https://www.dawn.com/news/1307254
Urdu has to be the most widely spoken non-standardized language, right?

>>1603
Really a lot of indo-iranian languages are shit shows, there's a reason English is becoming the language of education and the upper-class there

I really don't know languages much so I can't comment on it much. I think that one African tribe that communicates by clicking is pretty cool?

Speaking of desi languages, tamil script is a joke
ஏழை கிழவன் வாழைப் பழத் தோல் மேல் சருசருக்கி வழுவழுக்கி கீழே விழுந்தான்.

Also Mongolian script is an odd case. Mongolians in china use the old vertical alphabet (https://en.wikipedia.org/wiki/Mongolian_script)
While Mongolians proper use cyrillic officially but there is support for traditional script there too. https://en.wikipedia.org/wiki/Mongolian_Cyrillic_alphabet
ᠮᠣᠩᠭᠣᠯ ᠪᠢᠴᠢᠭ

>>1606
Looks cool, kind of reminds me of the LotR stuff, but I think that was based on Georgian or something in that area?

When I first saw Kartvelian script I definitely thought LoTR

>>1603
Hmmmm, I've found it kinda hard to find more info on Urdu's standards outside of this paki magazine. I imagine Devanagari is more regular given its massive importance, but there doesn't seem to be much info on that either.
Shame it doesn't mark vowels by default, that only really works with Semitic roots' being made up of consonants only.

>>1604
Sure, but that's not because of orthography, it's to have a lingua franca.
... Right?

>>1606
In terms of the characters themselves, I don't think Tamil is that bad. It has far more internal consistency than modern kana and way less base symbols to learn. The problem with all of these Vedic scripts is the same pre-war kana had: a thousand years without reforms. That's what makes it terrible.

>>1607
>vertical left-to-right, left-to-right
>the requirements of vowel harmony and syllable sequence usually indicate the correct sound
>actual ambiguities are rare
>All case suffixes, as well as any plural suffixes consisting of one or two syllables, are likewise separated
Huh, very nice, I like it. Breaks my browser, though.

Why is English so infamously bad though? Is the oft-repeated reason that the French messing with it true?
French in itself is also interesting, in that practicality was sacrificed for aesthetics. Arabic is the only language that does this too AFAIK.

>>1613
wouldn't it be the Viking's fault if anything?

Swedish orthography was very elegant before the 1906 spelling reform. For example, you used to write "skrifva" (to write) instead of the modern "skriva", even though both are pronounced the same. Before the spelling was standardised in 1801, you would normally spell it "skrifwa". The thinking in cases like this was that spelling should reflect the etymological relationship between words, so "skrifva" was spelled that way to tie it together with "skrift" (writing). This is why "ge" (to give) and "gåva" (gift) are both spelled with a g, even though "ge" is really pronounced with a y-sound.

>>1613
My understanding is that the Great Vowel Shift made many spellings that had been phonetic in Middle English no longer phonetic.

>>1603
Nope. That would actually be English! English orthography has no proper standard, instead being informally defined by dictionaries, style guides, and school curricula. That's also why different English-speaking countries spell certain words differently and have subtly different grammatical conventions.

>>1613
The first thing to consider is that the English lexicon has a lot of mixture from different language families. Celtic Bretons, Germanic Saxons, Nordic Danes, Frankish Normans, and romaboo academics. These words were introduced in different periods, in different ways, and with different sounds. Take the case of Japanese, for example, where compounds words using Chinese on'yomi can geminate, while Yamato kun'yomi compounds have rendaku. This kinda stuff can cause a lot of sound changes that aren't necessarily generalized.
Although, yeah, the French do have a lot to be blamed for. The "ti" in "nation" being a /sh/, the "su" in "measure" being a /j/, and their scribes changing things up. Anyways, compare French "gel", /jel/, with Germanic "get", /get/. gif amirite This alone would damn many a language.

The Great Vowel Shift did screw things up big time, as it happened around the time the printing press was introduced, which further simplified some spellings and somewhat paradoxically made them more resistant to change. But it's not like that was the only thing that warped their writing like cheap aluminum.
When you look into English phonology, you're gonna find a lot of splits and mergers. Like wine–whine, where the latter used to actually be pronounced /hw/. Compare "who" (hoo) with "what" (wat). Neither is the original. Then there's rhotic and non-rhotic (US "car" vs UK "caah"), t-glottalization making a flipping "t" into a ', like in "bri'ish" (sometimes it's actually an R instead, [ɾ]), trap-bath split lengthening a buncha vowels but only before certain consonants, or how the thorn (Þ/th) stands for both the voiced and unvoiced versions of the dental fricative (though vs through), and, speaking of gh, that's a combination that can be mute, an /f/, or a /g/, depending on its position, thank you very much cot–caught merger. Many such cases.
So one example you can find in Wikipeda is "getting better - [ɡɛʔɪŋ bɛʔə(ɹ)]". In more readable terms, "ge'ing be'ä." And this is actually pretty good as far as English sentences go.
(ps flip r-colored shit imagine having retroflex vowels lmaoooo)

>>1615
Keeping alive etymological elements is a must when it comes to keeping continuity, but at the same time after multiple sound changes have occurred and people can't tell they're related while speaking, that's the moment where updating spellings is justified. Especially when inconsistencies start to pile up.
Or you can go the opposite way and take the Jap approach of purposefully obscuring the fact that 休む and 安い are brothers, that 羨む is 裏+病む, and that 潔い is made up of 勇む+清い(itself likely 気+良い, but I haven't been able to confirm that).

language differences are a tool to isolate us from other people and promote rigid adherence to nationalist principles.

I hate all forms of language and wish we would standardize or obtain universal translators.

>>1618
Well if these things were not the case Japanese society and anime would not even exist. We would all be one homogenous culture and the world would be a much less interesting place for it.

>>1619
not being able to communicate with one another doesn't promote culture. That's like saying all imageboards are basically 4chan with simple differences

>>1618
I very much disagree that it is nationalist, as shown in this thread, language evolves based on cultural influence. Look at how North Korea has a lot of Chinese and Russian loanwords whereas the southern dialect has mostly English.
Linguistic nationalism is rare and with the exception of notable cases like Ambazonia (which is more about recreating colonial borders), they would find something else to rally around.

Globalism is just a prettier version of imperialism

>>1620
Most image boards are basically 4chan...
But ignoring that, you yourself said that it promotes nationalism, nationalism promotes culture.
Further ignoring that though, shared language enables ideas and culture to spread rapidly and with ease. If that is prevented such as in Japan then a culture becomes more insular and more distinct.

There is also the nature of language itself, languages themselves promote cultures.

>>1619
Ehh, "Japanese society" underwent a heavy Westernization process a long time ago due to regular contact with European traders and missionaries, and then the US's gunboat diplomacy and eventual occupation after WW2. Anime itself was copied inspired by stuff like Disney, too, so if the barrier was as strong as some people think it is then anime wouldn't exist at all. Just think of something like panties or schoolgirl outfits and how important they are in perverted otaku circles, and yet they're entirely a Western creation.
Cultures and people finding ways to communicate is one of the more interesting things in history. I'd love to see how that stuff happened, or at least the nice parts, because bridging the gaps seems like it would require immense intellectual and emotional understanding. It's something we'd like to know in the far future to if we somehow surpass expectations and become a spacefaring species that seeks to communicate with others.

Authentic "cultural exchange" is a thing of the past, almost everything has an agenda to gain "soft power" and its depressing

>>1626
Like, it wouldn't even surprise me if the anime industry was funded by the US to make nerds more pro-Japan-SK-Taiwan.

>>1618
>>1620
Languages are not unlike living beings.
There are fuzzy boundaries in species and in languages, where there's compatibility with the ones immediately close, but which is lost the further you travel. Ring species and language continuums. Small changes take place over periods of time, especially when there are pressures present, and the isolation of a group leads to ever increasing divergence. It's impossible for any language, constructed or otherwise, to cover the full breadth of all constructions that have been documented, just like you can't have a trout that's a tyrannosaurus, an oak, and a cyanobacteria all at the same time.
Most importantly, there are many things to learn from human speech. If a tribe in northern Australia can develop thirty semantic word classes and expresses this through a series of affixes, what does that say about human cognition? What does it say about a human's information processing capabilities when a language can get away with using just 13 phonemes, while another one has hundreds? What can you learn from degrees of animacy, from conjugations that express the validity of what's being stated, from verbs without tense? It can't be overstated how important the quest to confirm universality is.

Both biological and linguistic diversity are on the downturn, and every second that passes it becomes harder to gain a deeper understanding of them. Forcefully imposing a universal tongue would be the same as wiping away the Amazon to make way for more farmland.

>>1625
What you are talking about are material aspects mostly and westernisation in general is a complex topic. Japan's first phase of westernisation occurred because it felt inferior and felt that it had to westernise in order to compete, the second phase was due to American rule and was actually really just a continuation of the first, only now their goal was capitalism not conquest.

One can adopt a material aspect of a culture without adopting that culture as a whole, people do that all the time, with things like Yoga, Taichi, karate, Karaoke, bubble tea, sarongs, etc.

You mention school girl outfits, Japanese school girl outfits are based on early 20th century European sailor outfits. The Japanese adopted that before the second world war, yet here they are still using it while nobody else actually does. Because they are an insular society, they do things for their own reasons. Anime is another example of that, yes it is inspired by Disney but it is it's own thing completely, a product of Japanese society and not of the society they borrowed it from.

In case anyone wants to a dork and read about cool grammar, here's some of what was mentioned before:
https://en.wikipedia.org/wiki/Evidentiality
https://en.wikipedia.org/wiki/Animacy
https://en.wikipedia.org/wiki/Noun_class
Hell, here's a paper on the evolution of grammatical gender in Late Proto-Indo-European, where Hittite had conserved the older animate-inanimate system while the rest of the family developed its characteristic male-female-neuter structure. Spoiler: female is the youngest one.
It's a very nice article, going over how semantics mix with sound changes in all sorts of ways to form new categories while comparing the phonetic basis of gender to the semantic one of stricter noun classes, and it's only thirty pages long. There are technical terms like "nominative" and such, but you can easily gloss over them. It's still easy to get the gist of it.
Yes, the formal acronym is indeed PIE, and yes, its words are romanized with numbers.

>>1630
It's probably outside the scope of a typological(?) approach but reading through that paper I can't help but wonder why there wasn't any consideration for the speculative culture or myths and folklore of PIE/IE and it's relation to gender development. From what I could understand most of what is being argued is that if a language has X feature it will tend to develop in Y way and it won't develop in Z way; with little regard for the particulars of how the language was used and the particular importance/context its peoples placed on words. I recon there could be a lot of value gained from dissecting what we know about the stories and culture of PIE/IE peoples and what they stressed importance on; then applying this in conjunction with the typological approach.

>>1631
Typology is just the formal classifications of this and that, when something has some wild grammar going on it's said to be "typologically unusual." But linguistics does delve all the time into how culture shapes language, as it's impossible to describe something like a system of honorifics without explaining how it fits into its society. Take pic, from this paper about the uses of 方:
https://sci-hub.st/https://doi.org/10.1515/jjl-2012-0106
Another example from Japanese is how just a few decades ago pronouncing /g/ as [ŋ] (compare "sin" [sɪn] vs "sing" [sɪŋ]) was seen as prestigious and cultivated, an upper class person wouldn't pronounce the word 影 as [kage], it'd be [kaŋe] instead. These things can seem silly, but they still make an impact and people pay attention to it.

However, grammatical gender is different. The French "mont" and Italian "monte" are male nouns, as was Latin "mons" back in the day, but in Spanish "montaña" is a female noun. This isn't due to Iberians starting to relate mountains to mother Gaia, as the goddess was considered heretical by the time this was developed, nor did gender norms shift in any way. It's simply because the word ends in -a and -a is how female nouns end. We mustn't forget that the gender transition took place in Late Proto-Indo-European, and we know that their characteristic culture and mythology were already present in the previous stage of Early PIE.

In the same vein, Romance languages (except for Romanian) lost the neuter gender of Latin due to sound changes eroding away its distinct ending -um/-us, but they also lost case endings at the same time for the same reason. Furthermore, French became the only one in the family to lose its ability to unconditionally drop the subject because its verb endings got deleted as well later on, making it so they were no longer able to specify the subject with conjugation alone, meaning it was now necessary to add at least a pronoun to explicitly state who you were referring to. In PIE, the opposite happened, sound changes made it possible to reinterpret words and go from two genders to three.

The point is that grammatical gender is often semantically incoherent, especially when something that doesn't have a biological sex is classified as male or female, which is what the majority of Romance nouns do. It's important to understand that syntax and semantics are not the same thing and don't always have a straightforward 1:1 correlation. The utility of its current form largely comes from letting you distinguish, disambiguate words and their modifiers through their sounds, through agreement, regardless of their meaning.

>Another remarkable difference between classifiers and genders is the much more arbitrary character of the latter. As Corbett (1991) points out, in virtually all gender systems there is a ‘semantic residue’, distributed among genders in an arbitrary way. Such residue, which, as noted by Dahl (2000), only concerns inanimates, does not normally exist in classifier systems: hence, the classifying function of classifiers is much more straightforward.

For a recap and summary, these two videos explain it well (~13m at x2):
https://youtu.be/3AnG3tbwlIw
https://youtu.be/7kV_bLrfUxs
The first one is flashier and the second one is drier, but both are good.

If it's not corpus linguistics I'm not interested.

>>1622
Horseshoe theory proves true once again!

>>1633
what
>>1634
what

in the butt

You know, I was looking up stuff about calligraphy and kinda sorta ended up wandering into the Chinese Wikipedia article, where I learned from an oddly short section that wikipedians were practically forced to develop an automatic converter for Chinese characters and vocabulary, because it was the only way to address a whole host of issues:
https://en.wikipedia.org/wiki/Chinese_Wikipedia#Automatic_conversion_between_traditional_and_simplified_Chinese_characters
You see, this is interesting because IT'S INSANE.

For a variety of reasons, China, Taiwan, Hong Kong, Korea, Vietnam, Japan, they each use their own sets of characters that have throughout time ended up diverging both visually and semantically, even sometimes creating new characters wholesale that don't exist in other regions. Even inside of the mainland there are so many regional variants that people travelling to a place they've never been to may not be able to guess an unknown character's meaning. Now, certain traditional-simpified pairs as well as transnational variants do have a 1:1 correlation with each other that allow for a single encoded character to be accurately rendered by different fonts in its different forms, but there are many, many cases where these variants are encoded as entirely different symbols because there are simply too many of them, with too many unpredictable differences, asymmetric alternations, and even diverging meanings, as well as being handled differently by different encoding standards. On top of that,
>A number of surveys, such as [Xiandai 1986], have demonstrated that the 2000 most frequent SC characters account for approximately 97% of all characters occurring in contemporary SC corpora. Of these, 238 simplified forms, or almost 12%, are polygraphic; that is, they map to two or more traditional forms.
>Some preliminary calculations based on our comprehensive Chinese lexical database, which currently contains approximately three million items, show that more than 20,000 of the approximately 97,000 most common SC word-units contain at least one polygraphic character, which leads to one-to-many SC-to-TC mappings. This represents an astounding 21%. A similar calculation for TC-to-SC mappings resulted in 3025, or about 3.5%, out of the approximately 87,000 most common TC word-units.
Following this, a wikipedian commented:
>This means that approximately 20000 individual entries are needed for a good conversion system.
Fonts are qualitatively several levels away from coming close to a solution.

BUT THAT'S NOT ALL
Not only do the characters themselves diverge, so do the words they're used in. In Japanese, 手紙 is a letter, but in Standard Chinese it's toilet paper. Written Chinese is actually just Mandarin/Standard Chinese being used as a lingua franca, but EVEN THEN native SC speakers/writers from different regions predictably refer to the same things using different words, to the point that you have to convert vocabulary as well to make a neutral Standard Chinese Wikipedia that's readable for all SC users. You can see the table there displaying the SIX different sets of vocab they accounted for: China, Taiwan, Hong Kong, Singapore, Malaysia, and Macau, while at at the same time matching them with the appropariate type of rune, because the mainland may use both simplified and traditional, but in Taiwan the former is shunned. (their name for the traditionals is 正体字 rumaooooo)

Again, a historically unparalleled clusterfuck, millennia in the making. Existential horror whose calligraphy is an art. Imagine this: you have Einstein's biography written in Jamaican English while Newton's is in Hinglish, and over there Caesar gets two articles, one in Glaswegian with phonemic orthography, and another one in Nigerian. Those last two may be on the same site, but they don't know the other one exists. That might BEGIN to give you an idea of how much of a mess it must've been.
So how does Wikipedia fix this? With conversion templates consulting manually maintained tables, that transform the article's text bit by bit into one whose everything is appropriate for the reader's locale. Here's the code's explanation:
https://meta.wikimedia.org/wiki/Automatic_conversion_between_simplified_and_traditional_Chinese
All in all, mindblowing. I can't believe they actually did this shit, Jesus, it's hard to grasp. The site really is insane.

For further madness, delving deeper into the abyss, see The Pitfalls and Complexities of Chinese to Chinese Conversion:
https://cjki.org/c2c/c2cbasis.htm
This article is just about traditional-simplified conversion, and the many problems raised by that alone. "Chinese to Chinese", jeez, pure lunacy.
God bless Jack Halpern, there is a hardly a man for whom Moksha is more deserved. He's even carrying out these same efforts for the Arabic family.

>>1696
gonna be a good post to read again in next years winter

>>1696
It's this sort of thing that makes me pessimistic about learning another language, because it's organic it can get so messy and this stuff is also ironed for the average, local speaker by cultural osmosis.

Also on the topic, Chinese is such a special case that two classifications are not sufficient, but I think there's political pressure to not to stratify the language.

I will say, aside from maybe Arabic, I can't think of a language where regional dialects are mutually unintelligible. Maghrebi Arabic has such strong influences from European and Berber languages that I think it could be considered a Pidgin.

Dialects of Croatian apparently are mutually unintelligible.

And this is ignoring the Pinyin elephant in the room...

And before you say British and American English, heavy accents in both are becoming uncommon amongst younger people.

>>1700
Internet speech(which mostly seems to have an American or Australian origin) is becoming much more common amongst not just younger people but some older people as well, internet people I will say. I think people often wrongly assume that they are all actually young, my uncle sounds like and probably is a 4channer or something very close to it and he is 50.

>>1701
Australians also specifically seem to be making an effort to preserve accents and slang, but the UK, US, Singapore and to a lesser extent India and Nigeria have negative stereotypes associated with heavy accents and also may have government pressure to speak proper English (Singapore recently, and the US in the 20th century made public schools English only).

>>1700
>>1701
In the future, net english will consist of 10 words

>>1701
SHOOP DA WOOP NEPHEWFAG IMMA CHARGIN MAH LAZER

>>1699
>stratify
As in, decentralize? Or how? A solid chunk of these issues stem from the imparity between traditional and simplified characters, whose differences exist regardless of the language using them.
Though, unintelligibility is actually pretty common throughout the Old World even today, that's the whole deal with language continuums. Look at these percentages from Japanese:
https://en.wikipedia.org/wiki/Japanese_language#Dialects_and_mutual_intelligibility
The listeners were from Keio, I think that's worth mentioning. Germany gets mentioned fairly often whenever this topic is brought up, too, as does Italy, but I think the situation in the second one is different. Would appreciate if a kraut or a guido could chime in.
A lot about intelligibility comes down to phonetics, being able to interiorize the correlations between two dialects' sound systems, like the rhotic vs non-rhotic split mentioned above that no one ever complains about because connecting the dots is pretty straightforward. But in cases where multiple changes pile up, that's when intelligibility takes a nosedive, that's when this happens:
>The 1998 film by Ken Loach, My Name is Joe, is one of the few films recorded [almost] entirely in Glasgow dialect. As a result, the film had to be given subtitles when released in the United States and even for audiences in England.
But Scottish English has a syntax and lexicon that is largely identical to that of other English dialects, all you need to easily understand someone like Limmy is to turn on subs, then comprehension becomes trivial.

Going back to chink, "Chinese to Chinese" is kind of ambiguous now that I think about it. To be clear, Sinitic is 100% a full-on language family, like Romance or Baltic, while Mandarin/Standard Chinese is a single language inside of that group. Cantonese is a separate language, for example. (Itself in reality representing a group inside of which further unintelligibility exists, but that's life.) These sets of people, if they spoke in SC, would be perfectly able to understand each other. The issue lies, above all, in the runes. They're what throws a dragon-sized wrench into this melange of ink, paper, and bamboo.
As for Arabic, its status is debated afaik. I lean towards it being a family as well since ya gotta have diverged after a over thousand flippin years have passed, c'mon, but it's not something I've looked into.

Whereupon forenners and strangers do wonder at vs, both for the vncertaintie in our writing, and the inconstancie in our letters.
~ Richard Mulcaster's Elementarie, 1582

Let's revisit the mess that is English once more because I'm certain you haven't heard of this one. Please bear with me.


What's known and well understood is that its orthography, rather than being a single coherent system, is a set of sets. The Germanic <ch> of church, French chef, and Greek chaos are all pronounced differently not because of *English* sound changes but because they're using different orthographical standards all grouped under a single umbrella. If this were the extent of its irregularity it'd be a piece of work still, but nothing anywhere as crazy as what we have nowadays. The real kicker is when they attempted to generalize the irregularities in ahistorical cases, and then ITSELF APPLIED THIS IRREGULARLY.

Take all those words ending in -mb where the <b> is mute because it got dropped centuries ago, usually after its subsequent vowels were sheared off as in bomb (bombus) /bɒm/ and dumb (dumbaz) /dʌm/, their silence nowadays represented only informally as in the derived dummy. The problem is that this spelling was extended to words ending in -m where THE SOUND NEVER EVEN EXISTED, such as thumb (thūma) and limb (limu). There's also might and night, that had an h-sound in the middle like in German's current macht, nacht, which in English was dropped and the vowel lengthened in compensation as it often happens throughout history. Once again, scribes tried to apply this pattern to words that never had the ending -ght, leading to cases like dēlectāre -> deleiter -> delite -> delight(???), having nothing to do with the Germanic word light! "Despight" too briefly existed in place of despite.
During the second phase of importing French words, post-Norman 1300-1500, here too was the recent continental orthography adopted and haphazardly generalized, such as using <c> to represent the sound /s/ (centre, city) which was once more extended to Germanic vocabulary: īs->ice, nys->nice and the asymmetrical pair of mouse-mice from mūs-mȳs. Just one of way too many changes from that era.

Even etymologically appropriate re-spellings have caused a lot of confusion: doubt, debt, subtle, receipt, indict, in every one of these the consonant before -t was dropped too, and they're all mute today, you have cupboard instead of the assimilated cubborde, cobbarde, cobbourd, cobord, cubbard, cubbarde, and so on and so forth, the fake -l- in the middle of Italian colonel which actually came through French co-R-onnel and is pronounced as such, and yet another is the /l/s that got dropped, see walk, calm, half, should, all mute, cosmetically re-added to French saumon because back in Latin it was pronounced salmō(nem). But in the case of falcon, this restored /l/ came back and became the new spoken standard! It used to be faucon, the vowel was changed as well. Same thing happened to fault (French faute). In the process, Germanic island (iland) got mixed up with the unrelated Latin insula and now has a mute <s>, like isle's insula->ille->ile->isle.

See, the printing press was introduced to England in the final decades of the 1400s, at about 1476, thus in the 1500s was the anarchy unleashed, that which Mulcaster attempted to fight in 1582, as did another reformer, Bullokar, in 1580. At this time the Great Vowel Shift was still ongoing, you can see in the attached chart how it was in the 1600-1700 period where the major mergers of the epoch took place. Of course, this has all been in the works since 1066 when the Normans came to power, spiking later in the mid 1400s' FUCKING CHANCERY STANDARD establishing ouiaboo respellings.

The GVS is the reason behind another major part of our calamitous recipe, the fact that a notoriously large number of ordinary homophones have vowels differentiated solely in writing, since only a handful of dialects retain a handful of old distinctions each. Kruse writes that "it has been estimated that a staggering 30 per cent of monosyllables belonging to the lexical set FLEECE have a differently spelled homophone," some of these being beach–beech, beat–beet, breach–breech, cheap–cheep, creak–creek, dear–deer, heal–heel, lea–lee, leak–leek, meat–meet, peace–piece, peak–peek, peal–peel, read–reed, sea–see, seam–seem, steal–steel, team–teem and weak–week. This of course applies to other vowels too, with FACE we have rain–reign, slay–sleigh, strait–straight and wait–weight, GOAT roan–grown, moan–mown, nose–knows, road–rowed, sole–soul, throne–thrown, toe–tow, NORTH/FORCE border–boarder, for–four, horse–hoarse, morning–mourning, or–ore-oar, war–wore and warn–worn, that's 75 words right there. In non-rhotic English the loss of final /r/ led to even more of them: caught–court, cause–cores, cawed–chord/cord, awe–or/ore/oar (again), flaw–floor, fought–fort, gnaw–nor, laud–lord, law–lore, maw–more, pause–pores, paw–pore, raw–roar, sauce–source, saw–soar/sore, Shaw–shore, bawd–board, caulk–cork, sought–sort, stalk–stork and thaw–Thor, or alms–arms, father–farther, lava–larva, ma–mar, pa–par and spa–spar.
Lotta words, innit?

So far we've had fossilized spellings alongside respellings that looked towards the past to represent sounds that weren't there anymore, or that changed one standard for another. What we haven't seen is ones that proactively, consciously changed them in order to differentiate words that had become homophonous. For this, I need to bring up one final merger: whine-wine. The cluster <wh> used to have an h-sound, as is commonly known today through the two ways one hears "what," mainstream wat and archaic hwat. It also merged which-witch, wail-whale, weather-whether, weigh-whey, and several others.
Here's the kicker: in Old English you had the words hal and hol, eventually both came to be pronounced as /hoʊl/. They were spelled with <h> simply because that's the sound they had. But wait, what if we added a mute consonant to disambiguate them? What if we wrote them as... hole, and whole? The same happened with the pair of hoar and whore, where the <w> is a total fabrication spread on purpose by those scribes I despise so much. They even tried to write holy as wholy, contrasted with wholly, neither of which ever had a /w/. (The latter is doubly irregular: it uses repeated <ll> after a long vowel.)

With a millennium of fuckery established, we can reach the punchline: Thomas Wilson's 1707 Manx transcription.
Now, Manx Gaelic did not have a notable written tradition, and John Phillips' older 1610 translation of the Book of Common Prayer never managed to gain any traction to set an orthographical standard. Wilson, as bishop of Sodor and Man, was brought to the island to fix its state of disrepair and seemingly acted as virtuously as one can, he sounds like a genuinely great guy. By no means a stupid man, he learned the language and wrote a bilingual text in his Principles and Duties of Christianity.

What he did that was so astonishing is extend this practice they'd done with proactive h-wh before, arbitrarily picking Gaelic homophones and choosing to add some Hs or other letters to certain words, as you can see in the attached examples. He could've done the same as missionaries across America, faithfully represent its sounds first and foremost, but he chose not to. Sproat explains that:
>So the spellings themselves are not English, but what is like English is the idea that words that mean different things ought to be spelled differently even if they sound the same. Or to take another set of words that have the same or similar pronunciations, but different spelling: leigh ‘law’, leih ‘forgive’, lheiy ‘calf’ and lhiy ‘colt’.
This led him to conclude that:
>Manx orthography need not have been unpredictable. But if one views unpredictable spellings as at the heart of English orthography, then my host’s view is eminently sensible: basing an orthography for Manx on English would almost be guaranteed to yield a system that is unpredictable in how it chooses to spell words.
>Put another way, anyone familiar with English orthography and using that as a basis for a new spelling system, would not feel compelled to be consistent in how they spell given sounds; in contrast they would feel inclined to invent different ways to spell homophonic words, just to keep them distinct in writing.
Emphasis is mine, because the whole thing is simply insane. If you've managed to reach this point, I hope it entertained you.


This was taken from several chapters of The Routledge Handbook of the English Writing System, where the authors contemplate questions such as "to what degree can we say this bastardized script reaches ideogrammic levels?" It's a very fun read, and you can find the full exegesis of its influences in chapter 7, The etyomological inputs into English spelling.

>>2764
>The problem is that this spelling was extended to words ending in -m where THE SOUND NEVER EVEN EXISTED
Ehh, consistency is good. I think that's probably a good thing, although if English was actually consistent it would mean more.

>colonel
I hate this word for how nonsensical the spelling is compared to how it's said. Kernel.

>in contrast they would feel inclined to invent different ways to spell homophonic words, just to keep them distinct in writing
That's kind of funny. It's like furry OCs or something.

>>2767
>if English was actually consistent it would mean more.
Exactly. The problem is that it neither makes sense etymologically nor was it applied consistently, so you end up having /-aɪt/ represented as both -ight and -ite, and really it's the former that's more problematic given that it's not as general as the silent e and gets confused with sought, cough, or through which word differently.
>It's like furry OCs or something.
It's also very Sino-Xenic.

>>1626
It's the opposite.
Back in the day, people had to actually travel for long distances to have cultural exchanges.
Given that travel is a massive filter in a world without cars and lots of warfare, that means that the people who are traveling far are, on average, rich and powerful, or emissaries of the rich and powerful.
Most societies being class-based, the travelers would then usually deal with others of their own kind (rich and powerful, or emissaries of the rich and powerful).
During such meetings, gifts and performances were made to impress the other side.
For the most part, cultural exchange was driven by specific political aims.
This is also why specific parts of cities (government districts and the main paths to reach them) are built in such imposing fashion. That's not about the locals. Not even the local elites. It's about important visitors who are meant to be impressed.

In contrast, nowadays, you can just go on the internet and listen to music from the other side of the world, written for an audience of the other side of the world. If you want more than a digital experience, modern people can afford to just take a few weeks off to visit foreign countries. They may employ travel agencies who plan the trip for them, but there are also many that choose to directly interact with the local culture.

>>2766
>anyone familiar with English orthography and using that as a basis for a new spelling system, would not feel compelled to be consistent in how they spell given sounds; in contrast they would feel inclined to invent different ways to spell homophonic words
Beautiful. I love it.

>>2775
That's an interesting and very good point, although I think this part may be a bit of a stretch:
>That's not about the locals. Not even the local elites.
Looking at something like the pyramids, which I understand were built for reasons internal and specific to Egypt, it seems to me it's perfectly possible to pour that much time and effort into monumental things that either aren't focused on the exterior, or are supposed to marvel everyone all around. To display power, to prove something. I can't see why this wouldn't apply to constructions by the locals and for the locals, like great temples and palaces and the like.
>>2776
Happy to hear that. It's actually rearranged, the last part on Manx is discussed in chapter 3 as part of the nature of the script overall, the matter of standards and their history begins afterwards. But it makes for such a great conclusion, I love it too.

Oh, >>1697, did you remember?

>>2775
I would not say that's necessarily true. We overlook just how much poor people traveled. It was fairly common for the poor to go on pilgrimages over very long distances, plus the poor were the majority of the army as well, so any campaign that was sent to a far away land would see a lot of poor people travel(added to that you have the camp followers who were also poor).

If you just imitate other speakers it just werks

>>2782
>>2783
I would have to backpaddle a lot to salvage my argument(s).
You win.

Before /win/ closes down again, I also want to share an even less intuitive feature of English: homographs that are pronounced differently based on word class.


For example, verbs in English regularly take the -ed ending for their past tense and participle conjugations, but there are also several words where an identical ending was historically used to create adjectives. For whatever reason the latter remain fully enunciated while the former has been simplified, and due to even more historical happenstance there now exist several words that are written the same but pronounced differently depending on what role it is fulfilling:
Cursed = cursid, or curs't
Blessed = blessid, or bless't
Other pairs are learned, crooked, but it seems winged, supposed, and peaked have been neutralized. Something weird's going on with alleged, it seems that one's actually a new development by analogy. Then there's the similar naked, ragged, rugged, wicked, jagged, legged which I'm not confident in grouping here, and the cases of beloved and sacred where the original verb is obsolete and only the adjective remains. This still overlaps with common participles being used as adjectives, like bored from bore, which has a regular pronunciation.

Then there's <-ate>, which can be either /eɪt/ or /ət/. The verbs it's present in always have a long vowel, while with adjectives and nouns it's a tossup, no way to tell as far as I know. Richard Eaton, some 80 years old historian I just learned of, says Persianate with a schwa, while Latinate I've seen listed mainly with FACE even though you'd expect them to sound the same given how similar they are. (A younger speaker does use FACE for both.) It's possible /eɪt/ is becoming more common over time and as the spelling that makes sense, and which regularly applies to all chemicals bearing it.
Here some pairs you can find are graduate, incarnate, duplicate, correlate, (in)discriminate, elaborate, laminate, deliberate, federate, initiate, intermediate, moderate, and syndicate. Apostate appears to be pronounced both ways, maybe, but I don't know if that's tied to word class. I'm sure there are more though.

The most interesting one are those that are actually derived from each other. English has a method to derive nounds and adjectives from verbs, which consist of moving stress towards the start of the word. Take this set of original verbs: disCOUNT, inSULT, and imPORT, whose stress-derived nouns are DIScount, INsult, and IMport. In fact, the shifted DIScount is an adjective, too. It's called a SUPRAFIX 'cause it's suprasegmental i.e. it's not a specific phoneme but rather something that goes on top, though due to anglo phonetics shifting stress does cause the phonemes to change but the stress being moved comes first and that's what matters!
It's the most populous group, and productive too it would seem, here's a list of them:
https://en.wikipedia.org/wiki/Initial-stress-derived_noun#List

Oh, and for some reason "mouth" the verb isn't written as <mouthe> even thought it's /maʊð/, like loathe, bathe, lathe, seethe, lithe, writhe, etc., so that one's another weird homograph. Might have to do with /aʊð/ being an exceedingly rare sequence.


On the flipside, you have a peculiar ending that's pronounced the same but written differently: -tre/-ter, and -bre/ber. I'm not sure what the fuck happened here exactly, looks like something to do with the unstressed /r/ there dangling at the end being deleted off the map or some shit, which in rhotic dialects resulted in a schwar /ɚ/ that sounds exactly the same as <-er>, also /ɚ/. It must've merged or something, both ways of writing it were common across the Pacific before dictionaries fixed them in place. You can see it respelled in entre->enter, but retained in entry, entrance. Also compare tiger with tigris, monster with monstruous, and member with membrane. Anyways, there's centre/center from centrum, fibre/fiber from fibra, and sombre/somber from... sombre, sub-umbra.

Others are spelled with <-er> across all orthographies, like the first group of comparisons above, or chapter, fever, September, October. Inversely, ogre. The respelling was spared in <-cre> because <-cer> would make it sound like /sə/ rather than /kə/, as in acre, massacre, and mediocre. This may all stem from French bullshit, their final <-e> is mute and if you look at words ending in /-tʁ/ like mètre, théâtre, litre, lettre, the loaning languages add vowels where they feel like it, if at all. Seems like a huge chunk of <-re> endings come straight from frogs, whereas original <-er> in root words like anger and water, or any suffixing of comparatives, agents, and the like, are of Germanic origin.

>>2871
>This may all stem from French bullshit
That's what I always thought at least.

>>2872
I mean, yeah, when in doubt blame the French. It's a pretty reliable strategy.

How do Koreans deal with all the homophones they inherited from China anyways?

>>2879
In regular spoken language, I don't think it's much of a problem. As far as I know, the reason why Chinese went from this mythical state where so many words were monosyllabic to the current situation of most being dysyllabic compounds is because simplification made so many morphemes sound like each other, and the Chinese have to put up with it same as those who imported their vocabulary. Take the jouyou kanji: it includes like fifty characters with the reading shô, and you regularly encounter words like 少女, 勝負, 将来, 賞金, 正直, 証明, 症状, 昭和, 紹介, 召喚, 省略, 障害, every one of them starts with exactly the same しょう sound, but every time is stands for a different morpheme, each one means something different. What makes them distinguishable is that people learn them as full words, spoken shôjo is as distinct from shônen as it is from shôji, and that's why people recommend that you learn words rather than memorizing individual characters. Although the vast majority of characters are homophonous with each other, the same is not true for full compounds in daily use.

What Koreans use when it comes to technical written vocabulary is hanja (same word as kanji and hanxi, 漢字), they revert to using runes. The relationship that Chinese characters have with spoken language is a particularly strange one, and simply continuing to employ them when truly needed is the method that stuck.