On Hanying and Creole Adjustment

[bradrn]

For comparison, serifs have been around since the Romans, and yet they still aren't part of the prototype.

I can't agree, when before 1800 every printed font had serifs! Sans serif fonts seemed so odd when they were introduced that they were called "grotesque". (I don't mean people said they were ugly, I mean that was the actual category name.)

I would say that this point of view is dependent on what you consider to be the 'prototype'. I would consider a useful definition to derive from your 'cartoon' observation in the Lexipedia:

(I might add that if you ask a cartoonist to draw it, they’ll probably draw the prototype. From this we can learn that the prototypical hunk of cheese is Swiss, with really big holes.)

Rosenfelder, Mark. The Conlanger's Lexipedia (p. 54). Yonagu Books. Kindle Edition.

Starting from this, I would define the prototype of a letter as being the form it takes when you try to handwrite it most legibly with a monoline pen. According to this definition, serifs never have consistently been part of the prototype; one effect of this is that most people don't write the serifs on most letters.

Even in handwriting, people preserved things like the straight bars in lowercase u, m, n. Many people always write the serifs in capital I.

I never even thought about this! As someone who has always written I with two large serifs, I can safely say that for me that is the new prototype. I suppose that - for the odd letter - it isn't too implausible for the prototype to evolve slightly. But I still have trouble with the completely different prototypes of the Hanying alphabet.

(Cursive is really a whole 'nother category which has prototypes of its own. There's nothing odd about different writing styles having different prototypes-- compare italic and roman a or g, for instance.)

No, cursive doesn't have prototypes of its own - I would say it has the same prototypes as all other text. On the other hand, it's true that some letters have multiple prototypes - to borrow a word from your Lexipedia, I would call these letters polytypical.

But my personal hypothesis is that once you have printing, letter prototypes stop evolving - and that's even more true when you're using computers.

I agree that printing can exert a conservative influence. We still like the early humanistic fonts. But we also like variety, and there's been a load of new font ideas since 1500. Computers only add to the variety possible.

But compare those with Fraktur and other black-letter fonts. That was Gutenberg's own prototype, and basically those are gone except when we want an old-timey display face. Germany kept printing books in Fraktur well into the last century, but it would be extremely eccentric to do so today.

This is exactly what I mean! If not for printing, Fraktur would probably have become a different alphabet by now, but printing forced the prototypes to remain the same.

Even a little thing like the use of the long s seems very weird to modern readers; to me, it's a flaw in the otherwise astonishingly beautiful web version of Byrne's Euclid.

Long s is a completely different letter, not just a different prototype. I wouldn't call the loss of long s a change in the prototype; I would call it 'spelling reform'.

Besides the change from Fraktur, note the simplified characters for Chinese, or for that matter changes of alphabet in Turkey and elsewhere, or the switch to all-Hangul in North Korea, all post-printing developments.

These were top-down changes, enforced by the government, whereas prototype change is an organic process. If you had written something along the lines of "The new Hanying alphabet was created and implemented by the Areopolis government as a new 'simplified latin' alphabet", that would be a bit more plausible.

Here's a simple question: does the prototype for capital A include the crossbar? I have a Letraset catalog from 1977, and it's in every font, including the wackiest display fonts. Yet there are several modern fonts that omit it, and as I noted designers of 'future alphabets' love to leave it out. If these things look normal to us now, the crossbar is at least weakened within the prototype.

I can't argue with that. For me, the serifs on I are also weak compared to the prototype.

As Wikipedia notes, it's just too different to the modern English alphabet to be useful for teaching.

That's a poor lesson to take! There's fashions in teaching just as there are in font design. The ball-and-stick letters often taught in grade school are even more influential. For another example of a teaching alphabet, consider the use of pinyin in China. My understanding is that pinyin is taught first as a step to literacy, with characters introduced later. (The idea is not to replace hanzi.)

I've never heard of 'ball-and-stick letters'. And I didn't know that Chinese schools used pinyin as a teaching alphabet. I suppose it is conceivable that the Hanying alphabet could have been introduced for teaching.

[zompist]

Starting from this, I would define the prototype of a letter as being the form it takes when you try to handwrite it most legibly with a monoline pen. According to this definition, serifs never have consistently been part of the prototype; one effect of this is that most people don't write the serifs on most letters.

But you've chosen your definition to lead to your conclusion! Elizabethans didn't have monoline pens. Serifs are precisely a thing because of how calligraphic nibs work. I hope you wouldn't define the prototypes for cuneiform, or hanzi, based on how Bic pens work!

(Cursive is really a whole 'nother category which has prototypes of its own. There's nothing odd about different writing styles having different prototypes-- compare italic and roman a or g, for instance.)

No, cursive doesn't have prototypes of its own - I would say it has the same prototypes as all other text. On the other hand, it's true that some letters have multiple prototypes - to borrow a word from your Lexipedia, I would call these letters polytypical.

At this point we're mostly just disagreeing with each other, but I'd suggest taking a long look at this picture of Russian cursive, and comparing it to the printed letterforms. There are some very striking differences, and only a minority of the forms stick close to the printed forms. (Also note that these are very careful forms-- actual handwriting merges some of the forms, so that e.g. it's common to add a line above a lowercase т so it's not confused with ш.)

This is exactly what I mean! If not for printing, Fraktur would probably have become a different alphabet by now, but printing forced the prototypes to remain the same.

I don't think we can convince each other when we look at the same evidence and conclude opposite things! I have (somewhere) a document from one of my ancestors in cursive Fraktur-- it's basically unreadable. By your own criterion (handwritten forms), the prototypes are very different.

To me, fonts have changed dramatically and constantly in five centuries. And I think that will continue, over a timespan many times this length. If you disagree, things can go quite differently in your conworld.

[bradrn]

Starting from this, I would define the prototype of a letter as being the form it takes when you try to handwrite it most legibly with a monoline pen. According to this definition, serifs never have consistently been part of the prototype; one effect of this is that most people don't write the serifs on most letters.

But you've chosen your definition to lead to your conclusion! Elizabethans didn't have monoline pens. Serifs are precisely a thing because of how calligraphic nibs work. I hope you wouldn't define the prototypes for cuneiform, or hanzi, based on how Bic pens work!

I would say it is a valid definition though - it seems to work for most writing systems other than cuneiform. (Including hanzi - so-called 'Gothic' fonts use monolines.) On the other hand, if you have a better definition of the meaning of a 'letter prototype', I would be happy to use yours.

(Cursive is really a whole 'nother category which has prototypes of its own. There's nothing odd about different writing styles having different prototypes-- compare italic and roman a or g, for instance.)

No, cursive doesn't have prototypes of its own - I would say it has the same prototypes as all other text. On the other hand, it's true that some letters have multiple prototypes - to borrow a word from your Lexipedia, I would call these letters polytypical.

At this point we're mostly just disagreeing with each other, but I'd suggest taking a long look at this picture of Russian cursive, and comparing it to the printed letterforms. There are some very striking differences, and only a minority of the forms stick close to the printed forms. (Also note that these are very careful forms-- actual handwriting merges some of the forms, so that e.g. it's common to add a line above a lowercase т so it's not confused with ш.)

I completely forgot about Russian cursive. You're right about this.

This is exactly what I mean! If not for printing, Fraktur would probably have become a different alphabet by now, but printing forced the prototypes to remain the same.

I don't think we can convince each other when we look at the same evidence and conclude opposite things! I have (somewhere) a document from one of my ancestors in cursive Fraktur-- it's basically unreadable. By your own criterion (handwritten forms), the prototypes are very different.

After thinking it through, I do think you are right about this - the prototypes are indeed different.

To me, fonts have changed dramatically and constantly in five centuries. And I think that will continue, over a timespan many times this length. If you disagree, things can go quite differently in your conworld.

I wouldn't say fonts have changed all that dramatically. Here is a sample of text from 1487:


(from https://ilovetypography.com/2016/04/18/ ... man-fonts/)

And now compare it to the same text in Trebuchet:

ſenza altro conſiglio era
reſiſtetia ſifaceua grade
daua animici che dalloro

They're the same apart from the serifs in the original, slightly differing stroke widths at some points, and different gaps in the g (which is polytypical anyway).

____________________________

I'm starting to think that we've gotten a bit distracted from my original question:

Why is the Hanying alphabet so different to the modern-day alphabet? (You can't explain it using touch input, because that would only be a minority of devices, unless something happens to change this sometime in Incatena history.)

Potentially any further discussion of the evolution of prototypes should be relegated to some other thread (the 'Writing History Thread'?). It's certainly an interesting topic.

[zompist]

I would say it is a valid definition though - it seems to work for most writing systems other than cuneiform. (Including hanzi - so-called 'Gothic' fonts use monolines.) On the other hand, if you have a better definition of the meaning of a 'letter prototype', I would be happy to use yours.

Eh, just for discussion, I'd say "the most commonly accepted representation of the letter in the dominant medium for producing it."

In medieval times, that might be a scribe using a thick nib; in China, letters written with a brush.

In the printing era, I think the prototype is arguably a printed character. You may not agree, but think again about China, where characters are drastically and idiosyncratically simplified when written by hand; the prototype is surely something more like the kaishu hand, which underlies the printed characters. It's not someone's ballpoint-pen approximation.

(Note that prototypes can be formally tested... and should be! A prototype should be easily elicited, but also immediately recognized, and that would take research to find out.)

I wouldn't say fonts have changed all that dramatically. Here is a sample of text from 1487:

And now compare it to the same text in Trebuchet:

But that's cooking your examples! Yes, we still love the Italian humanistic letterforms, and they've prevailed in areas that once had different standards. German fonts of the same age are far less readable today. But look a bunch of modern display fonts!

You've chosen two fonts that are closely related except for serifs. Looking through that Letraset book, here's some readily apparent variation at the prototype level:

a: bow above the bowl, or no?
g: two closed curves, or just a swash below?
f: descender, or none?
z: flat or curved bottom?
Q: stroke or swash?
mnu: straight stroke at the side, or just curves?
t: tail or no?
v: right half straight or curved?
x: made of two straight lines, or two curves?
R: diagonal starts at far left, or is there a crossbar?

As for the original question, I think I've answered it several times over: I think letterforms will continue to change, because they always have.
People won't have the exact same design preferences in 3000 years!

[bradrn]

I would say it is a valid definition though - it seems to work for most writing systems other than cuneiform. (Including hanzi - so-called 'Gothic' fonts use monolines.) On the other hand, if you have a better definition of the meaning of a 'letter prototype', I would be happy to use yours.

Eh, just for discussion, I'd say "the most commonly accepted representation of the letter in the dominant medium for producing it."

In medieval times, that might be a scribe using a thick nib; in China, letters written with a brush.

Yes, that's a better definition.

In the printing era, I think the prototype is arguably a printed character. You may not agree, but think again about China, where characters are drastically and idiosyncratically simplified when written by hand; the prototype is surely something more like the kaishu hand, which underlies the printed characters. It's not someone's ballpoint-pen approximation.

I'm not sure I agree with this. Why is the handwritten simplification the prototype with hanzi, but not with Latin letters?

(Note that prototypes can be formally tested... and should be! A prototype should be easily elicited, but also immediately recognized, and that would take research to find out.)

Yes, that would be ideal.

I wouldn't say fonts have changed all that dramatically. Here is a sample of text from 1487:

And now compare it to the same text in Trebuchet:

But that's cooking your examples! Yes, we still love the Italian humanistic letterforms, and they've prevailed in areas that once had different standards. German fonts of the same age are far less readable today. But look a bunch of modern display fonts!

You've chosen two fonts that are closely related except for serifs. Looking through that Letraset book, here's some readily apparent variation at the prototype level:

a: bow above the bowl, or no?
g: two closed curves, or just a swash below?
f: descender, or none?
z: flat or curved bottom?
Q: stroke or swash?
mnu: straight stroke at the side, or just curves?
t: tail or no?
v: right half straight or curved?
x: made of two straight lines, or two curves?
R: diagonal starts at far left, or is there a crossbar?

I wouldn't say that that's variation at the protoype level - I would say that's variation 'above' the prototype level. Let's look through that list in detail:

a, g: As already noted, these are polytypical, which is the 'prototype variation' you're describing.
f: The prototype would be the Trebuchet form for me; your variations are just varying how far down the prototype descends.
t: The primary distinguishing feature of this prototype is a cross at the top of the letter, and no 'curving over' at the top (which distinguishes t/f). A tail isn't part of it, for me.
x: Depending on your perspective, this could be called polytypical. However I would argue that it's just variation 'above' the prototype again - the primary distinguishing feature of x is four lines/curves radiating out from a central point at roughly NW/NE/SW/SE angles.
R: I'm not sure what you're trying to say here.
zQmnuv: I think the observed variation with these is in a feature where the prototype is ambiguous as to curvature/shape/precise placement.

As for the original question, I think I've answered it several times over: I think letterforms will continue to change, because they always have.
People won't have the exact same design preferences in 3000 years!

At this point I'm beginning to think it's simply a difference in opinion. You think letterforms will continue to change, because they always have; I think that they won't, because the prototypes haven't changed all that much since Roman times. There's no way to know who's right - no-one can predict the future.

[zompist]

In the printing era, I think the prototype is arguably a printed character. You may not agree, but think again about China, where characters are drastically and idiosyncratically simplified when written by hand; the prototype is surely something more like the kaishu hand, which underlies the printed characters. It's not someone's ballpoint-pen approximation.

I'm not sure I agree with this. Why is the handwritten simplification the prototype with hanzi, but not with Latin letters?

Well, you've pointed out yourself the influence of printing on Western letterforms! I expect most of us see printed letters far more than we see handwritten ones.

As for Chinese, it seems pretty clear that the printed forms are close imitations of (very careful) brushed strokes. There are modernistic or display faces, but they've stayed far closer to the calligraphic tradition than we have.

[bradrn]

In the printing era, I think the prototype is arguably a printed character. You may not agree, but think again about China, where characters are drastically and idiosyncratically simplified when written by hand; the prototype is surely something more like the kaishu hand, which underlies the printed characters. It's not someone's ballpoint-pen approximation.

I'm not sure I agree with this. Why is the handwritten simplification the prototype with hanzi, but not with Latin letters?

Well, you've pointed out yourself the influence of printing on Western letterforms! I expect most of us see printed letters far more than we see handwritten ones.

As for Chinese, it seems pretty clear that the printed forms are close imitations of (very careful) brushed strokes. There are modernistic or display faces, but they've stayed far closer to the calligraphic tradition than we have.

Oops; my bad. I was lazy and didn't look up kaishu; those were indeed what I was thinking of as 'hanzi prototypes'.

And, after reflection, I do think you have a point about Western printed letterforms. My prototypes are probably close to geometric sans-serifs. As for when sans-serif fonts didn't exist, we can only speculate about what the prototypes looked like.

[Yiuel Raumbesrairc]

At this point I'm beginning to think it's simply a difference in opinion. You think letterforms will continue to change, because they always have; I think that they won't, because the prototypes haven't changed all that much since Roman times. There's no way to know who's right - no-one can predict the future.

Prototypes have changed a lot since Roman times. The biggest change did occur before printing, as one can see between the Roman book hand,and the carolingian minuscule, which is still quite far from our own minuscule in style :



Can printing slow down the evolution of characters? Obviously; it's one of the effects of standardization. But it doesn't stop it completely. French was full of that awful s, written like an f without the middle stroke, until very recently. "a"is another weird one, where we seem we cannot decide on either the fanciful form giving on this board, or a simple circle with a vertical stroke on the right, as with the other IPA character giving us the back open unrounded vowel.

Note that simplification has also occured in kanji, the Japanese version of hanzi, and these simplifications were created from the bottom and spread to the standard, all this while printing still existed.

Also, we don't have much experience in having over a thousand years since printing. China has the longest history, and while a standard has been created, it relates mostly to an educational standard that has maintained itself, not without a few changes. Despite lifespans getting longer in the Incatena, I wouldn't be surprised if, at one point, some kind of lettershape reform would occur just because certain lettertypes would be easier to write on pads with fingers.

[bradrn]

At this point I'm beginning to think it's simply a difference in opinion. You think letterforms will continue to change, because they always have; I think that they won't, because the prototypes haven't changed all that much since Roman times. There's no way to know who's right - no-one can predict the future.

Prototypes have changed a lot since Roman times. The biggest change did occur before printing, as one can see between the Roman book hand,and the carolingian minuscule, which is still quite far from our own minuscule in style :

True. I suppose I really meant that they haven't changed much since the invention of printing (blackletter aside).

Can printing slow down the evolution of characters? Obviously; it's one of the effects of standardization. But it doesn't stop it completely. French was full of that awful s, written like an f without the middle stroke, until very recently. "a"is another weird one, where we seem we cannot decide on either the fanciful form giving on this board, or a simple circle with a vertical stroke on the right, as with the other IPA character giving us the back open unrounded vowel.

This is probably more accurate than my view - that nothing has changed. As for 'a', this letter has multiple prototypes (refer to the above discussion with zompist for more on this topic).

Note that simplification has also occured in kanji, the Japanese version of hanzi, and these simplifications were created from the bottom and spread to the standard, all this while printing still existed.

I'm not all that familiar with kanji - do you have any resources on this? It sounds very interesting.

Also, we don't have much experience in having over a thousand years since printing. China has the longest history, and while a standard has been created, it relates mostly to an educational standard that has maintained itself, not without a few changes.

Good point - we simply don't have enough history to discuss this subject properly.

Despite lifespans getting longer in the Incatena, I wouldn't be surprised if, at one point, some kind of lettershape reform would occur just because certain lettertypes would be easier to write on pads with fingers.

But this was my original query! As I was saying to zompist:

It could be several things... but hey, sometimes you gotta drive the authorial fiat. So...

At some point, probably in the late Douane era, some devices took gestural (or hand-drawn) input. For speed and to simplify the input space, letters had to be single strokes. The look caught on, helped by the relatively small population.

This sounds intriguing. The dominant input device today is the keyboard; how then did some devices come to use gestural/hand-drawn input?

[Pedant]

Another small query: if neurimplants rely on physical gestures to make the words...well, what's happened to various sign languages over the years? Likely in the 50th Century someone could very easily get new ears made, but how about in the intervening years where various forms of digital communication and sweep-gestures were becoming more popular? Is there an HSL one could have learned, and how well would it map to writing as opposed to "speaking"?

[zompist]

Another small query: if neurimplants rely on physical gestures to make the words...well, what's happened to various sign languages over the years? Likely in the 50th Century someone could very easily get new ears made, but how about in the intervening years where various forms of digital communication and sweep-gestures were becoming more popular? Is there an HSL one could have learned, and how well would it map to writing as opposed to "speaking"?

Different devices and time periods are involved here.
1. The solar era and early Incatena, 2300 - 3000. This is when the gestural alphabet was important, with certain machines, not neurimplants. I picture these as controls or ATMs, where designers actually wanted deliberate, physical signalling. (At this time, if nothing else, not everyone had neurimplants.) You might also use this way to teach children, who you want to actually learn the letterforms, rather than just learn to type.
2. The middle and mature Incatena, after 3000. Now everyone has a neurimplant, which respond to mental gestures. All those earlier devices are obsolete, except perhaps the educational ones.

Sign languages would not, most likely, line up with the boundaries of spoken languages. If, say, Valles had more Deaf kids from Anglophone backgrounds, it'd use ASL, while if Areopolis had more from Chinese families, and used CSL. (Perhaps both northern and southern CSL survived.)

Possibly at least some space communities wold actually teach everyone Sign, as a backup to radio in environments where you can't use sound: outer space, the surface of the moon, the oceans of Europa, etc.

I'd really expect a full writing system for Sign to develop in this century-- and not just the hand gestures, but facial expressions, tempo and aspect, anaphoric pointing, etc.

[BGMan]

A couple questions:

1) Is there a reason English numerals took over instead of Chinese? I could imagine Chinese numerals being more popular due to being monosyllabic and not having any awkward combinations like "thr". After all, they managed to take over even in the Thai language, which otherwise relies more on Sanskrit.

For example, here's my attempt to create Sino-English numerals, admittedly relying more on Cantonese than Mandarin:
it, nee, som, see, goh, lok, (t)set, bat, gow, sup

And even if they don't displace English, I could still imagine them co-existing with English numerals as they do in other East Asian languages.

2) Is there any real-world influence from Singlish, or Macanese Patua?

[mèþru]

Macanese Patua is already mostly dead in the present, and a revival is less likely than unilateral denucleartization of Russia

[zompist]

1 - It could have gone either way, but I expect more late-21C Chinese scientists to know English numbers, than English ones to know the Chinese. Very basic words in OH are taken from English.

2 - Some of the sentence-final particles are taken from Singlish.

[Arkasas]

Today in typos: the translations in the negative section of MH morphology are all positive. Also, there's a broken image at the top, which I'm guessing is meant to be Hanying written in its native alphabet.

[bradrn]

Another typo (I think): In the sound changes section of Modern Hanying, you have köni → kwöni caused by ö → wö /_k. Shouldn't this be ö → wö /k_ (i.e. you've swapped the _ and k)?

(Also, while I'm talking about sound changes, would it be possible to get the changes from HC to MH?)

[zompist]

Problems should be fixed; thanks!

OK, sound changes. There were a few I applied by hand; the one I remember is eliminating any remaining /j/.

C=pbtdkgfsxžčjmhnŋñrlwyvʔš
V=aeiouãẽĩõũɛɔəöüæ
O=ouɔ
E=eiɛ
A=aãẽĩõũwɛɔə
S=ptkbdgčj
F=fzhwrɣšž
N=ãẽĩõũ
U=aeiou
H=iuü
M=nmŋñ
J=aiu
∆=æeo

x/š/_
o/ö/_Ci
u/ü/_Ci
a/ə/_H
a/ə/#C(C)_#
e/ye/#C(C)_C
o/we/#C(C)_C
ö/wö/#C(C)_C
J/∆/#C(C)_
w/u/_w
kw/p/_
tw/p/_
gw/b/_
S/F/V_V
M//_C
C/Ca/_Ca
C/Co/_CO
C/Cə/_CE
wu/uw/_V
wo/ow/_V
y/š/V_V Þ
w/g/V_V
e/ə/C_#
N/U/_
k/ʔ/_#
s//_#
č/š/#_
j/š/_#
a//_#
l//V_V
k/č/_E
t/č/_E
d/l/#_
ž/d/_O
ñ/y/_
v//V_V
a/ə/VCVC_
ö/e/_
ü/i/_

[Neon Fox]

Sənágmehan.
2-not-passive-irr-see
I may not be seen.

The third example in the Modern Hanying Syntax section: Shouldn't the translation be "You may not be seen"?