On making a conscript that doesn't suck.

If you're new to these arts, this is the place to ask "stupid" questions and get directions!
clawgrip
MVP
MVP
Posts: 2234
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: On making a conscript that doesn't suck.

Post by clawgrip »

Sights wrote:Don't mean to take attention away from Chagen's budding script, but I have a noob question concerning what Clawgrip said...
clawgrip wrote: Part 3
Every time I've seen someone design a conscript and show the characters from a logographic script that it supposedly descend from, the characters shown are invariably basic pictographs or ideographs. This simply isn't how it works. If a society has an entire functional logographic script from which to choose source characters for a syllabary, why would they seek out only basic pictographs or ideographs for this?
I thought that's exactly how a number of characters from the latin alphabet originated. Doesn't the current "A" shape trace its origins to a character resembling an ox head in Phoenician or Proto-sinaitic? My understanding (or total lack thereof) was that certain ideographs eventually became the phonetic characters for the sounds which made up the corresponding word in the first place (most often the first sound of said word, ASFAIK). What I guess Clawgrip is pointing at is that... it would be strange that only certain ideographs were "picked" instead of others, particularly when a complex logographic system was already in use? It sounds unlikely now that I think about it, but I thought the reason for this was that the ideographs concerned represented very basic dimensions of human life: things like "water", "sun", "house" and other words likely to come up in Swadesh lists and things like that. Am I mistaken?

I ask because I'm developing a script too (an alphabet), and I thought that this logic was sound enough [:|]
Let me clarify. I was definitely not saying that it is impossible for phonetic glyphs to originate from basic pictographs (as indeed, many Latin letters do), but that it's unlikely there won't be several non-basic ones <E> is supposed to come from the sign for Egyptian "jubilation", and <S> from "hunt".

Egyptian very early on developed phonetic writing based heavily on pictographs, and avoided compounding glyphs and this influenced a lot of scripts in the area, which is why a lot of Latin letters can be traced back to basic pictograms (although Cuneiform did take advantage of compounding). Chinese is quite different, because it didn't really develop a phonetic script and was heavily into compounding, so when hiragana was created, it borrowed from complex characters.

Let's try an example. Imagine, if you will, that English were written in a logographic script, and that a syllabary was slowly beginning to develop from it. If we think, what word would likely get used for the syllable /biː/, it would likely be "bee". You could use a pictograph of a bee, or maybe not. This is 蜂 in Chinese/Japanese, a compound character that does not contain a pictogram of a bee (but it has one of a snake!). If we were going to choose one for /beɪ/ we would probably choose the character for "bay". It could be a pictogram of a bay, but again, This is 灣/湾 in Chinese/Japanese, which also does not contain a pitogram of a bay (just one for water). If we chose a glyph for /teɪ/ we might go with "tame" (馴 in Chinese), "take" (取 in Chinese), etc.

Also, we should remember that the Swadesh list is about basic words and has nothing to do with writing. My main point is that already literate cultures are not necessarily going to choose from their most basic words when designing a phonetic writing system, they're probably going to choose whatever seems easiest to remember (like the glyph for "owe" in our hypothetical English logography is a natural choice for the syllable /oʊ/, even though the "owe" glyph would likely be complicated due to the difficulty of representing that concept visually), or whatever has most cultural significance, or whatever looks nicest.
I ask because I'm developing a script too (an alphabet), and I thought that this logic was sound enough [:|]
I may have overstated somewhat. If we look at the basic phonetic pictographs for Egyptian, we see that a lot of them are relatively simple, not Swadesh-level basic, but still not compounded or anything.

Egyptian single consonant characters give us: vulture, reed leaf, arm, quail chick, leg, stool, horned viper, owl, water, mouth, reed shelter, wick, and so on.

Hiragana characters are a little more random and complicated, giving us: easy/safe, more than, eaves, clothing, at/in/on, add, how much, long time ago, plan/measure, self/I/you, left, of/this, unit of measurement, world, before/formerly, fat, know, river, heaven, stop, etc.
Last edited by clawgrip on 18 Apr 2014 03:44, edited 1 time in total.

User avatar
Thrice Xandvii
runic
runic
Posts: 2696
Joined: 25 Nov 2012 10:13
Location: Carnassus

Re: On making a conscript that doesn't suck.

Post by Thrice Xandvii »

Based on my somewhat limited knowledge, I would say that in effect, you are both right. Clawgrip (I think) is merely pointing out that to have EVERY glyph in an alphabet/abjad descend from VERY simple pictorgraphs in a (presumably) well fleshed out and extensive logographic system would be quite bizzarre. Take a look at a list of frequently used Chinese characters, for instance, while you will see some fairly simple ones like 人 (man, person) topping out that list, you will also see some far more complex and non-representational ones like 謝 (to thank) in there too. One would have to almost create an entire logographic system to be able to believably use it to derive an alphabet.

On the other hand, you are also correct. You CAN trace back some of the Latin script to things like an ox, but I don't think it is safe to say that ALL of the components of the modern Latin alphabet can be traced to the simplest representational items either. "H" for instance, may owe it's shape to the pictogram for the word "courtyard" which is common enough, but not necessarily the most simple of ideas either.
Edit: Looks like Clawgrip said everything I was going to say better... [:)]
Image

User avatar
Sights
sinic
sinic
Posts: 211
Joined: 04 Jan 2014 20:47

Re: On making a conscript that doesn't suck.

Post by Sights »

I see then. So I guess I'll have to jumble up my script and include irregularities, like characters without any representantional origin, ones with non-basic meanings, etc. Thanks to the both of you [:)]

User avatar
DesEsseintes
mongolian
mongolian
Posts: 4170
Joined: 31 Mar 2013 13:16

Re: On making a conscript that doesn't suck.

Post by DesEsseintes »

@Clawgrip: you are very good at these guides! [:D] Thanks for taking the time!

User avatar
eldin raigmore
korean
korean
Posts: 5605
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: On making a conscript that doesn't suck.

Post by eldin raigmore »

clawgrip wrote:....
- The roundness of Oriya, Telugu and Kannada scripts is a direct result of scribes being careful not to damage the palm leaves they were writing on, and is a clear contrast to the rather angular Brahmi script, from which they are descended. Brahmi's angularity, on the other hand, likely arises from being incised into rock.
- Futhark is another obvious example: as it was mostly incised into wood or stone or what have you, it ends up being very angular, like Brahmi. It also lacks horizontal lines, probably to avoid having strokes run along the wood grain.
….
I was going to add "the nature of the writing surface matters too, as much as the nature of the writing implement"; but I see you already included the examples I was thinking of.
So I'll just sort of high-light them.

btw Doesn't carving the letters on stone with hammer and chisel kind of push writers to write right-to-left? (whether top-to-bottom within right-to-left, or right-to-left within top-to-bottom).

And doesn't writing with anything that's still smearable for a bit after it's applied -- e.g. ink -- kind of push writers to write left-to-right, and push them even more strongly to write top-to-bottom, so they end up either left-to-right within top-to-bottom, or top-to-bottom within left-to-right?
Aren't most right-to-left scripts that are written with ink vertical scripts, so that characters are top-to-bottom in each line but lines are right-to-left on each page?

DesEsseintes wrote:@Clawgrip: you are very good at these guides! [:D] Thanks for taking the time!
What DesEsseites said. [+1]

As for my (so far only) conscript, I've done some of what clawgrip recommends and not followed some other recommendations.
It would be easy to learn my script on a training grid such as is used by CJK learners.
It's all straight lines, and they can run in only eight directions. (You could call these North-South, West-East, Northwest-Southeast, Northeast-Southwest, NNW-SSE, NNE-SSW, WNW-ESE, ENE-WSW.)
The letters are "square" and each must "fill up" a square (that is, touch all four sides).
There are also some simple restrictions to make it easy to create a cursive form of the script.

Then there are some distinctly unnaturalistic features.
It's bottom-to-top boustrephedon within right-to-left. That is, the first line starts at the bottom right corner and goes up the righthand side to the top right corner; then the second line starts at the top just to the left of the end of the first line and goes down parallel to the first line until it gets to the bottom; then the third line starts at the bottom just left of the second line's end and goes up … etc. …
(I think if I were creative enough I could come up with a reason this script would have devoloped that way; but I haven't done so yet, so maybe I'm not creative enough.)

Because of the above, no two different characters are mirror-images of each other, neither right-to-left nor top-to-bottom. If a character is reflected through either a horizontal or a vertical line, the reflection still means the same thing. Note: by no means will it still look the same, at least not usually; most of the characters are not even two-fold symmetric, much less four-fold symmetric.


Quite possibly this script sucks. I'll have to work on it some more.

But in the mean time, does anyone have an idea why the script should have been developed more-or-less as I've described, in a general kind of way?

User avatar
Xing
MVP
MVP
Posts: 4256
Joined: 22 Aug 2010 18:46

Re: On making a conscript that doesn't suck.

Post by Xing »

I'm a bit of a newbie when it comes to actual glyph creation. When I started to work on the Gaku script, I wasn't really satisfied with the appearance of the glyphs. They looked to unsystematic - as if I had just thrown various kinds of shapes together, without any underlying 'feel', that gave the script a distiguishable appearance.

I think I'm becoming more and more satisfied now. I scribble quite a lot, and the glyphs start to get more and more of a 'common feel'. At first, pretty much any shape was possible. I drew each new glyph without much though of whether this or that kind of line was 'permitted' or not. The result was a rather unsystematic script. Then I started to sort out the kind of 'basic' shapes that I decided shoud be allowed. As one scribbles, one might decide to scrap/merge certain kinds of basic shapes, and to add others - which might have been 'forbidden' in the first draft of the script.

I don't bother too much about the writing materials available to the hypothetical speakers of the language. There are undoubtedly some features of the appearance of a script that may show clear influence from writing material, But (1) there are many features of the script that is independent thereof, and (2) those features that are dependent on writing material could be treated as different typefaces of the same underlying script. The Roman alphabet can be written in many different ways - which may be suited for different available writing material. But many of those ways could still easily be recognised as forms of the Roman alphabet. In the same way, I could create versions of the Gaku script suited for carving or brush-writing, which would nevertheless preserve much of the underlying structure of the glyph.

User avatar
Lambuzhao
korean
korean
Posts: 5988
Joined: 13 May 2012 02:57

Re: On making a conscript that doesn't suck.

Post by Lambuzhao »

Xing wrote:I'm a bit of a newbie when it comes to actual glyph creation. When I started to work on the Gaku script, I wasn't really satisfied with the appearance of the glyphs. They looked to unsystematic - as if I had just thrown various kinds of shapes together, without any underlying 'feel', that gave the script a distiguishable appearance.
Do folks feel that scripts where types of sounds (e.g. alveolars, velars, labials, etc) have a common shape are more logical/systematic/preferable, or of not much consequence, or more confusing/illegible?

User avatar
eldin raigmore
korean
korean
Posts: 5605
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: On making a conscript that doesn't suck.

Post by eldin raigmore »

Lambuzhao wrote:Do folks feel that scripts where types of sounds (e.g. alveolars, velars, labials, etc) have a common shape are more logical/systematic/preferable, or of not much consequence, or more confusing/illegible?
In real life "featurographies" (where if two phonemes share some characteristic feature, then the characters representing them do too) turn out to be confusing/illegible.
That doesn't, and IMO shouldn't, keep people from trying, though.
Take a look at Alexander Melville Bell's "visible speech", for instance, and tell me what you think of it.
I think I could read it, though I imagine "speed-reading" might run into problems.

If your character-set is a syllabary, though, it might make sense for syllables with the same onset to share some graphic feature; syllables with the same nucleus share some other feature; and (assuming there are codas) syllables with the same coda should share some third graphic feature.
AIUI Hange'ul (sp?) is like that, and that even makes it easier, not harder, to read it.

Does any of that help?

User avatar
Sights
sinic
sinic
Posts: 211
Joined: 04 Jan 2014 20:47

Re: On making a conscript that doesn't suck.

Post by Sights »

eldin raigmore wrote: Then there are some distinctly unnaturalistic features. It's bottom-to-top boustrephedon within right-to-left. That is, the first line starts at the bottom right corner and goes up the righthand side to the top right corner; then the second line starts at the top just to the left of the end of the first line and goes down parallel to the first line until it gets to the bottom; then the third line starts at the bottom just left of the second line's end and goes up … etc. …
(I think if I were creative enough I could come up with a reason this script would have devoloped that way; but I haven't done so yet, so maybe I'm not creative enough.)
But in the mean time, does anyone have an idea why the script should have been developed more-or-less as I've described, in a general kind of way?
I don't think this will be of much use but since, as you well said, those are already somewhat unnatural traits, I'll share the very capricious reasoning behind the script I'm designing, which is also bottom-to-top boustrophedon (but left-to-right). The whole thing is centered around one theme: fire. The characters are supposed to be subtly reminiscent of the movement and unstability of flames, so a good deal of variety in shapes, within reason and aesthetic coherence, is seen as a good thing. As I see it, that also makes the bottom-to-top initial directionality a bit more believable: flames grow upward. Granted, in that regard keeping bottom-to-top direction would make more sense, but boustrophedon allows for more practical reading and writing (I think so, at least?). The script is still in the early stages of development, and I might change a lot of things later on, but some of these unlikely explanations have already grown on me. [;)]

(Particularly, having the script inventor turn out to be nothing but an overpraised pyromaniac with a strange outlet)

I suppose you could also play the "cryptograph" card, and explain some of the weirder bits by saying that the script originated as purposefully difficult/complex writing system, meant to limit the abilities of writing and reading to a specific group, which would make some anthropological or historical sense. I feel these are the kind of things that can later be tweaked and improved by having some more realistic considerations about the evolution of writing systems in mind, but some features could be kept and explained by ceremonial or cultural, rather than practical, reasons. Hope that helped in some way [:)]

User avatar
Lambuzhao
korean
korean
Posts: 5988
Joined: 13 May 2012 02:57

Re: On making a conscript that doesn't suck.

Post by Lambuzhao »

eldin raigmore wrote:
Lambuzhao wrote:Do folks feel that scripts where types of sounds (e.g. alveolars, velars, labials, etc) have a common shape are more logical/systematic/preferable, or of not much consequence, or more confusing/illegible?
In real life "featurographies" (where if two phonemes share some characteristic feature, then the characters representing them do too) turn out to be confusing/illegible.
That doesn't, and IMO shouldn't, keep people from trying, though.
Take a look at Alexander Melville Bell's "visible speech", for instance, and tell me what you think of it.
I think I could read it, though I imagine "speed-reading" might run into problems.
[<3] Visible Speech. Have two fonts for it on my cruddy ole PC.

Still, the less attention I pay to it, the more it looks like hoop earrings and xmas ornament hooks.
If there was a way to accentuate the differences more, without sacrificing the general shapes,
I think that would go a long way.

If your character-set is a syllabary, though, it might make sense for syllables with the same onset to share some graphic feature; syllables with the same nucleus share some other feature; and (assuming there are codas) syllables with the same coda should share some third graphic feature.
AIUI Hange'ul (sp?) is like that, and that even makes it easier, not harder, to read it.

Does any of that help?
Sort of, yes. I was actually thinking to mention Hange'ul, but I don't know a lot about it. I looked it up, and the consonant glyphs behave pretty much optimally in the way I was thinking. That is a clever writing system-
Kamsam hamnida!
[:D]

Plus, I'm thinking of the / " ° / consonants of Hiragana and Katakana, and to some degree the Inuktitut Syllabary, and insofar as established :con: go, Tengwar.

User avatar
eldin raigmore
korean
korean
Posts: 5605
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: On making a conscript that doesn't suck.

Post by eldin raigmore »

Sights wrote:I suppose you could also play the "cryptograph" card, and explain some of the weirder bits by saying that the script originated as purposefully difficult/complex writing system, meant to limit the abilities of writing and reading to a specific group, which would make some anthropological or historical sense. I feel these are the kind of things that can later be tweaked and improved by having some more realistic considerations about the evolution of writing systems in mind, but some features could be kept and explained by ceremonial or cultural, rather than practical, reasons. Hope that helped in some way [:)]
Thanks!
This inspires me to make the real motive, be my in-story motive.

My real motive for making it that way was to save as much physical effort on the part of the writer as possible.

Assuming s/he's right-handed, s/he can at least begin writing without even moving his/her hand.
And when s/he does lift his/her hand, the writing of the first line goes right along with that.
And then the writing of the second line just follows putting his/her hand back down.

So if I just make the early adopters of this writing system be a bunch of lazy-assed slackers who didn't even want to see what they'd already written if that took any extra effort, that will explain it.

Wha'd'ya think?


[hr][/hr]

Lambuzhao wrote:.... That is a clever writing system....
[+1]
Next time I see him, I'll be sure to tell Emperor Sejong that you said you like his work.

clawgrip
MVP
MVP
Posts: 2234
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: On making a conscript that doesn't suck.

Post by clawgrip »

eldin raigmore wrote:
Lambuzhao wrote:Do folks feel that scripts where types of sounds (e.g. alveolars, velars, labials, etc) have a common shape are more logical/systematic/preferable, or of not much consequence, or more confusing/illegible?
In real life "featurographies" (where if two phonemes share some characteristic feature, then the characters representing them do too) turn out to be confusing/illegible.
That doesn't, and IMO shouldn't, keep people from trying, though.
Take a look at Alexander Melville Bell's "visible speech", for instance, and tell me what you think of it.
I think I could read it, though I imagine "speed-reading" might run into problems.

If your character-set is a syllabary, though, it might make sense for syllables with the same onset to share some graphic feature; syllables with the same nucleus share some other feature; and (assuming there are codas) syllables with the same coda should share some third graphic feature.
AIUI Hange'ul (sp?) is like that, and that even makes it easier, not harder, to read it.

Does any of that help?
I would hesitate to say that featural scripts definitely make a script difficult to read; it depends on how you do it, and limited, obvious featural elements can be useful. The diacritics of Roman alphabet itself can make it a partially featural script, and the dakuten of hiragana and katakana that indicates voicing is also a featural element that works just fine. Same with tone marks of various southeast Asian scripts. It's just a matter of how extreme you get, with visual speech taking it to a ridiculous extreme.

User avatar
eldin raigmore
korean
korean
Posts: 5605
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: On making a conscript that doesn't suck.

Post by eldin raigmore »

clawgrip wrote:I would hesitate to say that featural scripts definitely make a script difficult to read; it depends on how you do it, and limited, obvious featural elements can be useful. The diacritics of Roman alphabet itself can make it a partially featural script, and the dakuten of hiragana and katakana that indicates voicing is also a featural element that works just fine. Same with tone marks of various southeast Asian scripts. It's just a matter of how extreme you get, with visual speech taking it to a ridiculous extreme.
What "featurography" means to me is that the glyph for each phoneme is composed entirely out of the sub-glyphs for that phoneme's distinctive features.

For many abugidas, each character represents an open CV syllable with a one-consonant onset; the syllable represented by that character has an "inherent vowel" if there's no diacritic.
A diacritic can change that vowel, and/or change the syllable from CV to VC, and/or make the consonant stand alone without any vowel.

That's a featural element, and it doesn't hurt legibility nor writability. But it also doesn't turn the entire abugida into a featurography.

I agree that a few "featurographic" additional notations can make a script that's still highly usable.

But I think that if, say, more than 50% (just to pull a number out of my hat) of the typical phoneme's graphic representation is made up out of such featurographic symbols, then it becomes maybe a little too easy to get mixed up while reading; or, rather, that seems to be the gist of what I've read about other people's experiences with such scripts.

Still, I think that conlangers and neographers (folks who make "con"scripts) should not be discouraged from trying to "do it right".

clawgrip
MVP
MVP
Posts: 2234
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: On making a conscript that doesn't suck.

Post by clawgrip »

In that case I guess we agree.

A large family of scripts I made started out as a fairly featural script. I had fun trying to obscure and obliterate the featural elements as much as possible in the descendant scripts.

User avatar
Chagen
runic
runic
Posts: 3381
Joined: 03 Sep 2011 05:14
Location: Texas

Re: On making a conscript that doesn't suck.

Post by Chagen »

So this thread has strayed far from my conscript in topic...
Nūdenku waga honji ma naku honyasi ne ika-ika ichamase!
female-appearance=despite boy-voice=PAT hold boy-youth=TOP very be.cute-3PL
Honyasi zō honyasi ma naidasu.
boy-youth=AGT boy-youth=PAT love.romantically-3S

User avatar
Ahzoh
mongolian
mongolian
Posts: 4025
Joined: 20 Oct 2013 02:57
Location: Canada

Re: On making a conscript that doesn't suck.

Post by Ahzoh »

Chagen wrote:So this thread has strayed far from my conscript in topic...
yes, it has.
Image Ӯсцьӣ (Onschen) [ CWS ]
Image Śāt Wērxālu (Vrkhazhian) [ WIKI | CWS ]

User avatar
eldin raigmore
korean
korean
Posts: 5605
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: On making a conscript that doesn't suck.

Post by eldin raigmore »

Chagen wrote:So this thread has strayed far from my conscript in topic...
Haven't the majority of posts on the second half of this thread been relevant at least to the thread's title? (Seems that way to me.)
Or even to your original post? (Maybe not?)

User avatar
Xing
MVP
MVP
Posts: 4256
Joined: 22 Aug 2010 18:46

Re: On making a conscript that doesn't suck.

Post by Xing »

eldin raigmore wrote:
Chagen wrote:So this thread has strayed far from my conscript in topic...
Haven't the majority of posts on the second half of this thread been relevant at least to the thread's title? (Seems that way to me.)
Or even to your original post? (Maybe not?)
IMO, they have at least been related to the topic in the OP. But anyway, we could alway return to the original question.

Spoiler:
Chagen wrote:Feels embarrassing to post in the Beginners' Corner but I don't know if this is suitable in the Conlangs forum.

I've been trying to make a script for Sunbyaku. However, I am completely stuck. Conscripts are easily the hardest part of a conlang for me to the point where I just don't make them often. But I feel like I need to try at least ONCE.

Anyway: Sunbyaku's native Sunkirnahashi is a syllabary much like Japanese's native kana. Indeed, the entire thing is supposed to be asthetically like Hiragana--lots of flowy lines combined with straight strokes. However, Hiragana was formed from simplifying a bunch of chosen kanji. Sunkirnahashi, on the other hand, was completely home-made in Sunzaku and thus came from nothing before. Thus, I'm finding it hard to come up with a suitable history for each letter even if the actual syllabary itself has a history (it was devised by a monk as his lord asked him to come up with a suitable writing system that was easy to learn and logical, based off proto-writing found in said lord's lands). It doesn't help that I need a LOT of glyphs (roughly 67), and they need to be easily distinguishable as this is not using an logography like Japanese does. Thankfully I have no equivalent to katakana...

So how exactly can I start out with this? Because I don't even know how to BEGIN.
Could you state some general 'rules' for what the glyphs must look like?
Are certain shapes forbidden? (Like, ascenders and descenders? Loops? Circles? Perfectly symmetric characters?)
Could the permitted shapes/strokes be arranged into some basic kinds? (Like, 'horizontal stroke', 'vertical stroke', 'downward curve', etc.)
Is there a standard stroke order in hand-writing?

(As the script evolves, it's almost certain that some of the answers to the above questions will change!)

I know the problem - I can be quite a challenge to create a script where the glyphs have enough in common for the script to have a distinct appearance, and yet distinct enough to make reading easy.

In my Gaku script (which is about to be revised at the moment), I have some constraints:
-The script is monospace - each character takes up an equally-sized square.
-At least one stroke in each glyph much reach each the four sides of the square (top, bottom, left and right).
-There are no ascenders or descenders.
-If two straight lines meet, they must do so at a 90˚ angle.

Hopefully, these constraints will give the script a distinct appearance. Various means are utilised in order to make the individual glyphs more distinct from each other:
-There are but complex/clutterd glyphs with many strokes, and simpler ones with only a few strokes.
-Both straight and curved lines, as well as loops and circles, are allowed.

clawgrip
MVP
MVP
Posts: 2234
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: On making a conscript that doesn't suck.

Post by clawgrip »

Xing wrote:I'm a bit of a newbie when it comes to actual glyph creation. When I started to work on the Gaku script, I wasn't really satisfied with the appearance of the glyphs. They looked to unsystematic - as if I had just thrown various kinds of shapes together, without any underlying 'feel', that gave the script a distiguishable appearance.

I think I'm becoming more and more satisfied now. I scribble quite a lot, and the glyphs start to get more and more of a 'common feel'. At first, pretty much any shape was possible. I drew each new glyph without much though of whether this or that kind of line was 'permitted' or not. The result was a rather unsystematic script. Then I started to sort out the kind of 'basic' shapes that I decided shoud be allowed. As one scribbles, one might decide to scrap/merge certain kinds of basic shapes, and to add others - which might have been 'forbidden' in the first draft of the script.
I think this is a good way to do it. It's possible to impose specific rules if you know what you want, but it's also possible to let them form themselves. Chagen has already specified a couple rules or characteristics (namely, flowing lines combined with straight strokes). You can keep these as a guiding principle, and then rewrite the glyphs repeatedly, and see if they start to fall in line with each other.

User avatar
Chagen
runic
runic
Posts: 3381
Joined: 03 Sep 2011 05:14
Location: Texas

Re: On making a conscript that doesn't suck.

Post by Chagen »

It seems that this has been turned into the general noob question thread for conscripts. I'm fine with that.

Clawgrip: I've been actually meaning to ask you, does my script as is actually get the hiragana-influenced feel down? I've noticed that bunch of my characters are way simpler than hiragana, nothing as complex as む め み な を ほ ふ, to name a few examples. I keep thinking that I can't have too many strokes, but the Japanese seem to do fine.

Hiragana has this quality though that just plain eludes me. I can figure out all of the things it has that make it unique and different from, say, western scripts; a lot more lines, emphasis on smooth curves, characters are disparate lines just sort of floating together with no obligation to connect (<を> just looks like two characters awkwardly crammed together yet somehow it works, same for <け れ せ そ, etc. Yet, when trying to make Sungirna-hashi I can't make those same elements work together. My glyphs right now look like some sloppy travesty of hiragana.
Nūdenku waga honji ma naku honyasi ne ika-ika ichamase!
female-appearance=despite boy-voice=PAT hold boy-youth=TOP very be.cute-3PL
Honyasi zō honyasi ma naidasu.
boy-youth=AGT boy-youth=PAT love.romantically-3S

Post Reply