How should I define the lexicographic order for my conlang?

Question

Though I've had a tough time choosing the phonology and the orthography of my conlang, I think I've finally confirmed it: 12 consonants, 10 vowels, Cyrillic script.

The consonants are:

• Пп for [p]
• Тт for [t]
• Цц for [t͜s]
• Чч for [ʈ͡ʂ], or [t͜ɕ] when iotified
• Кк for [k]
• Ҁҁ (koppa) for [ʡ]
• Фф for [f]
• Ѳѳ (fita) for [θ]
• Сс for [s]
• Шш for [ʂ], or [ɕ] when iotified
• Мм for [m]
• Нн for [n]

The vowels are:

• Аа for [a] / Яя for [ja]
• Ээ for [e] / Ее for [je]
• Ыы for [ɯ] / Ии for [i]
• Оо for [ɔ] / Ёё for [jɔ]
• Уу for [u] / Юю for [ju]

But how should the words of my conlang be ordered on a dictionary? It seems that the best solution is to follow the Russian lexicography, while inserting the missing letters owing to their Phoenician origins. That would give the lexicographical order of АЕЁИѲКМНОПСТУФҀЦЧШЫЭЮЯ. Or is there a better option?

(Sorry for bad tagging; there is no "lexicography" tag.)

Answer 1

People aren't especially consistent about this with natlangs; it tends to depend on the target audience. For example, Akkadian words are conventionally sorted by English alphabetical order, with Š and Ṣ inserted after S. This is convenient for English-speakers, who will see a letter that looks like an S, and find it in the S-esque section of the dictionary. Hebrew-speakers, who are used to separate letters for S, Š, and Ṣ, would probably prefer to put S (samekh) after N, Ṣ (tzade) after P, and Š (shin) after Q. But the majority of people consulting an Akkadian-to-English dictionary are probably not familiar with Hebrew alphabetizing conventions.

In-universe, the speakers of your conlang can get used to any lexicographic order. They tend to be pretty much arbitrary, after all, and children memorize them without issue. So I would think about out-of-universe convenience. Who is going to be consulting this dictionary? What ordering will they find most intuitive? The target audience is probably you more than anyone else (you're going to be adding to this dictionary, after all), so if referencing Phoenician alphabetical order works for you, I'd go with that.

Answer 2

I think your proposal sounds well-considered--especially if you want your computer to sort your vocabulary lists in alphabetical order rather than you having to do it all by hand.

If I were you, I'd (a) write your alphabet with each letter inserted into the cell of a single column table, (b) have the computer sort the items in the table, and (c) adopt the results of the sorting as alphabetical order.

Answer 3

For the Western alphabets (including Cyrillic), the lexicographic order is essentially arbitrary. If there was once a reason behind the original order of the Semitic alphabet, it is long forgotten and buried in the dust of time.

There are some scripts with a more rationalised lexicographic order, Indic scripts like Devanagari separate vowels from consonants and order the consonants by place of articulation. The conscript Tengwar by JRR Tolkien has adopted this approach. Also, the Universal Alphabet by Francis Lodwick followed this approach.

The decision is essentially yours: Reasons for some preference can be convenience of usage for the potential users of your conlang, æsthetic criteria like elegance of the sorting order or grouping similar shapes of the letters together, or linguistic criteria as in Devanagari.

Answer 4

It's ultimately an arbitrary choice. But, I assume that you want a "naturalistic" order to make it look like your version of the Cyrillic alphabet evolved in parallel with Russian, Ukrainian, Belarusian, etc., I would make one change: Move Ҁ (Q) to between П (P) and C (S), to match its position in the Phonecian, archaic Greek (Ϙ), and Latin alphabets.

АЕЁИѲКМНОПҀСТУФЦЧШЫЭЮЯ

An alternative ordering can be obtained by treating your alphabet as a more direct descendant of the Phonecian and Greek alphabets, treating Ц and Ч as counterparts to Tsadi/San (between P and Q), and Ш as Shin/Sigma. This gives:

АЕЁИѲКМНОПЦЧҀСШТУФЫЭЮЯ

And then there's the lazy programmer's approach of just using the Unicode code point order:

ЁАЕИКМНОПСТУФЦЧШЫЭЮЯѲҀ

Answer 5

dan04's answer gives several good options, but I will suggest some others, which are more likely if the letter was re-adopted after having fallen out-of-use. When a new letter is added to an existing alphabet, there seems to be a strong tendency to insert it at the end. This happened with the latin "Z" which was inserted at the end of the alphabet even though its Greek counterpart is towards the beginning of the alphabet. That would give you:

АЕЁИѲКМНОПСТУФЦЧШЫЭЮЯҀ

Another option is to put it with a letter that sounds similar, probably К:

АЕЁИѲКҀМНОПСТУФЦЧШЫЭЮЯ

Finally, you can put it with a letter that looks similar, like С (this is how the Latin letter G ended up coming after F).

АЕЁИѲКМНОПСҀТУФЦЧШЫЭЮЯ