Category: conlanging

Creating a Fictional Language Family Part 6: Last Sound Changes in Language 1b

Nilakash Roy-Faderman

I have a much less strong idea of the phonaesthetic of language 1b as possible. As such, most of the sound changes I’ll make will be trying to differentiate this language from language 1a. We’ll see what comes out once it’s done, and I can go back and tweak things if something needs to be changed for the phonaesthetic.

I already stated that I didn’t like voiced plosives for language 1a. While that is true, it’s probably more accurate to say that I don’t like voiced plosives in general. (Of course, they can be nice-sounding in some contexts, but I often find myself trying to get rid of them.) Fortunately, they’re easy to dispose of! This process will occur differently in language 1b from 1a, which will further differentiate them. 

C{+obstruent,+voiced} → C{+obstruent,-voiced} / _ C{-voiced}, C{-voiced} _

C{+obstruent,+voiced} → C{+obstruent,-voiced} / _ #

b, d, dʑ, g → β, ð, ʑ, ɣ

β, ð, ɣ → β̞, ð̞, ɣ̞

Now, we have something that looks like Spanish, where voiced stops are often realized as approximants. But unlike Spanish, the “voiced stops” are never realized as voiced stops – they’re always either voiceless stops or voiced approximants. Because of this, I think it’s more reasonable to treat the approximants as proper phonemes.

Like the voiced stop shifts in language 1a, this change will create some fun alternations when it’s applied to things that can both prefix and suffix. In addition, these forms will look really different from the ones in language 1a. An affix that could take the form d- or -d in Protolanguage 1 will take the form t- or -z in language 1a, but ð̞- -t in language 1b!

The rest of the shifts are fairly simple:

ɕ, ʑ → ʃ, ʒ

∅ → A / $ C _ C

Note the use of A in the second rule. This is a symbol representing the underspecified low vowel, which could be ɜ, ɞ, ɑ or ɒ, depending on the vowel harmony of the word. (This rule is also written horribly – I believe this is how you’re supposed to write epenthesis, but it feels really clunky.)

And with that, the sound changes for language 1b are done! Our final table looks like this:

Lang 1b Labial Dental/Alveolar Palatal Velar
Stop/Affricate p t tʃ k
Fricative Voiceless f s ʃ x
Voiced v z ʒ ɣ̞
Approximant β̞ ð̞ l j
Taps – ɾ – –
Nasals m n
Vowels Front Back
Unrounded Rounded Unrounded Rounded
Close i iː y yː ɯ ɯː u uː
Open ɜ ɜː ɞ ɞː ɑ ɑː ɒ ɒː

Next installment, I’ll begin to tackle morphology!

Creating a Fictional Language Family Part 5: A Discussion of Phonaesthetics, and Language 1a Consonant Shifts

Nilakash Roy-Faderman

I already touched on phonaesthetics in Language 1a before, when I was talking about vowel hiatus. If you aren’t familiar, a phonaesthetic is a specific type of overall feeling to the sound of a language. I’ve already noted my borrowing from Latin and Old English for some diphthongs to give it a more archaic feeling. The use of uvulars in this language was also initially inspired by Classical Arabic. This is about as far as I’m willing to go with using real-world languages as phonaesthetic inspiration – much more would feel to me like laziness and stereotyping. The rest of the phonaesthetic of this language is much harder to describe. I think the easiest way is discussing the sounds I like and dislike for it.

I like the voiceless stops in this language, especially (q) and (k), but I also like the voiced fricatives, especially word-initially and word-finally. However, I really don’t like the voiced stops. My initial plan was to have the protolanguage have no voiced stops, but I realized that if I have them in the protolanguage, they can be lenited in some positions, which yields more voiced fricatives.

b, d, dʑ, g, ɢ → β, ð, ʑ, ɣ, ʁ / V _ V, _ C, _ #

b, d, dʑ, g, ɢ → p, t, tɕ, k, q

β, ð → v, z

An upside of this alteration is that affixes that can be both prefixed and suffixed may take different forms in different positions. For example, an affix that could take the proto-forms d- and -d would have the modern forms t- and -z. This adds a bit of variety to the sounds of commonly used affixes, as well as some interesting unclarity in related forms.

Apart from that, there are a few minor changes to make the language fit the phonaesthetic I’m going for:

tɕ → ts

ɕ, ʑ → ç, ʝ

And with that, we’re done with sound changes for language 1a! Our final table looks like this:

Consonants

 Bilabial Alveolar Palatal Velar Uvular

Stop/Affricate

 p

t ts

   k q
Fricative

Voiceless

f

 s

ç

 x χ
Voiced v z ʝ ɣ ʁ
Liquids

–

ɾ l

 – – –

Nasals

 m

n

Vowels

Front

 Back
Close

i iː

u uː
Mid e eː o oː

Open

 ʌ ɑ ɑː

 

Creating a Fictional Language Family Part 4: Rounding Harmony

Nilakash Roy-Faderman

Language 1b’s vowel harmony system was looking pretty boring, so let’s change that by introducing an additional axis of harmony: rounding!

The approach I took to introducing front/back harmony in this language was very non-traditional. The rounding harmony will occur in a much more standard fashion. The roundedness of the vowels [u], [y], [ɞ], and [ɒ] will transmit their roundedness to other vowels in the word. This is, by the way, part of why I introduced the rounded vowels in the last part: I imagine that rounding harmony will happen more easily if rounded/unrounded pairs already exist. Vowels affected by rounding will transform in this way:

i, ɯ, ɜ, ɑ → y, u, ɞ, ɒ

There’s a catch, though – unlike emphasis harmony, which spreads bidirectionally by default, rounding harmony often will only spread in one direction. In fact, emphasis harmony itself may be the result of two different processes, one spreading backwards (regressively) from the emphatic consonant and one spreading forwards (progressively) – which is why it often has different spreading patterns progressively than regressively. 

I want to keep vowel harmony bidirectional in this language; the language makes extensive use of both prefixes and suffixes, and I think it would get confusing if they worked too differently. However, I’ll have the progressive spread be halted by the vowel [ɑ] – it’s such an open vowel that you need to move your lips significantly to make the rounded or unrounded version, and since you’re already moving them, you might as well unround them. This wouldn’t happen regressively, because the regressive harmony is based on anticipation of the rounding, not lingering on it.

I also don’t want this harmony to spread from affixes – that would cause all kinds of  stuff that, while interesting, is not what I’m trying to do with this language. So instead, I’ll make this stress-dependent. Stress in language 1b always falls on a syllable in the stem (for reasons we’ll get into later), and so the vowel harmony will always be triggered by a stem vowel.

Our final table looks like this:

Front

 Back

Unrounded

Rounded

 Unrounded Rounded

i iː*

y yː†

ɯ ɯː

 u† uː†

ɜ ɜː

ɞ ɞː

 ɑ§ ɑː§ ɒ ɒː

* opaque progressively to backing harmony

† sometimes transparent to rounding harmony

§ opaque progressively to rounding harmony

(Note that the long rounded vowels will be transparent to rounding harmony outside the stem if they came from a diphthong. This is weird, and I might change it later, but it was the simplest solution.)

This overall system of vowel harmony might look too clean – there are two basic vowel forms, each one with four potential forms for the four potential combinations of backness and roundedness that correspond pretty exactly on the vowel chart. But I’m fine with this system for this language; the complex spreading rules and multiple hybrid vowels (vowels that are sometimes neutral) make this system weird enough for me. Besides, we can mess it up more in descendant languages. 

Again, I have nothing to recommend for further reading, at least apart from what I’ve already recommended – Rose and Walker’s “Harmony Systems” is an excellent overview, and one of the only scholarly works on harmony that I’ve found that was concise and comprehensible to someone without a linguistics degree. That’s all for now! In the next installment, I’ll be giving the consonants some love. (And by “giving them love”, I mean killing a few of them in the name of phonaesthetics.)

Creating a Fictional Language Family Part 3: Vowel Hiatus, Diphthongs and Monophthongization

Nilakash Roy-Faderman

Proto-lang 1 (the ancestor of the two modern languages we’re working on) permitted “vowel hiatus” – that is, it allowed two vowels to exist next to each other. The only rule governing vowel hiatus was that the first vowel must be short; apart from that, any vowel combination was legal. But vowel hiatus is pretty unstable, and it doesn’t survive into the daughter languages. 

Language 1a deals with vowel hiatus by creating semivowels, in this series of sound changes:

V → V(+semivowel) / _ V(+long)

i, u → i̯, u̯ / _ V

i, u → i̯, u̯ / a _

This results with a table of diphthongs that looks like this:

  i u a
i N/A u̯i ai̯
u i̯u N/A au̯
a i̯a

u̯a

 N/A
iː N/A u̯iː a̯iː
uː i̯uː N/A a̯uː
aː i̯aː u̯aː

N/A

These diphthongs will be affected by vowel harmony. After vowel harmony occurs, you get something that looks like this:

  i u a
i N/A u̯i, o̯e ʌi̯, ae̯
u i̯u, e̯o N/A ʌu̯, ao̯
a i̯ʌ, e̯a u̯ʌ, o̯a N/A
iː N/A u̯iː, o̯eː ʌ̯iː, a̯eː
uː i̯uː, e̯oː N/A ʌ̯uː, a̯oː
aː i̯aː, e̯aː u̯aː, o̯aː

N/A

As you can see, this language has a lot of potential diphthongs, including some pretty weird ones. These diphthongs hit the precise phonaesthetic I’m going for. The lowered ones in particular remind me of Latin (oe, ae) and Old English (eo, ea), appropriate for a language I want to have an archaic feel. 

Language 1b follows a slightly different path to resolve vowel hiatus than 1a:

short i, u → j, w / V _

short i, u → j, w / _ V

a → ∅ / _ V(+long)

This results in the following table:

  i u a
i N/A uj aj
u iw

N/A

 aw
a ja wa

N/A
iː N/A wiː iː
uː juː N/A uː
aː jaː waː N/A

There’s another step, though. Some of these diphthongs undergo monophthongization, which in some cases results in the creation of new vowels:

uj, aj → yː, ɜː

yˤː → uː

iw, ɯw, ɜw, ɑw → yː, uː, ɞː, ɒː

The final table looks like this:

  i

u

 a
i N/A yː, uː ɜː, ɑː
u yː, uː N/A ɞː, ɒː
a jɜ, jɑː wɜ, wɑ N/A
iː N/A wiː, wɯː iː, ɯː
uː juː N/A uː
aː jɜː, jɑː wɜː, wɑː

N/A

This means that [uː] is now a hybrid vowel. Sometimes it’ll be transparent to front-back harmony, and sometimes it’ll alternate with [yː]. 

With this final chart, we can also see that the syllable structure of this language has changed. Keep in mind that, unlike the semivowels in Language 1a, [j] and [w] are true approximants, acting as consonants and existing outside the syllable nucleus. While the monophthongization has prevented them from occurring at syllable codas, they can still occur at syllable onsets. Thus, the syllable structure shifted from (C)V(C) to (C)(W)V(C) (where W is a glide).

That’s all for now! (Normally, I would recommend some literature on resolving vowel hiatus here, but honestly, most of what I did here was pretty much intuitive.)

Creating a Fictional Language Family Part 2: Evolving Vowel Harmony

Nilakash Roy-Faderman

Hello! Now that I’ve shown off the protolanguage (and done what was probably a far too lengthy explanation of emphasis harmony), I’m getting into the descendant languages, creating the vowel harmony systems as they evolve from emphasis harmony in the two descendants: Language 1a and 1b.

Language 1b

(Yes, I know, I’m explaining Language 1b before I explain Language 1a. But I’d already decided on the naming convention, and this one makes more sense to introduce first.)

In Language 1b, emphasis harmony causes vowels to back. The emphatic vowel [iˤ] goes to [ɯˤ], and the emphatic vowel [aˤ] goes to [ɑˤ]. [a] also changes when not emphatic, going to [ɜ]. (I was initially skeptical of the realism of this last change, but a similar thing happens in Arabic. Many dialects of Arabic have non-emphatic [a] going to [æ] in most environments, and in Tunisian Arabic, it shifts all the way to [ɛ].) None of the vowel changes in Language 1b are affected by stress.

This leaves us with a paradigm that looks like this:

Front

Back

i, iː¹

 ɯ, ɯː
u², uː²

ɜ, ɜː

ɑ, ɑː

¹opaque progressively

²sometimes transparent bidirectionally

Language 1a

In language 1a, emphasis causes the lowering of vowels. The emphatic vowel [iˤ] goes to [eˤ], the emphatic vowel [uˤ] goes to [oˤ], and the emphatic vowel [aˤ] goes to [ɑˤ]. Unemphatic [a] goes to [ʌ] when short, but it goes to [ɑː] when long. This makes [ɑː] a neutral vowel, and, interestingly, means it’s transparent to vowel harmony – an unusual characteristic for a long low vowel, but one that makes sense given the background history. This leaves us with a system that looks like this: 

High Low
i, iː e, eː
u, uː o, oː
ʌ ɑ
ɑː²

¹opaque progressively

²transparent bidirectionally

But wait! There’s a twist!

See, emphasis doesn’t just alter vowel quality – it can alter consonant quality too. Emphatic /kˤ/ often goes to /q/, and it wouldn’t be a stretch to say that this would impact other velar consonants. If this change occurs after emphasis harmony is established, it could create a velar-uvular consonant harmony system:

proto xaːkˤ > xˤaˤːkˤ > modern χaːq

proto xaːk > modern xaːk

Velar-uvular consonant harmony isn’t common, but it does occur in some Totonacan languages. But what we have here isn’t just velar-uvular consonant harmony – it’s a system of consonant-vowel harmony. Uvulars only occur in words with low vowels, and velars only occur in words with high vowels. This sort of harmony, “faucal harmony”, is very rare, though it occurs in Interior Salish. However, emphasis harmony leaves behind an interesting and (as far as I know) unprecedented property: faucal harmony can occur in words without any uvulars in the stem.

proto tˤuv > tˤuˤv > modern tov

proto tˤuv-ik > tˤuˤvˤ-iˤkˤ > modern tov-eq

I’m not too worried about this, despite its apparent lack of precedent. The sound changes that arrived at this point are all naturalistic, and I don’t think that it would be likely to get removed via assimilation, given the clear link between low vowels and uvulars in other contexts. 

This leaves us with two pretty fun vowel harmony systems evolved out of the old emphasis harmony system. The harmony system in language 1b is still pretty bland, but there are things coming that will make it a bit more interesting. For further reading on emphasis harmony and faucal harmony, I recommend Ananian and Nevins’ Postvelar Harmonics: A Typological Odyssey, which goes into far more detail than I can about emphasis harmony in Palestinian Arabic and faucal harmony in Interior Salish. I also recommend Rose and Walker’s Harmony Systems, an excellent guide to consonant harmony, vowel harmony, and vowel-consonant harmony – especially good if you’re considering implementing some type of harmony in your language. In the next installment, I’m going to further complicate the vowels by discussing diphthongs and vowel coalescence! 

Creating a Fictional Language Family Part 1: Proto-phonology, and a Discussion of Emphasis Harmony

Nilakash Roy-Faderman

Hello! This time, as promised, I’ll be actually getting into the conlanging. Let’s start with some protophonology:

Lang 1-proto Consonants

Bilabial

 Alveolar

Avleolopalatal

Velar
Stop/Affricate

Voiceless

p

 t tɕ

k

Emphatic

pˤ

 tˤ tsˤ

~

 kˀ

Voiced

 b d dʑ g
Fricative

Voiceless

f

 s ɕ x

Voiced

 v

z

ʑ

 ɣ
Liquids –

ɾ l

    
Nasals m

n

 –

–

Lang 1-proto Vowels

Front

Back

Close

i iː

u uː
Open

a aː

Phonotactics: Default (C)V(C) syllable structure. Allows word-initial syllable structure of sPV(C), where P is a voiceless plosive. Allows vowel hiatus of up to two vowels.

On the whole, this is a fairly standard phonological system: it’s got a classic three-vowel system with a length distinction, full series of voiceless and voiced plosives at the three major places of articulation, and a fourth palatalized place of articulation (though note the alveolo-palatal obstruents instead of the more common palato-alveolars). But you’ve probably noticed the “emphatic” category. Let me explain that.

Some Afroasiatic languages, such as Arabic, have a series of consonants with a pharyngeal secondary articulation*. These consonants are referred to as emphatic consonants, and they derive from a series of obstruents in Proto-Afroasiatic that contrasted with separate voiced and voiceless series, much like the one you see here.

These pharyngealized consonants often have the interesting property of spreading their emphasis to nearby sounds. In some dialects, this may only affect adjacent consonants – in others, it can affect the entire word, in a process called “emphasis harmony”. Even more intriguingly, vowels affected this way tend to change their quality. Sometimes vowels affected this way will back, sometimes they will lower, sometimes they do both. This starts to create something that looks like vowel harmony.

However, there’s nothing happening at the phonemic level. These vowel shifts are purely allophonic, still being determined by the presence of an emphatic consonant. However, if the emphasis distinction were to be lost after emphasis harmony has taken effect, we might end up with a system of actual vowel harmony. In fact, we could end up with several different systems in different descendant languages, depending on how emphasis harmony affected the vowels. The two descendant languages that I’m creating differ in how they handle emphatic vowels, resulting in two very different harmony systems.

In these systems, vowel harmony will spread like emphasis harmony did in the protolanguage, resulting in some unusual properties. Emphasis harmony tends to spread bidirectionally. When it spreads regressively, it is most often unimpeded, but when it spreads progressively, it often is blocked by high front segments. This includes the vowel [i], but also consonants like [j] and even [ʒ]. For this protolang, I’ll follow this pattern. Emphasis harmony isn’t blockable regressively, but is blocked progressively by [iː], [ɕ], [ʑ], [tɕ], and [dʑ]. These patterns will persist into the descendant languages, meaning they’ll end up with vowel harmony being blocked by consonants!

In the next installment of this, I’ll be getting into the actual languages, evolving out vowel harmony systems for two descendent languages. Future posts on this won’t be nearly as research-heavy. Emphasis harmony is a niche enough feature that I thought it merited a fair bit of explanation; most of the other changes won’t need a whole blog post of background.

*It’s actually uvularization in Arabic, not pharyngealization. However, it’s almost always represented with the pharyngealization symbol, and a lot of places will still refer to it as pharyngealization instead of uvularization, so I’ll follow the standard. Also take note that references to “emphatic” consonants aren’t always talking about pharyngealization/uvularization. In Proto-Semitic, the emphatic consonants were ejectives, and they retained this quality in some modern languages. Emphatic consonants also gave rise to implosive consonants in some non-Semitic Afroasiatic languages. 

Editor’s Note: I’m completely guessing at the date on this one. The concept of evolving emphasis harmony into vowel harmony goes all the way back to Yiksighe (though I did it really weirdly there). I’m dating this to December 8, 2022 – the creation date for the earliest documentation I have of the protolanguage – but most of the research here was done earlier for Yiksighe.