(Conlangs) Q&A Thread - Quick questions go here

A forum for all topics related to constructed languages
User avatar
Ahzoh
mongolian
mongolian
Posts: 4106
Joined: 20 Oct 2013 02:57
Location: Canada

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Ahzoh »

My conlang has three basic verbal distinctions: the reallis, the subjunctive, and the imperative/jussive. I intend to create a tense distinction by adding a "future" prefix, so that there now becomes a basic verbal distinction of nonfuture and future. Would it then make sense for the imperative/jussive mood to turn into a subjunctive mood? I think when it stands alone it will still have imperative force, and optative force in the future, but when used with other verbs it will have a subjunctive force.

Additionally, I'm thinking of having a noun pattern to express a "modal agent" as it ultimately derives from verb froms:
paruḫtessu / paruḫtāsu "she who speaks"
parḫattessu / parḫattāsu "she who must speak"

paruḫtāsu (conjugated verb) > parāḫ-um (agent)
parḫattāsu (conjugated verb) > parḫās-um (imperative agent?)
(Not sure if these are plausible sound changes either)

I'm not sure how a mood-based agent deverbal would work.
Like I could get how it could work with tense, with your former-wives and husband-to-be's
Plausibly I could only see: ṣâb-um "killer" versus ṣābās-um "one who must be killed" > "public enemy number-one, bounty target" or "suspected killer"
Image Śād Warḫālali (Vrkhazhian) [ WIKI | CWS ]
User avatar
Creyeditor
MVP
MVP
Posts: 4336
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor »

Maybe someone who is prone to kill?
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]
User avatar
Omzinesý
runic
runic
Posts: 3079
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Omzinesý »

Ahzoh wrote: 03 Jun 2021 06:15 My conlang has three basic verbal distinctions: the reallis, the subjunctive, and the imperative/jussive. I intend to create a tense distinction by adding a "future" prefix, so that there now becomes a basic verbal distinction of nonfuture and future. Would it then make sense for the imperative/jussive mood to turn into a subjunctive mood? I think when it stands alone it will still have imperative force, and optative force in the future, but when used with other verbs it will have a subjunctive force.
One alternative is condition. "Go to the castle and you see the princess." -> 'If you go to the castle, you will see the princess.'

I think those actor nationalizations are quite lexeme-specific.
'One who must sleep' = 'sleepy one, tired one'
'One who must work' = 'one without capital income'
'One who must live' = an euphemism for 'critically ill'
'One who must read' = 'student'
'One who must masturbate' = 'one without relationship'
Now I'm stopping.
My meta-thread: viewtopic.php?f=6&t=5760
Salmoneus
MVP
MVP
Posts: 2352
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus »

Ahzoh wrote: 03 Jun 2021 06:15 My conlang has three basic verbal distinctions: the reallis, the subjunctive, and the imperative/jussive. I intend to create a tense distinction by adding a "future" prefix, so that there now becomes a basic verbal distinction of nonfuture and future. Would it then make sense for the imperative/jussive mood to turn into a subjunctive mood? I think when it stands alone it will still have imperative force, and optative force in the future, but when used with other verbs it will have a subjunctive force.
Why would it turn into a subjunctive when you already have a subjunctive?

So far as I'm aware, it's much more common for a subjunctive to become an optative/jussive/etc than vice versa, for obvious reasons: people will say "I'd have a drink" to mean "I want to have a drink", or even "I should have a drink", but people would rarely say, say, "Whether I ought to have a drink or not, I'd like a meal" to mean "Whether I'm having a drink or not..." - because fundamentally jussives and the like have more content than a bare subjunctive, and it's more common to add content than to lose it. But having said that, the reverse is certainly plausible, if less probable.

Additionally, I'm thinking of having a noun pattern to express a "modal agent" as it ultimately derives from verb froms:
paruḫtessu / paruḫtāsu "she who speaks"
parḫattessu / parḫattāsu "she who must speak"

paruḫtāsu (conjugated verb) > parāḫ-um (agent)
parḫattāsu (conjugated verb) > parḫās-um (imperative agent?)
(Not sure if these are plausible sound changes either)

I'm not sure how a mood-based agent deverbal would work.
You've just told us! For instance, "she who speaks", vs "she who must speak". Vs "she who I want to speak", "she who ought to speak", "she who might speak", etc.

There's a hint of this even in English: -ee and -ed and -ate/-ete usually mark indicative patients, while -end/-and/-endum/-andum in theory marks subjunctive patients: a conjugate has been conjoined, whereas a conjugand is to be conjoined, and so on. In English, this doesn't work well, because all these suffixes are less than fully productive, and the gerundives tend to lose their modal meaning once the word is widely adopted (dividends and addenda have usually in practice already been divided/added). But there's no reason the distinction couldn't be systematic: a 'divorcee' could be a divorced person, for instance, while a 'divorcend' could be someone who ought to be divorced, or who someone has decided to divorce.

Latin made this distinction, at least in adjectives, which could often be used substantively: "deletus", something destroyed, vs "delendus", something to be destroyed. Carthago delenda est!
Like I could get how it could work with tense, with your former-wives and husband-to-be's
Plausibly I could only see: ṣâb-um "killer" versus ṣābās-um "one who must be killed" > "public enemy number-one, bounty target" or "suspected killer"
Yes, that's a gerundive - although of course you don't need the de-productivising final derivation there. It could just mean 'one who must be killed' - an important legal concept. "The one who must be killed is to be given a last meal"; "you are one who must be killed!", etc. In the same way that "killer" just means "one who kills", and doesn't have to have semantic drift to "executioner" or "poultry farmer" or whatever.

However, be careful: the distinction you make is not primarily one of mood, but one of voice! Killer is an agent, one who must be killed is a patient. But of course you could have a modal agent: one who must kill. [for instance: "When a criminal of noble blood is to be killed, he-who-must-kill must be of equivalent rank to he-whom-must-be-killed"].

Bear in mind that one reason modal agents might not seem useful in English is that our agents are ambiguous - either indicative OR subjunctive. "Killer" can be a person who has killed or does kill, OR a person who is to kill or ought to kill, or colloquially someone who is able to kill. ["wow, that guy's really agressive!" - "yeah, he's a real killer!"... doesn't necessarily mean he HAS or even WILL kill someone, just that it seems they COULD]. But you could systematically distinguish.
So for instance, going back to patients for a moment: we can say "the killer's last victim was found in a field". "victim" here is the patient of a past indicative act. "We decoded the message - you are the next victim!" - 'victim' here is indicative future. But "the killer is on the hunt for a new victim" - 'victim' is both future and non-indicative.

In fact, you can even create an ambiguity in this way. "The killer is looking for their next victim" can have either an indicative patient (the killer has decided their next victim will be Fred, but cannot find Fred, so is looking for them), or a non-indicative patient (the killer is trying to find someone (as yet unknown to them) who should, could, might, would, be their next victim).

Similarly: "the ideal applicant has ten years of experience" - do you mean the indicative ideal applicant (an actual applicant who happens to be ideal), or the optative ideal applicant (an ideal person you wish would apply)?

There's no reason a language can't systematically mark these modes: "she who has applied" vs "she who ought to have applied"; "he who will be victimised" vs "he who would be victimised (if they happened to be the first person the killer saw)", and so on.
User avatar
Ahzoh
mongolian
mongolian
Posts: 4106
Joined: 20 Oct 2013 02:57
Location: Canada

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Ahzoh »

Salmoneus wrote: 03 Jun 2021 16:22 Why would it turn into a subjunctive when you already have a subjunctive?
The subjunctive will turn into the future and there will be no subjunctive form anymore. That leaves there only being indicative mood and imperative/jussive. I'm also iffy how much remaining modal force I would want the new future tense form to have.

I just need one of the forms to be used in subordinate clauses like "want that..." and "suggest that..." and one to be used bare in order to be used to issue commands. Ideally, it would be the same verb form.

If the old jussive became the new subjunctive, then the modal agent could also take on a subjunctive meaning.
You've just told us! For instance, "she who speaks", vs "she who must speak". Vs "she who I want to speak", "she who ought to speak", "she who might speak", etc.
I've never heard of such nouns and they don't feel right
However, be careful: the distinction you make is not primarily one of mood, but one of voice!
Well yes, the "to be Xed" examples are something of a mental typo that I never got to correct. I only have agent deverbals and no patient deverbals.
Bear in mind that one reason modal agents might not seem useful in English is that our agents are ambiguous - either indicative OR subjunctive.
Explains why the whole idea feels wrong to me.
Image Śād Warḫālali (Vrkhazhian) [ WIKI | CWS ]
Salmoneus
MVP
MVP
Posts: 2352
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus »

Ahzoh wrote: 03 Jun 2021 19:32
Salmoneus wrote: 03 Jun 2021 16:22 Why would it turn into a subjunctive when you already have a subjunctive?
The subjunctive will turn into the future and there will be no subjunctive form anymore.
Oh, OK. I thought you said you were going to form the future with a prefix.
Bear in mind that one reason modal agents might not seem useful in English is that our agents are ambiguous - either indicative OR subjunctive.
Explains why the whole idea feels wrong to me.
Sure, but sometimes a conlanger has to be willing to have their language do something differently from how English does it!
User avatar
Ahzoh
mongolian
mongolian
Posts: 4106
Joined: 20 Oct 2013 02:57
Location: Canada

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Ahzoh »

Salmoneus wrote: 03 Jun 2021 19:44
Ahzoh wrote: 03 Jun 2021 19:32
Salmoneus wrote: 03 Jun 2021 16:22 Why would it turn into a subjunctive when you already have a subjunctive?
The subjunctive will turn into the future and there will be no subjunctive form anymore.
Oh, OK. I thought you said you were going to form the future with a prefix.
Yea, I am going to form the future with a prefix on the base irrealis stem, but the prefix-less irrealis stem will either fall into disuse or merge with the realis stem because it is identical to the relais form.
Image Śād Warḫālali (Vrkhazhian) [ WIKI | CWS ]
User avatar
Omzinesý
runic
runic
Posts: 3079
Joined: 27 Aug 2010 08:17
Location: nowhere [naʊhɪɚ]

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Omzinesý »

This is not a precise question but anyways.

How should I start building complex aspect system? I could go a bit kitchen-sinky but don't know how.
I have usually just gone with three-way distinction: habitual, progressive, perfective.
I read the aspect section in Mithun's book on American languages and then checked how Ithkuil does it, but it's still quite messy.
Should I theoretically distinguish between aspect (internal time structure) and results (how the patient is affected)?
Apparently, aspects can be shared to two categories and they can be combined, like perfect being expressed be perfective+stative.
How do you make complex aspect systems?
My meta-thread: viewtopic.php?f=6&t=5760
Salmoneus
MVP
MVP
Posts: 2352
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus »

I'm not sure there's really any complicated answer to this. If you want your verbs to mark aspect, you can decide which aspects you want marked, and then decide how to mark them. There's not really a right or wrong answer.

For extra complexity and realism, of course, you could begin by asking what aspects were marked in an earlier form of the language, and then consider how that marking has evolved into the 'curren't form of the language. This could encompass either morphological development (agglutination, fusion) or semantic drift, or both.

You may want to consider how various sentences would be translated in this language. What distinctions would seem to be helpful to make?

Verbs have inherent aspect. How does this interact with morphological aspect - are all combinations possible, or does the significance shift?

You can certainly combine aspects - English does it, after all. ["I had been starting to eat"]. Can you combine all aspects in your language, or only some? Are some paradigms defective?
User avatar
Creyeditor
MVP
MVP
Posts: 4336
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor »

Salmoneus wrote: 07 Jun 2021 15:33 You may want to consider how various sentences would be translated in this language. What distinctions would seem to be helpful to make?
This! I think it might make sense to start with possible situations and how you could lassify them according to properties that are important for aspect. What properties are relevant for the aspect system in your conlang and what properties are not. What happens if two aspect markers fit a certain situation (e.g. progressive and imperfective aspect). Are both forms in free variation? Can you use only one form and why this form? Also, how do tense and aspect interact?

I once did some elicitations on aspect with a native speaker. We started out "knowing" some basics facts and having a specific question, but it turned out that you could not understand one part of the aspect system without understanding the other. Everything was much more interelated than we thought. We started out with some basic facts: the language had a three-way graded past system and a perfect marker. Our question was why the perfect marker was licit in resultative situation and recent past contexts but not in experiential contexts. It turned out that the medial past markers that we thought were in free variation could actually be distinguished in one context: experiential situations. Exactly the context that the perfect marker did not occur in. Even though experiental and non-experiental past markers did not block each other in all other contexts (and where in free variation), the experiental past marker was able to block the perfect marker.

So, I guess what I wanted to say is that aspect and tense markers are not really a simple list, but more like an interactional network.
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]
Salmoneus
MVP
MVP
Posts: 2352
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus »

Creyeditor wrote: 07 Jun 2021 20:59
Salmoneus wrote: 07 Jun 2021 15:33 You may want to consider how various sentences would be translated in this language. What distinctions would seem to be helpful to make?
This! I think it might make sense to start with possible situations and how you could lassify them according to properties that are important for aspect. What properties are relevant for the aspect system in your conlang and what properties are not. What happens if two aspect markers fit a certain situation (e.g. progressive and imperfective aspect). Are both forms in free variation? Can you use only one form and why this form? Also, how do tense and aspect interact?

I once did some elicitations on aspect with a native speaker. We started out "knowing" some basics facts and having a specific question, but it turned out that you could not understand one part of the aspect system without understanding the other. Everything was much more interelated than we thought. We started out with some basic facts: the language had a three-way graded past system and a perfect marker. Our question was why the perfect marker was licit in resultative situation and recent past contexts but not in experiential contexts. It turned out that the medial past markers that we thought were in free variation could actually be distinguished in one context: experiential situations. Exactly the context that the perfect marker did not occur in. Even though experiental and non-experiental past markers did not block each other in all other contexts (and where in free variation), the experiental past marker was able to block the perfect marker.
Bloody experientials. You're of course familiar with English, in which the experiential is unambiguously marked only for one single verb, and is otherwise merged with the perfect! An extreme example of how TAM can depend on verb class...

[I'd instinctively call both the experiential and the perfect 'tense' rather than 'aspect', though; an experiential is just a past indefinite...]
So, I guess what I wanted to say is that aspect and tense markers are not really a simple list, but more like an interactional network.
Absolutely. The whole of TAM is really a complex bag of many different concepts, and 'aspect' is what remains once the relatively clearcut bits of 'tense' and 'mood' have been substracted.

I would define 'aspect' in a 'narrow' sense as indicating the internal temporal structure of an event; but even this subsumes multiple dimensions:
- is the total temporal period specified (imperfective) or unspecified (perfective)?
- if specified, is it extended or punctual?
- if extended, is the extent bound?
- if extended, is the event continuous within its period or interruptive?
- if continuous within the period, is it functionally uniform within the period (a state), or evolving throughout the period (a progressive action)
- if interruptive, is it repetitive or evolving?


And then 'aspect' as commonly used includes not only aspect in this narrow sense, but also telicity, transitivity, definiteness, pluractionality, and what we might call 'external' aspect (does the event also extend beyond, or recurr beyond, the temporal period of the verb?)...


But having said all that, aspect systems don't have to be complicated - you could just have a simple distinction, perfective vs imperfective, for instance. Just be aware that lexical aspect (and tense, mood) often interfere even with simple aspect systems (eg, can future events be perfective? can punctual events be imperfective? and so on...)
User avatar
Thrice Xandvii
runic
runic
Posts: 2700
Joined: 25 Nov 2012 10:13
Location: Carnassus

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Thrice Xandvii »

I don't know how "quick" either of these questions are, but:

1) Is there any naturalistic way that a language could have a very simple and pretty symmetric inventory with unvoiced stops... and then have /g/ with no /k/? It seems super odd to me, but I'm not sure just how odd it might be.

2) How might one combine Welsh's and Gaelic's phonological inventories together? I'm looking for a suggestion here that also doesn't just mash all the sounds together into one gigantic mess of an inventory?




(Maybe this'd be better in L&N or somewhere? Not entirely sure.4)
Image
Salmoneus
MVP
MVP
Posts: 2352
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus »

Thrice Xandvii wrote: 10 Jun 2021 12:05 I don't know how "quick" either of these questions are, but:

1) Is there any naturalistic way that a language could have a very simple and pretty symmetric inventory with unvoiced stops... and then have /g/ with no /k/? It seems super odd to me, but I'm not sure just how odd it might be.
It's very odd. Voicing is harder to accomplish and distinguish the further back you go in the mouth, so /g/ is by far the most likely voiced stop to be missing (other than the voiced uvular stop, which, being further back, is even rarer).

But it certainly could happen naturalistically. The two obvious reasons would seem to be:

- a voiced set has arisen, but /k/ has already shifted. So, for instance, if we imagine /k/ > /?/, and then /w r j/ > /b d g/ (perhaps only initially/finally?). The initial shift in place only applies to stops, and since at that point there isn't a /g/ yet, there's nothing to shift.

- a full voiceless set has shifted to voiced, and a new voiceless set has been created that lacks /k/. So, /p t k/ > /b d g/; you could say, for instance, that initially these stops were aspirated (later becoming fricatives), while the unaspirated intervocalic stops lenited through voicing. Then you replenish the voiceless stops from, say, ejective or pharyngealised stops. But it's possible for ejective or pharyngealised /k/ to be absent - perhaps it's already lost this quality (and so merges into the earlier voiceless stops), or perhaps it is backed to a glottal stop, or a pharyngeal or the like. So the new voiceless set lacks /k/.

I don't imagine this would be a very stable system - I'd expect /g/ to simply devoice to /k/! - but it could conceivably arise, yes.
2) How might one combine Welsh's and Gaelic's phonological inventories together? I'm looking for a suggestion here that also doesn't just mash all the sounds together into one gigantic mess of an inventory?
What do you mean, 'combine'? Combining the inventories would indeed result in a single, larger inventory. You could of course just take some sounds from one inventory and put them in the other - but that kind of depends which ones you want in your language, doesn't it?

In this case, in particular, Welsh and Irish have similar inventories, except that Irish has a systematic slender/broad distinction that Welsh lacks. Welsh also has voiceless nasals (through mutation), a lateral fricative, and its dorsal fricative is uvular, whereas in Irish it's velar. Oh, and it retains its interdentals, which have been lost in Irish (though still written). And Irish has its weird palatal voiced fricative/stop/approximant/thing. But Welsh has central vowels.

So to 'combine' the inventories, take some of the features of one and put them in the other, or vice versa?
User avatar
Pabappa
greek
greek
Posts: 490
Joined: 18 Nov 2017 02:41
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Pabappa »

Thrice Xandvii wrote: 10 Jun 2021 12:05
1) Is there any naturalistic way that a language could have a very simple and pretty symmetric inventory with unvoiced stops... and then have /g/ with no /k/? It seems super odd to me, but I'm not sure just how odd it might be.
Mongolian does it:

https://en.wikipedia.org/wiki/Mongolian ... Consonants

No two ways about it .... the /k/ is just not there.

The distinction in the alveolar region is aspiration rather than voice, but Im sure that wont be a problem since they could easily shift from /t th/ to /d t/ without disturbing the velars..
I'll take the theses, and you can have the thoses.
User avatar
Creyeditor
MVP
MVP
Posts: 4336
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor »

Thrice Xandvii wrote: 10 Jun 2021 12:05 I don't know how "quick" either of these questions are, but:

1) Is there any naturalistic way that a language could have a very simple and pretty symmetric inventory with unvoiced stops... and then have /g/ with no /k/? It seems super odd to me, but I'm not sure just how odd it might be.
It is odd and probably diachronically unstable. A quick fix is to have an aspiration distinction instead of a voicing distinction.

Moni (Trans-New-Guinea, Indonesia) did something similar. /k/ lenited becoming /h/ and deleting eventually. But /g/ eventually devoiced word initially to [k] leaving [g] to only occur intervocally.
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]
User avatar
Thrice Xandvii
runic
runic
Posts: 2700
Joined: 25 Nov 2012 10:13
Location: Carnassus

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Thrice Xandvii »

Thanks all for the food for thought with regard to /g/ lacking /k/ inventories! Mongolian does have an interesting sound distribution and it is intriguing that /k/ only appears there in loanwords it would seem. So, it sounds to me like a language can do it, but it would be odd and quite uncommon and not likely to be a super stable state to exist in. Which, that's fine by me, I don't really need to do anything diachronically with the 'lang I'm thinking of. It would serve more as a vehicle for the weirdo script I am trying to make happen.
Salmoneus wrote: 10 Jun 2021 13:48 What do you mean, 'combine'? Combining the inventories would indeed result in a single, larger inventory. You could of course just take some sounds from one inventory and put them in the other - but that kind of depends which ones you want in your language, doesn't it?

In this case, in particular, Welsh and Irish have similar inventories, except that Irish has a systematic slender/broad distinction that Welsh lacks. Welsh also has voiceless nasals (through mutation), a lateral fricative, and its dorsal fricative is uvular, whereas in Irish it's velar. Oh, and it retains its interdentals, which have been lost in Irish (though still written). And Irish has its weird palatal voiced fricative/stop/approximant/thing. But Welsh has central vowels.

So to 'combine' the inventories, take some of the features of one and put them in the other, or vice versa?
I think what I mean is how would one take a bit of the "defining features" of both 'langs in such a way that a bit of the "flavor" of both shines through without it becoming a jumbled mess or just smashing them both together. Obviously one of the defining features would be the slender/broad distinction, I would think, but was just wanting to solicit some thoughts to see how others might do it or approach such a task.
Image
User avatar
Creyeditor
MVP
MVP
Posts: 4336
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor »

Maybe the answer is /ɬˠ/ vs. /ɬʲ/
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]
Salmoneus
MVP
MVP
Posts: 2352
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus »

Thrice Xandvii wrote: 11 Jun 2021 03:38 I think what I mean is how would one take a bit of the "defining features" of both 'langs in such a way that a bit of the "flavor" of both shines through without it becoming a jumbled mess or just smashing them both together. Obviously one of the defining features would be the slender/broad distinction, I would think, but was just wanting to solicit some thoughts to see how others might do it or approach such a task.
In that case, the inventories may be the wrong place to look.

In terms of inventories, I would guess that to 'Welshify' Irish, it would help to have interdental fricatives, the lateral fricative, and a devoicing mutation. And maybe central vowels.

But I think the bigger problem is the general feel of the two languages. Welsh is a lot... softer? (muddier, stickier), whereas Irish is relatively 'hard' (sharp, clear). Having those extra fricatives would help a lot, to be sure, but I'd guess you'd probably also want to have more voiced consonant, for a start? And more front and mid vowels, and umlaut? And diphthongs?


For comparison, here's the UDHR in Welsh:
Genir pawb yn rhydd ac yn gydradd â'i gilydd mewn urddas a hawliau. Fe'u cynysgaeddir â rheswm a chydwybod, a dylai pawb ymddwyn y naill at y llall mewn ysbryd cymodlon

And in Irish:
Saolaítear na daoine uile saor agus comhionann ina ndínit agus ina gcearta. Tá bua an réasúin agus an choinsiasa acu agus dlíd iad féin d'iompar de mheon bráithreachais i leith a chéile.

I've bolded what are usually considered the hardest or harshest sounds: voiceless stops, voiceless sibilants, and voiceless dorsal fricatives (just by sight, I don't speak either language). As you can see, these are a lot more common in Irish than in Welsh!
Post Reply