
Khemehekis
runic
runic
Posts: 3160
Joined: 14 Aug 2010 09:36
Location: California über alles

Re: Conlanging Features you Avoid

Post by Khemehekis »

Salmoneus wrote: 08 May 2022 00:21 Apparently the situation is even more interesting with the German example: there, the pronoun usually (though colloquially sometimes not) agrees with the neuter gender of maedchen when the pronoun and the noun occur in the same sentence, but often does not agree when the pronoun and the noun are in different sentences.
I actually remember reading something about this! They did a study of writing and utterances with the noun Mädchen, and German speakers were more likely to refer to Mädchen with the pronoun sie (she) instead of es (it) the older the girl was. So a 17-year-old girl, as they said, would be much more likely to be referred to with sie than a 7-year-old girl. Perhaps that is because (social) gender is clearer in adolescents/young adults than in children, or perhaps that is because Mädchen goes under Frau (a 17-year-old girl is a young woman) for older girls, and Frau is a feminine noun, while Mädchen goes under Kind (a 7-year-old girl is a child) for younger girls, and Kind is a neuter noun.

One of us may be able to find more information on the study with some googling.
♂♥♂♀

Squirrels chase koi . . . chase squirrels

My Kankonian-English dictionary: 79,516 words and counting

31,416: The number of the conlanging beast!
User avatar
Lorik
cuneiform
cuneiform
Posts: 87
Joined: 27 Nov 2021 12:30
Location: Brazil

Re: Conlanging Features you Avoid

Post by Lorik »

Sequor wrote: Also, I think I've sometimes seen Spanish "su majestad" 'Your Majesty' with feminine agreement, in spite of a masculine referent, due to the noun majestad being feminine!?? I'm not quite sure though.
No idea about Spanish, but I've already seen Portuguese "Vossa Majestade", which also means 'Your Majesty' and is feminine, being used that way.
Native: :bra: | Fluent: :eng: :fra: | Intermediate: :rus:
Image Лохдан [ɫəɣˈdã] is the conlang I'm currently working on. It's pretty different from its now abandoned predecessor, Lohdan [loʀˈdɑ̃ː].
User avatar
WeepingElf
sinic
sinic
Posts: 276
Joined: 23 Feb 2016 18:42
Location: Braunschweig, Germany
Contact:

Re: Conlanging Features you Avoid

Post by WeepingElf »

It is not really such that I choose to avoid features in my conlangs, but with each of my conlangs, I make a conscious decision what to use in that conlang, and I generally do not use things I feel about that I do not understand them well enough.
... brought to you by the Weeping Elf
User avatar
Creyeditor
MVP
MVP
Posts: 4539
Joined: 14 Aug 2012 19:32

Re: Conlanging Features you Avoid

Post by Creyeditor »

I was actually thinking the same. Also, I probably miss some features by chance and don't even know about it.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
Post Reply