Ionioka embȳr?

Logographic systems work quite a bit differently from most ‘phonetic’ writing systems. Your inventory of signs will be much larger than other systems. Think a couple of hundred. Tepatic glyphs ended up with about 200 distinct signs, which is appoximately the number of Chinese radicals.

However, it also works similarly to ‘phonetic’ systems, or more similarly than you might expect. Some of the stuff you need to do will be the same as for any writing system / conlang - for example, consider the writing medium (paper, stone etc.) and tools (brush, etc.) and how they affect your system. Also, know your conlang’s phonology and basic grammatical structure – logographic or not, this will be important (as you’ll see).

First, I would like to talk about what logographs are, and how they work. Since you, Anon, asked the question, I’m guessing you might already know what logographs are, hence you are asking, but in case you don’t, I want to go into what logographs are, because I think many people don’t quite get what they are, and unless they are Chinese (or Japanese), don’t know how they work, and this is going to cause problems in making the script work if they try to make logographs for a conlang. In case you already understand logographs well (or you’re Chinese) you can skip over what I’m writing. Or read it anyway, if you want to hear about how other systems work. I advise you reading up a lot about how other systems work.

Earth has at least four families of logographic writing systems: Egyptian hieroglyphs, Sumerian cuneiform, Chinese characters, and Maya glyphs. They’re all very different, but have some similar principles, and one of them is they have a phonetic component, just like any other script. Also, logographs (also called logograms) are not ideographs or pictographs (‘picture writing’).

Many people think that hieroglyphs or Chinese somehow represent thoughts directly through some kind of picture, bypassing “words” and spoken language, but they don’t. This is no doubt exacerbated by the fact most hieroglyphs undeniably LOOK like ‘things.’ So it’s natural to think that an obvious picture of an owl must represent the idea of an owl, but often it doesn’t. Confusion about this will make you not get how to make your system work especially when expressing something abstract that you can’t draw. Logographic systems are intimately tied to the languages they are used to write. Hence, yours will be tied to your conlang. You should already have a good idea of your conlang. It doesn’t need to be complete (when are they ever?) but you should know the phonology, and some other parts of the grammar may be helpful (is it very inflectional, word order, etc.)

Logographic systems contain signs that can represent whole words (or roots), which may or may not be recognizable ‘pictures.’ They also have phonetic elements (which may also look like pictures). The way the phonetic elements develop is usually by what we call the rebus principle, which means a sign is borrowed to write something else which sounds similar.

I’d like to illustrate this with an example. I wrote the following English sentence in pictures:

If you’re having trouble, try saying the name of each thing depicted. You should get:

eye can knot bear two leaf ewe.

Now try saying it faster and see if you get it.

What I’m trying to say is, “I can not bear to leave you.”

This is something I still didn’t quite get when I started working on Yuk Tepat.

You will note, that none of these little pictures are actually pictures of the concept referred to by the word they are used to spell. They are pictures of the thing whose word sounds like the word for the thing you want to represent, but can’t easily, so you get at it by using something that sounds similarly. (I know that sentence is hard to parse, please read it again.) This example only works because “I” and “eye” sound the same in English.

Now, imagine for a minute that English-speaking America is pretty much the same as in our world but English is written logographically. Nuclear war or something happens and the country gets buried under mounds of ash and dust and dug up hundreds of years later by archaeologists for some human civilization from elsewhere, unrelated to America, and their texts are deciphered. In this future civilization’s version of Youtube, people start making videos explaining their mysterious, cool writing system.

“The ancient American sages represented the word for oneself, I, as a picture of an eye. And if you think about it, the self is a perceiving self; without a subject able to look around and observe an outside, objective world, there is no self, no ‘I.’ The self begins with a perception of the world. This is the profound and mystical wisdom of the long-lost ancient civilization of The United States of America. Every letter hides a profound truth about the world.”

The simple, possibly boring explanation is just that ‘I’ and ‘eye’ sound the same in English, so they could be written the same way. But laugh as you might, there are tons of people online (invariably non-Chinese) making similar fluff about Chinese characters. Beware.

People may think to write ‘be’ logographically you have to think up some cool symbol of the concept of ‘being’ but the likely solution for an English logography would simply be to draw a bee (the insect that makes honey) and have it represents the copula.

I think what a lot of people don’t realize is that it is very common (even usual?) for the word the sign is borrowed for to completely take over that sign, and that a semantic indicator will be added to indicate the ORIGINAL meaning of that sign, or a new character created for it. The sign will completely shift over; it is no longer a picture-drawing of its meaning, even in a highly stylized way.

The typical development often goes:

- A picture is drawn to represent the word for the same thing that the picture represents. This is still pre-writing; you cannot draw a picture of everything you might want to talk about.

- The picture is borrowed to write another word or syllable in the language which cannot be depicted, but which sounds like the word. This is going to be different for each language because each language has different words and different words will be homonyms; hence you cannot separate a logographic script from the spoken language it was made for.

- Some of these signs develop into phonetic signs.

- The system reintroduces ‘pictures’ (or their relics) to clarify the meaning of words which might phonetically be spelled similarly.

This has happened many times in Chinese characters. It is probably the main way Chinese abstract, grammatical words get characters. The character 水 is ‘water,’ from an image of 3 parallel water flows. 永, representing probably a person floating on water, meant ‘swim’ (Old Chinese *ɢʷraŋs). This was then used to write the similar-sounding *ɢʷraŋʔ ‘forever,’ an idea which is hard to depict visually, don’t you think? ‘Forever’ completely took over 永. Then when they wanted to actually depict ‘swim,’ they added three water dots (itself ultimately a form of 水) to the left side to make a new character 泳. This is essentially ‘that ɢʷraŋ-ish sounding word that has to do with water.’

Signs like the three dots are called a radical in Chinese, and the remainder is called the phonetic. Some radicals are extremely common and form a great classificatory system, marking characters containing them as glyphs referring to words for people, animals, birds, plants, liquids, earth, etc. The vast majority of Chinese characters are of this type: composite characters made up of a part indicating roughly the sound, and a part assigning the meaning to some semantic domain.

Similar signs to radicals are found in hieroglyphs and cuneiform, the main difference being that they are called determinatives and they are still written as separate signs rather than being compressed into a single character like Chinese.

(I will note, radicals are basically a lexicographic tool for indexing characters, and not an etymological classification. But the radical usual corresponds to the semantic component of a character.)

If you understand this, you understand one of the key principles behind logographic writing and are ready to proceed.

So, logographic scripts generally have three components:

(1) Logograms (logographs) – signs that represent a word (or root) in the language

(2) Phonograms – signs that indicate sounds

(3) Semantograms / Determinatives – signs that mark

The way this works out in particular scripts varies. Not all of them will have all of the parts. I think Maya doesn’t really have a system of semantic determiners. And it may very well be that these may be functions and a language may use the same sign in each of these functions – but perhaps with a system for indicating when it is working as a logogram, a phonogram, or a taxogram. In Chinese and Maya, signs are generally compressed into a ‘block’ to indicate they are functioning together – as we saw in the Chinese example. In Egyptian, a particular sign may perform any of the functions, but when it is used as a logogram, a small vertical mark is added to it to indicate it is a logogram.

Systems can get vastly more complex: Old Persian or Japanese. Japanese is probably the most complicated writing system ever – certainly the most complicated in use now. But I will set that aside, and notably both of those systems are the result of existing logographic systems being repurposed to suit completely unrelated languages.

You can tweak this overall system a bit. In Yuk Tepat, I have incorporated a class of marks I have variously called ‘operators’ or ‘modifiers,’ which are attached to and modify the meanings of logograms. For example, adding various different marks to the sign ‘hand’ forms signs for finger, knuckle, thumb, palm, have, lack, etc. You see this sporadically in Chinese, for example the characters 刀 ‘knife’ and 刃 ‘blade,’ or 木 ‘tree,’ 末 ‘branch,’ and 本 ‘root.’ In Chinese though, this is a sporadic thing, while in Yuk Tepat this kind of marking is very systematized.

Since logographic systems are still at least partly phonetic, the phonetic / phonemic part can be classified by the different ways it works, same as phonemic writing systems. Hence, you can have logo-alphabets, logoabugidas, and logosyllabaries. Egyptian hieroglyphs are logoconsonantal (an abjad). The other systems are syllabic. While I don’t know any logoabugidas, you can always make one. Think about which one suits your language. Egyptian, e.g., is logoconsonantal probably because it’s Afroasiatic and abjads seem well-suited to their style of consonantal roots. The only difference between these and fully phonetic writing systems is that the phonetic part of logographic systems usually has much less phonetic precision, hence why semantic signs are needed to disambiguate similar-sounding words.

Your system doesn’t have to work just like any of these. You can figure out your own type, although it will likely mix logograms, phonograms, and taxograms in some form. I’ll grant that, of course, you don’t have to do even that. If you want something ideographic, you can try to make it. People constantly imagine fantasy languages that work completely differently from any real language, and that’s their prerogative. But if by ‘logographic,’ you meant a system like Egyptian/Sumerian/Chinese/Maya then it will be somewhat like I describe.

So to wrap up, logographs =/= pictures, they don’t represent translinguistic ‘ideas,’ they have a phonetic component that reflects the spoken language, and use semantic signs to disambiguate words. Once you know this, you basically understand the mechanics of logographs.

With that, I’m going to stop for a bit, then get back with Part II where I will get to some more particular points about making a script.

Some languages do have words that refer to it directly, with terms like Asturian mexaderu, "urination place," or Navajo chąąʼ báhooghan, "house for feces." Sometimes terms like these are considered vulgar, more like English shitter, but others seem to be completely acceptable.

One common strategy is to refer to washing or cleaning, like English washroom or Arabic مرحاض (mirḥāḍ), "cleaning-tool." Some include specific types of washing such as taking a bath, like English bathroom or Indonesian kamar mandi "bathing room" - however, a lot of these words are used specifically for rooms with a bathtub and/or shower, not for public restrooms. Japanese 手洗い (tearai) and Persian دَسْتْشویی (dastšuyi) both mean "handwashing," and Kazakh дәретхана (däretxana) means "establishment for Islamic ritual washing."

Similarly, some words refer to the cleanliness of the room itself, like Vietnamese nhà vệ sinh, "sanitation house," or Urdu صحت خانہ (sihat xāna), "health place." My favorite of these is Spanish and Tagalog inodoro, "odorless," because yeah I sure hope it is!

Some, like Cornish attesva "comfortable place," Japanese 便所 (benjo) "place of ease," and Burmese အိမ်သာ (imsa) "pleasant house" seem to come from the fact that visiting it makes one less uncomfortable - indeed, in Philippine English, the term comfort room is common. This is probably where restroom fits, along with Arabic مستراح (mustarāḥ), "relaxing place."

Similarly, it can be referred to by the fact that going there is sometimes necessary. Serbo-Croatian nužnik means "needed place," Latin necessarium means "necessary place," Tajik ҳоҷатхона (hojatxona) means "need house," and the necessary is occasionally used in English.

Another activity that gets referred to as a substitute is getting dressed, which is where we get English toilet. Armenian զուգարան (zugaran) means "decorating place," Ukrainian вбиральня (vbyrálʹnja) means "dressing place," and Korean 화장실 (hwajangsil) means "makeup room."

A strategy that seems to be common in Southeast Asia is to call it the "water room," like Lao ຫ້ອງນ້ຳ (hǭng nam) or Malay bilik air. Wiktionary proposes that some of these may be calques of English water closet, but there's enough that it seems more likely to me that they calqued it from each other.

Some other references to activities in the room that seem worth mentioning include Bengali পায়খানা (paẏkhana), "footrest site," Saanich EMETÁU¸TW̱, "sitting house," and Georgian საპირფარეშო (saṗirparešo), "place where servants are found."

Another way to describe it is as a place that one goes away to or excuses oneself to, like Finnish käymälä, "a place one visits," Spanish excusado, "excused," or Polish wychodek, "little exit," which seems slightly more common in Europe, but there's also the Alaskan language Lower Tanana, with yo'ena' ninotr'ededekde, "where we go back out beyond."

Similarly, it's also referred to by its location outside or behind a house, like Tagalog palikuran "going to the back," Latin forica "outside," Ancient Greek ἀφεδρών (aphedrṓn), "away seat," and Korean 뒷간 (dwitgan), "back room." I learned a new English word from this one - reredorter, a latrine behind a medieval monastery or convent, literally "rear of the dormitory." Outhouse is also similar, and while it refers to a specific type of restroom, I can easily imagine it broadening.

Spanish común, "common, shared" might be related to the description of it being outside, or perhaps to public restrooms; Wiktionary simply presents it without explanation. Interesting to compare to English privy and Brazilian Portuguese privada "private," which focus on the exact opposite property.

Another physical property that's often use as a descriptor is its size. Welsh tŷ bach means "small house," informal French petit coin means "little corner," Indonesian kamar kecil means "small room," and Tok Pisin smolhaus and liklik haus both mean "little house."

Of course, one other strategy is loanwords. English influence around the world has resulted in variations on toilet and water closet or WC being very common. Arabic خلاء (ḵalāʔ) "empty, void" has spread a bit as well. Russian нужник (núžnik), "needed place" and Classical Persian حاجتخانه (hājatxāna) "need house" are surprisingly parallel, with similar meanings and some spread as a loanword, but neither is really used in the language they originally came from.

Loans of water closet often get clipped down, resulting in things like French waters and German Klosett, which have definitely caused some confusing situations! Another one that's definitely confused people is Thai ส้วม (sûuam), which as a result of clipping and/or lots of avoidance literally just means "room," and the cognates in its sister languages still mean "room" or "bedroom."

Finally, a few weird ones.

English ladies' room and men's room are a relatively uncommon strategy, but Finnish has naistenhuone and miestenhuone, with the exact same meanings.

The exact etymology of Estonian peldik is unknown, but there's a theory that it's related to Germanic spelt and means "spelt barn," because they were sometimes used as restrooms (this seems like a stretch to me).

Azerbaijani ayaqyolu apparently means "leg path." No other explanation is given on Wiktionary.

Persian تشناب (tašnāb) appears to mean "thirsty water."

Polish łazienka means "little bathhouse."

Vietnamese cầu means "bridge," and Malay jamban may be a contraction of jambatan, also meaning "bridge." This is thought to be because latrines were built over water, similar to bridges.