Split from Scripts for Wesnoth

For writers working on documentation, story prose, announcements, and all kinds of Wesnoth text.

Moderator: Forum Moderators

Whiskeyjack
Posts: 476
Joined: February 7th, 2015, 1:27 am
Location: Germany

Re: Split from Scripts for Wesnoth

Post by Whiskeyjack »

Durin, that looks like an awesome concept! Knowledge could have tool as its essence (not optimal, but I think tool should make its way in there. An alternative option would be to have craft as yet another part of the inner circle).

The only thing I really don´t like is your "interaction". How do the dwarves know about diamond synthesis and the like (nonetheless at the languages inception)? Shouldn´t language come before cultural elements like money (the buy connection)? How is stone rusted metal or rusted metal stone (not a chemist, but that is new to me)? Sheltering fuel sounds nice, but mining doesn´t really fit into the equation for me.
Might be a good fit to connect with the above idea of tool/craft (though that is once again at odds with the time factor of language and invention/culture). One might hold against my time argument the invention of written language at a later point, but that does not apply to the diamond thing.

On that note: craft might be an interweaving element that is not of the four but instead adds an aspect to every interaction (basically culture meeting raw elements):
  • the mining of stone-sheltered fuel
  • the forming of metal through fuel (switching places of gem and metal)
  • the ornamentation of metal by gem (this one is weak, couldn´t think of something better on the fly)
  • the detection of gem hiding in stone
That would add another four runes. The problem that remains is that sheltering would fit for every element in connection to stone and that both stone interweavings depend on the direct connection while those of metal do not.

Just throwing out some ideas this gave me... feel free to discuss/discard/modify them.
Under blood-red skies, an old man sits
In the ruins of Carthage - contemplating prophecy.
Durin_the_great
Posts: 50
Joined: November 6th, 2015, 6:39 pm

Re: Split from Scripts for Wesnoth

Post by Durin_the_great »

Whiskeyjack,

Even as I was typing my last message, I started thinking about the outer loop. It would be much more sensible to make it clockwise:

Fuel hides in stone;
Stone (ore) is purified to metal;
Metal is used to cut gems or earn gems;
Gems can be used to buy fuel;

It would make the interactions a bit more understandable in medieval universe.
Craft could be inserted among the priorities (between wealth and life) alongside with tools (among gems and fuel). Then it would be:

Fuel hides in stone;
Stone is purified to metal;
Metal earns/cuts gems;
Gems buy tools;
Tools extract fuel (mine coal, cut trees);

Life ends with death;
Death gives honour;
Honour brings wealth;
Wealth encourrages craft;
Craft supports life;

It goes quite smooth this way. Oh, and then the "tools are used in craft" relation appears. We get 2 more runes, just the runes and their letters (sounds) would be needed.
Caladbolg
Posts: 198
Joined: January 1st, 2016, 4:40 pm
Location: Hopelessly trapped within the Submachine

Re: Split from Scripts for Wesnoth

Post by Caladbolg »

Wow, you've really thought this through! It's great :)
Durin_the_great wrote:I wanted to stay less complicated, so I limited the number of runes in the set. This might prove ill, since this runeset is not enough to write all dwarvish Wesnothian names....These runes represent consonants, the main part of dwarvish language....The vowels might then be represented as diacritic marks.
I agree that keeping things simple is a good idea, but isn't just 9 symbols a bit too few? Also, it's a bit strange that you're making it into an abugida of sorts considering runes were mostly used in alphabets. I like that though :)
I used four vowels: a, i, u and h. Yes, h would be a vowel that comes after a consonant, eg -kh-, -gh-, -dh-.
How would that h sound as a vowel? Like a breathy sound behind the consonant?
Durin_the_great
Posts: 50
Joined: November 6th, 2015, 6:39 pm

Re: Split from Scripts for Wesnoth

Post by Durin_the_great »

Yeah, I know that 9 are few. But the funny thing is that if we count h as a vowel, Wesnothian dwarvish names only use 12 consonants, but a couple are only seen several times.

Yes, that's what I meant about "h". It kind of suits the style.

The system of vowels might need consideration, since we also lack e and o which are seen in a few names or titles.
User avatar
Celtic_Minstrel
Developer
Posts: 2166
Joined: August 3rd, 2012, 11:26 pm
Location: Canada
Contact:

Re: Split from Scripts for Wesnoth

Post by Celtic_Minstrel »

Those runes look fairly nice, though I feel like Dwarvish needs a few more than just those. Perhaps that could be the oldest form of the alphabet, and over time dwarves have added to it,

What you're describing as a radical system sounds rather like an abugida (or abjad if the vowel markers are optional). A radical system is something used by Chinese, and the radicals don't represent a specific vowel IIRC; I don't think it's a good fit for dwarves in any case. I see no reason why an abugida could not be used by dwarves, though, but I do object to describing H as a vowel. You could rather consider <kh>, <gh>, <dh>, <th> as separate, unique consonants; <h> alone is also a consonant (though I dunno whether it appears in Dwarvish). Mind you, it's still reasonable to have a modifier in a Dwarvish rune set that transforms <k> -> <kh>, <g> -> <gh>, <d> -> <dh>,<t> -> <th>. You could even add to that list: <p> -> <ph> (ie <f>) and <b> -> <bh> (ie <v>) are reasonable extensions of the system.

With this in mind, and considering your mention of using / for <h>, the set could be extended to something like the following:
Possible variation of Durin_the_Great's runes.
Possible variation of Durin_the_Great's runes.
I changed <b> to <p> and <z> to <s> based on the occurrence of these consonants in the existing Dwarvish name base. I've also used <c> for the /k/ sound for the same reason. Some of these are purely hypothetical and could easily be switched out or just omitted. It's still missing <t> and <th>, though. The right column shows how vowel markers might work, including the possibility of combining two vowel markers to produce diphthongs (rather than using an extra X symbol as you described before).

It was only until after I made that image that I noticed the second page of this thread, where you mention thinking of <h> not as a vowel or fricatizer but instead as a breathy-voice modifier. There's nothing wrong with that! You could still use some of the above altered rune forms for breathy-voiced versions, if you wanted. I'd still point out that your system is missing /t/ though and should probably have /p/ and /s/ rather than /b/ and /z/ if you want the existing Dwarvish names to fit in. You could treat <f> as a breathy-voiced /p/, though I'll note that none of the names in the Dwarvish list actually contain <f>, so you can just ignore it if you want. You're also missing <ng> and <o>. So basically, you need at least two more runes for full coverage, and one more vowel marker.

EDIT: Oh, I just realized the dwarf names do use <e>. It doesn't show up until about halfway through them, so I missed it. So I guess you'd need a vowel marker for <e> too.

EDIT2: Another thing I just thought of, based on names in the list such as "Aigdrlos" - perhaps dwarves treat /r/ more as a vowel than a consonant. If it's the English /r/, it's already an approximant, which is a consonant that's vowel-like; even the Spanish R or French R could be treated in some ways like a vowel though, since they are trills. Under this proposal, the name "Aigdrlos" would be 3 syllables and would probably sound to an English speaker something like "Ayg-der-lohs".
Author of The Black Cross of Aleron campaign and Default++ era.
Former maintainer of Steelhive.
Durin_the_great
Posts: 50
Joined: November 6th, 2015, 6:39 pm

Re: Split from Scripts for Wesnoth

Post by Durin_the_great »

Yeah, these points need consideration.

1) The "h" case: what if we make "h" a consonant and an operator? It wod have a rune and a mark. Whenever it is used after a vowel or a vowel complex, the rune is used and it goes like a consonant. If it comes after a consonant, it is used as an operator that modifies the sound of the consonant. Then the mark is used. I suggest an underscore, since it would not interact with other marks.

2) The "r" case: I think same solution as "h" would do. By the way, have you noticed that in some languages r sounds nearly as h? Then the modifiercould be the common one, the r-h operator (underscore _).

3) As for the assigned sounds: I actualy don't remember why and how I assigned the eunes particular sounds. It basically has no effect, the sounds can be easily edited, the cultural background is something that I would try to maintain.

4) The lack of runes: as Whiskeyjack has suggested, we can add tools and craft. It preserves the relations of runes and adds sounds.

5) The vowel case: it's even more flexible here. There is no rule of vowel relations, so we can have as many marks as needed. It's a good idea to have seperate symbols for vowel complexes. Then we have "a", "e", "i", "o", "u", "au", "ai" and operator r-h.
Caladbolg
Posts: 198
Joined: January 1st, 2016, 4:40 pm
Location: Hopelessly trapped within the Submachine

Re: Split from Scripts for Wesnoth

Post by Caladbolg »

A few thoughts on the matter:

1) Yeah, it'd be good to have a modifier that transforms /d/ into /ð/ (dh), /k/ into /x/ (kh), etc. These were a part of Norse phonology (says wiki) so it'd fit thematically. That could be an operator adding a breathy 'h', but 'h' could also be a standalone consonant.

2) (Disclaimer: personal preference ahead) For 'lh', I'm not sure if /ɬ/ is a good idea. It doesn't sound "dwarvish" to me. For 'sh', I'd prefer /ʃ/ over /ɕ/, but maybe /ʂ/ would be even better. In any case, I don't think it matters that /m/ doesn't have a pair. After all, it is not necessary that this h-modifier be aplicable to all consonants.

3) I'm not sure it'd be good for 'r' and 'h' to be the same modifier. It'd sound too French, I think. Voiced alveolar trill (/r/) would probably be the best for 'r' as it was used in Old Norse and can easily function as a vowel (as is the case in some Slavic languages today). It also has that rugged sound that'd fit the dwarves. 'r' would also need a standalone rune if it can appear as a consonant.(edit: oops, 'r' rune is already there)

4) I'm a bit divided on the 'ng' issue. It could be /ŋ/ as in english "sing", but having it as two consonants /n/+/g/ would sound a bit rougher (depending on how you want your dwarvish to sound).
Durin_the_great
Posts: 50
Joined: November 6th, 2015, 6:39 pm

Re: Split from Scripts for Wesnoth

Post by Durin_the_great »

2. I totally agree. The h operator would only affect some consonants, e.i. it would be used with particular consonants. Standalone h consonant is needed. Also, I think ng sounds really elvish, reminds of sindarin.

3. Sure.

4. I suggest making it rough, but subtle rough. Like a music that not all cultures understand as smooth.

Another idea: having a (symbol) operator that makes consonants more ringing could help. It would transform k to g, p to b, s to z and t to d. Then there would be two operators and 9 runic consonants.

Summary. Right now we have:
12 runic consonants (or 9):
c(/k/), g, l, m, n, d, r, h, t and s, but b is also needed for Burin (or c, l, m, n, r, p, h, t, s).

One operator (or two):
The breathy
(And maybe the soundy)

5 vowels:
a, e, i, o, u

And 3 vowel complexes:
ai, au, oi (for Glonoin).
User avatar
Celtic_Minstrel
Developer
Posts: 2166
Joined: August 3rd, 2012, 11:26 pm
Location: Canada
Contact:

Re: Split from Scripts for Wesnoth

Post by Celtic_Minstrel »

Durin_the_great wrote:1) The "h" case: what if we make "h" a consonant and an operator? It wod have a rune and a mark. Whenever it is used after a vowel or a vowel complex, the rune is used and it goes like a consonant. If it comes after a consonant, it is used as an operator that modifies the sound of the consonant. Then the mark is used.
I'm not too sure what you mean here...
Durin_the_great wrote:I suggest an underscore, since it would not interact with other marks.
So basically a line beneath the rune?
Durin_the_great wrote:2) The "r" case: I think same solution as "h" would do. By the way, have you noticed that in some languages r sounds nearly as h? Then the modifiercould be the common one, the r-h operator (underscore _).
I've noticed no such thing, but there's no particular reason I know of to prevent r and h being variations of the same phoneme in Irdyan Dwarvish. There doesn't seem to be much use of h as a standalone constant in the official namelist though. (I didn't see any at all.)
Caladbolg wrote:1) Yeah, it'd be good to have a modifier that transforms /d/ into /ð/ (dh), /k/ into /x/ (kh), etc. These were a part of Norse phonology (says wiki) so it'd fit thematically. That could be an operator adding a breathy 'h', but 'h' could also be a standalone consonant.
I also agree that this fits them thematically.
Caladbolg wrote:2) (Disclaimer: personal preference ahead) For 'lh', I'm not sure if /ɬ/ is a good idea. It doesn't sound "dwarvish" to me. For 'sh', I'd prefer /ʃ/ over /ɕ/, but maybe /ʂ/ would be even better. In any case, I don't think it matters that /m/ doesn't have a pair. After all, it is not necessary that this h-modifier be aplicable to all consonants.
I literally only used /ɬ/ because it was the only "fricatization of L" that I could think of. Of course it doesn't have to be in the Dwarvish language (the L rune may have no h-modified form). As for the palatal sibilant, my only reason for omitting the retroflex is that I didn't want to show all three; I'd be fine with it.
Caladbolg wrote:4) I'm a bit divided on the 'ng' issue. It could be /ŋ/ as in english "sing", but having it as two consonants /n/+/g/ would sound a bit rougher (depending on how you want your dwarvish to sound).
I think it'd be quite weird for it to be [ng], but there's also the possibility of <ng> representing [ŋg]. In that case there doesn't really need to be a phonemic /ŋ/, and the sound would only ever occur before <g> or <k>; similarly, the regular [n] sound would never occur there. If it's not a separate phoneme then it doesn't really need a separate rune. If <ng> doesn't get a rune and <r> is demoted to a modifier or merged into the <l> rune, then the wealth rune can be used for <t> and no new runes are needed.

Building on Durin's concept of <r> having h-like qualities, the life rune plus H-modifier could produce <r>. Admittedly, h-like qualities don't really work for <r> when it's a trill, though; it might actually better to swap it, using the plain life rune for <r> and the h-modified one for <l>.

Durin_the_great wrote:Another idea: having a (symbol) operator that makes consonants more ringing could help. It would transform k to g, p to b, s to z and t to d. Then there would be two operators and 9 runic consonants.
Well, first of all, we don't actually need <b> as far as I know. There is no <b> whatsoever in the official Dwarvish names in 1.12. We don't really need a standalone <h> either. The only consonant letters used in the official dwarvish names (unless I missed one) are <g t th s l ng c d r n m p>. So, there's a c/g pair and a t/d pair, but for labials there's only p. Given this, I don't feel that a voicing operator is required. In fact, the fricatization operator demonstrated in my previous post isn't really needed either, as the only fricatives are s and th. (It's unknown whether these are soft or hard th though, so that could be three fricatives total.) Of course we don't need to limit the inventory to only what's in the official names, though. I think the /x/ (ch/kh) and /ɣ/ (gh) are harsh and may be quite suitable for dwarves, and if we're adding those then (IMO) we might as well also add /f/ (ph) and maybe an sh.
Durin_the_great wrote: Summary. Right now we have:
12 runic consonants (or 9):
c(/k/), g, l, m, n, d, r, h, t and s, but b is also needed for Burin (or c, l, m, n, r, p, h, t, s).
I'm only counting ten consonants there, excluding /b/. Where is Burin? It's not in the official name list, but if it's the name of a mainline campaign character than we should account for it.
Durin_the_great wrote: And 3 vowel complexes:
ai, au, oi (for Glonoin).
Where is Glonoin? I don't see that in the name list either. By the way, these are called "diphthongs". :)
Author of The Black Cross of Aleron campaign and Default++ era.
Former maintainer of Steelhive.
Caladbolg
Posts: 198
Joined: January 1st, 2016, 4:40 pm
Location: Hopelessly trapped within the Submachine

Re: Split from Scripts for Wesnoth

Post by Caladbolg »

Celtic_Minstrel wrote:I think it'd be quite weird for it to be [ng], but there's also the possibility of <ng> representing [ŋg]. In that case there doesn't really need to be a phonemic /ŋ/, and the sound would only ever occur before <g> or <k>; similarly, the regular [n] sound would never occur there. If it's not a separate phoneme then it doesn't really need a separate rune.
Sounds good.
Building on Durin's concept of <r> having h-like qualities, the life rune plus H-modifier could produce <r>. Admittedly, h-like qualities don't really work for <r> when it's a trill, though; it might actually better to swap it, using the plain life rune for <r> and the h-modified one for <l>.
Hmm... r+h and l+h combinations don't appear in the dwarvish name list. Modifying <r> into <l> or vice versa could work if <r> were a tap (/ɺ/) but that has nothing to do with breathiness so it wouldn't fit this modification system. Maybe it's better to just look at <r> and <l> as seperate runes, neither of which can be modified by 'h'.
Well, first of all, we don't actually need <b> as far as I know.
It's not on the list but Burin is a mainline character (in 'The Rise of Wesnoth', I believe).
We don't really need a standalone <h> either. The only consonant letters used in the official dwarvish names (unless I missed one) are <g t th s l ng c d r n m p>.
True.
I don't feel that a voicing operator is required. In fact, the fricatization operator demonstrated in my previous post isn't really needed either, as the only fricatives are s and th. (It's unknown whether these are soft or hard th though, so that could be three fricatives total.) Of course we don't need to limit the inventory to only what's in the official names, though. I think the /x/ (ch/kh) and /ɣ/ (gh) are harsh and may be quite suitable for dwarves, and if we're adding those then (IMO) we might as well also add /f/ (ph) and maybe an sh.
I think having a more complete system makes it seem more realistic and internally consistent so I support adding extra sounds even if they don't appear on the list (maybe the list can be revised if this takes off).
User avatar
Celtic_Minstrel
Developer
Posts: 2166
Joined: August 3rd, 2012, 11:26 pm
Location: Canada
Contact:

Re: Split from Scripts for Wesnoth

Post by Celtic_Minstrel »

Caladbolg wrote:Hmm... r+h and l+h combinations don't appear in the dwarvish name list. Modifying <r> into <l> or vice versa could work if <r> were a tap (/ɺ/) but that has nothing to do with breathiness so it wouldn't fit this modification system. Maybe it's better to just look at <r> and <l> as seperate runes, neither of which can be modified by 'h'.
It's not necessarily required for the h-modifier to be 100% consistent. Maybe Irdya dwarves realized that R / L can't have an h-modifier according to their language's phonology, but since the sounds are closely related (especially so if the R is trilled!), decided to use a rune modified by the h-modifier to represent one of the two sounds, and the unmodified rune for the other.
Caladbolg wrote:I think having a more complete system makes it seem more realistic and internally consistent so I support adding extra sounds even if they don't appear on the list
Sure, I'm on board for keeping all those fricatives! I'll also add that /x/ is a bit like /h/ and may fit a Nordic/Germanic flavour (since the former evolved into the latter in some Germanic languages, including English). I guess we do also need B, after all.
Caladbolg wrote:(maybe the list can be revised if this takes off).
The old list may never be updated since it's pretty much been obsoleted by Dugi's rule-based generation, but said rules could definitely be updated. Actually, they're currently missing <p> rather than <b>. (More accurately, there's a single instance of <p>, but several of <b>.)
Author of The Black Cross of Aleron campaign and Default++ era.
Former maintainer of Steelhive.
Durin_the_great
Posts: 50
Joined: November 6th, 2015, 6:39 pm

Re: Split from Scripts for Wesnoth

Post by Durin_the_great »

I have an idea for x. For the style, we actualy could use /x/. I mean the breathy russian х. It could be a standalone h affected by the modifier. A voiced h made into a breathy h.

Also, the name Kuhnar suggests that standalone h is needed (SoF if not mistaken).

Totally agree with your ideas about the h modifier.

Tomorrow (12-03) all my devices will be taken from me till 12-~11, so I will have no connection. Sorry, an olympiad.
Caladbolg
Posts: 198
Joined: January 1st, 2016, 4:40 pm
Location: Hopelessly trapped within the Submachine

Re: Split from Scripts for Wesnoth

Post by Caladbolg »

Celtic_Minstrel wrote:It's not necessarily required for the h-modifier to be 100% consistent. Maybe Irdya dwarves realized that R / L can't have an h-modifier according to their language's phonology, but since the sounds are closely related (especially so if the R is trilled!), decided to use a rune modified by the h-modifier to represent one of the two sounds, and the unmodified rune for the other.
So, basically, the modifier would change the sound represented by the rune into a "similar" sound (not necessarilly h-modification)? Yeah, that'd be simple yet pragmatic. I like it.
The old list may never be updated since it's pretty much been obsoleted by Dugi's rule-based generation, but said rules could definitely be updated.
I was not aware of that. So, if I'm getting this right, this would mean that we can allow ourselves more freedom when making runes, use a bottom-up approach and then names consistent with the language we make can easily be generated by tweaking a few rules? If so, that's awesome :D
User avatar
WTrawi
Posts: 136
Joined: July 11th, 2016, 6:04 pm
Location: My location is not public, but I know YOURS and I can use it against you.

Re: Split from Scripts for Wesnoth

Post by WTrawi »

Wow, you wrote a lot of stuff while I was not there. :D
Not very active on the forum anymore, but I still read it, and I also still play Wesnoth and draw a bit.
User avatar
Celtic_Minstrel
Developer
Posts: 2166
Joined: August 3rd, 2012, 11:26 pm
Location: Canada
Contact:

Re: Split from Scripts for Wesnoth

Post by Celtic_Minstrel »

Durin_the_great wrote:Also, the name Kuhnar suggests that standalone h is needed (SoF if not mistaken).
It suggests nothing of the kind. There are at least two alternative interpretations. The h in that name could be merely a modifier on the vowel (perhaps making it longer?). Or, it could be representing the sound /x/.

Incidentally, English takes the first of those two interpretations (at least in the dialects I'm familiar with - an <h> sound cannot be the coda of a syllable, so the <h> in the name <Kuhnar> would be assumed to be part of the vowel; in the case of English it'd probably laxify the vowel, ie transform it from /u/ to /ʊ/. Note that tense-lax distinction is entirely arbitrary as far as I know and would depend on the language, which is why I suggested lengthening instead.
Caladbolg wrote:I was not aware of that. So, if I'm getting this right, this would mean that we can allow ourselves more freedom when making runes, use a bottom-up approach and then names consistent with the language we make can easily be generated by tweaking a few rules? If so, that's awesome :D
Uhh. Potentially? I'd personally prefer to (at least mostly) preserve the flavour of the old names, though. The current ruleset is pretty simplistic, but in theory the rules could probably be made more complex, with power close to this word generator... though I'm not entirely sure if there'd be performance hits for more complex rulesets. To be precise, the rules for the new name generator are specified as a context-free grammar.

As an aside, it'd be cool to add rules for female dwarf names, too. Sure, there are no female dwarves in mainline (I think? I haven't played any of the harder campaigns though), but I presume they exist in the setting.
Author of The Black Cross of Aleron campaign and Default++ era.
Former maintainer of Steelhive.
Post Reply