In the rest of this article, the following notation is used:

  • C refers to a standard consonant.
  • V refers to a standard vowel.
  • Cc refers to a geminated/long consonant.
  • Vv refers to a long vowel.

Consonants

Labial Alveolar Post-alveolar Palatal Velar Glottal
Nasal _ m ⟨m⟩ _ n ⟨n⟩ _ (ɲ) _ ŋ ⟨q⟩
Plosive p b ⟨p b⟩ t d ⟨t d⟩ k g ⟨k g⟩ ʔ _ ⟨'⟩
Affricate (tʃ dʒ)
Fricative s z ⟨s z⟩ (ʃ ʒ) h _ ⟨h⟩
Approximant _ (w) _ l ⟨l⟩ _ (j)
Tap _ ɾ ⟨r⟩
Trill _ r ⟨rr⟩

Hot take: The "whatever rhotic" is highly problematic and should not be used. It is so awful and I cannot believe people still permit it. Don't you realize how many rhotics there are? This is why I enforce rhotics in Sekko to only be the alveolar tap and alveolar trill. No other rhotics are permitted.

Thoughts: I might permit the palatal and retroflex fricatives to be allophones of the post-alveolar fricatives.

Phonology help for (Standard American) English speakers:
/ʃ/ is the sh sound in shell
/ʒ/ is the j in French (think j in j'aime). It's the voiced version of ʃ.
/tʃ/ is the ch sound in chin.
/dʒ/ is the j sound in jam.
/ɾ/ is the r sound in Japanese. It is not the r sound in English, which is the voiced alveolar (or postalveolar) approximant /ɹ/. It's similar to the Spanish trilled r, except that you only touch the tongue to the palate once, rather than multiple times.
/r/ is the Spanish trilled r sound.
/j/ is the y sound in yuck or yum.
/ɲ/ is the ny sound in "nyan cat". It's the ñ sound in Spanish.
/ŋ/ is the ng sound in going or boing.
/ʔ/ is the glottal stop, which is the catch in the throat in uh-oh.
The rest of the sounds you should already know.

In the earlier stages of Sekko development, there were more sounds represented with non-ASCII characters: ň ž š ņ, the last one being the velar nasal (which remains). Changes caused the other three to be removed, leaving ņ as the only non-ASCII character. Because of this, the velar nasal is now written as q, following Toaqic and Ceqlian tradition. There is a wide variety of sounds assigned to letters c, and q since they don't really have canonical pronunciations, and IPA does not use them. Sekko used to have c to mark tʃ, but it was removed.

Phonemes in parentheses are permitted allophonically (see further down).

All consonants may geminate and have length distinction, except for h and '. Geminated consonants are written twice. The length of a geminated consonant is twice the length of a short consonant.

Due to the structure of allophones (see below), the following consonants also cannot geminate: w j ʃ ʒ ɲ

The geminated form of the alveolar tap is the alveolar trill. At least two taps must be spoken for the alveolar trill.

Vowels

Front Back
Close i _ ⟨i⟩ _ u ⟨u⟩
Close-mid _ o ⟨o⟩
Open-mid ɛ _ ⟨e⟩
Open a _ ⟨a⟩

Thoughts: I might permit the back open vowel and have a front-open back-open vowel distinction. For now, I consider roundedness to be non-allophonic. You must have rounded u and o, and you must unrounded i e a. I may change this, but this means that the approximants have to have allophones as well. Phonology help: Roundedness refers to the shape of the lips when pronouncing a vowel. Notice how when you say o, your mouth puckers up. But when you say i, the lips relax. Many languages have phonemic roundedness distinction. Compare German i and ü. (Okay, ü is slightly centralized in German but my point stands!)

Vowels have length distinction. Long vowels are written twice. The length of a long vowel is twice the length of a short vowel.

Only /Vi/ /iV/ /uV/ /Vu/ diphthongs are allowed. Length distinction is forbidden in diphthongs, and all vowels in diphthongs are to be pronounced short. (e.g. /uua/ is forbidden)

Allophones permitted:

  • /niV/ == /ɲV/
  • /tiV/ == /tʃV/
  • /diV/ == /dʒV/
  • /iV/ == /jV/; /Vi/ == /Vj/
  • /uV/ == /wV/; /Vu/ == /Vw/
  • /siV/ == /ʃjV/; /suV/ == /ʃwV/ (Note: Bare /ʃV/ is forbidden.)

Syllable structure

Sekko has a CcV+(C(c)) syllable structure. Consonant clusters may only occur across syllable boundaries; consonant clusters cannot occur in the same syllable (ignoring approximants). The nucleus may be of arbitrary length, so long as all pairs of vowels are legal diphthongs. (e.g.: In /kuioua/, /ui/ /io/ /ou/ /ua/ are all legal diphthongs.)

Null onsets are forbidden.

See this Google Sheets chart for a chart of permitted consonant clusters (sadly, I cannot figure out how to embed pictures in the editor).

Orthography

Sekko uses Latin script as of now. Spaces are optional (the self-segregating morphology system is sufficient to find word boundaries unambiguously) but highly recommended for ease of reading.

Punctuation, such as the question mark to mark interrogative sentences, commas for connection, and periods for sentence endings is optional (the sentence fence suffixes are sufficient to find sentence boundaries unambiguously).

Capitalization to mark sentence beginning and names is optional.

The apostrophe, representing the glottal stop, has no upper-case form.

In Lojban, capitalization and punctuation are not used (in the English style, at least). However. Toaq does use English-style capitalization and punctuation (but it is not mandatory). Sekko takes the Toaq approach and makes it preferred but optional to have English-style capitalization in Latin orthography.

2

8 comments, sorted by Click to highlight new comments since: Today at 4:24 PM
New Comment

Lukewarm takes:

Phonology should be significantly optimized for aesthetics, as long as the loglangishness doesn't suffer. The sheer ugliness of Lojban is IMO a big reason why it's not as popular as it should be. As a second point on the "optimize for popularity" topic, if there's ever a conflict between ease of pronounceability for English speakers versus any other language, err on the side of English.

Having any character not in the a-z range has two major drawbacks – the first is that it's going to be really annoying to type for a vast amount of people. Typing "Ņ" with my Swedish keyboard requires me to do the awkward hand movement of pressing AltGr+, then releasing the keys to press Shift-N. I'd rather have it be any other character that's available.

Secondly, and this applies to the apostrophe too, is that a lot of things that has to do with computers doesn't deal with those characters very well. Anything with e.g. an apostrophe will be hard to google for, it will often need escaping if it's inserted in a string, and likely won't be useable as tokens (e.g. a variable name in programming languages, a browser user-agent, computer usernames...) – and even in the cases where it is usable, it requires ugly hacks to get working (domain names, filenames).

Hm, you're right about the apostrophe. Toaq and Eberban use the apostrophe for the glottal stop. Lojban uses the period. What other glyph do you suggest? x? Of the letters in the English alphabet, the following are not used by Sekko: j f v w x.

Yeah, x seems the most appropriate candidate. It sufficiently rare in English to not trip people up too much, from a cursory glance at Wikipedia it's at least used for that purpose in Pirahã, and it even looks like a little pictographic "stop" symbol.

Edit: Oh, apologies, I completely misunderstood the part where "ņ" was actually written with the letter "q". Nevermind that part!

I would prefer if the post would contain the argument for why you make the choices in the language design that you make and not just states which choices you make. 

In truth, the phonology is unimportant for making a loglang. For making an artlang or an auxlang (auxillary language), phonology is very important -- but not for loglangs. It's more of an aesthetic choice. Also, I'm not very good at phonology. I used to have a loglang partner who co-created the language with me, but she left early on because of mental health issues.

Conlang making is weird because it's less "design" and more like...feeling what feels right. There are certain things that seem correct and certain things that feel wrong based on which elements you've put down previously. You "fix" some points, and the rest of the design falls into place. In that sense, it's more like discovery. I'll tell you a little history about Sekko's phonology, though.

Sekko originally had a Finnish-esque phonology. Voicing distinction on plosives and affricates (we had tʃ dʒ), no voicing distinction on fricatives. At that time, we also had the labiodental fricative f. Vowel and consonant length distinction were already present. The approximants j w were treated as separate consonants and not treated as diphthong allophones -- diphthongs were instead pronounced with hiatus. The palatal nasal ɲ was also a separate consonant and not an allophone.

There was no voicing distinction on fricatives because we didn't want a v-w distinction (and because Finnish didn't have it either, but we soon stopped caring about faithfulness to a Finnish-esque phonology. Now it's more of a weird mix of Japanese and Mandarin because of the labialization and palatization of consonants).

Now, this was all fine, except that I didn't like how diphthongs had to be pronounced with hiatus. It's much better to permit them to be said with glides. That entailed removing j w as true consonants.

I didn't like that we had no ʒ but we had dʒ. The labial fricative was removed, and voicing distinction for fricatives permitted.

Because consonants can be palatized now, we had a tj tʃ distinction (think of English "tube"). I didn't like this, so the affricates also lost their status as true consonants and became allophones.

Now there was less reason to remove ʃ and ʒ (I actually might re-add them). It's the same situation as before, where there was a sj ʃ ʃj distinction. I didn't like that very much, so the post-alveolar fricatives had to go.

In truth, the phonology is unimportant for making a loglang. For making an artlang or an auxlang (auxillary language), phonology is very important -- but not for loglangs. It's more of an aesthetic choice.

I don't see any reason to think of strict categories like auxlang and loglang that are mutually exclusive.

The notion of a loglang means that you want your language to fulfill certain criteria and are thus willing to make it a bit harder to learn, but that doesn't mean that ease of learning the language is irrelevant.

Lojban itself goes through exercises like making it's terms resemble existing words to be easier to learn.  

I know it's phonology and spelling rather than phonetics you're mainly talking about in this post, but you talk about allophones anyway, so I feel I may ask this. Will tone / pitch or stress play a role in the language?

I thought very much about having a simple high-low tone accent system like Japanese or Ancient Greek. It would make SSM very, very easy and simple, given that I can just make word beginnings take the high tone and continuing syllables take the low tone (or the reverse). However, I'm unduly biased against tonemes, so I didn't include them. Toaq is a loglang with a seven tone system, if you're into tonemes.

Stress will not. I already have phonemic vowel length and consonant length. The only remaining differences possible are tone (which is pitch accent), and volume. I don't want to have volemes (do those even exist?).

New to LessWrong?