Connic 1 - Choosing the Phonology

Originally published on: 2021-07-10

Intro

In this post, I'll be covering thge phonology of my first serious conlang: Connic (the name is a work in progress). Right now, I'm designing a proto-lang that I'll then evolve into a modern version to make it more naturalistic and create some possibly unintended, but fun side-effects.

Goals

Language is a core part of what makes us humans, and yet most people don't even think about its inner workings. Not syntax, not grammar, not phonology. We just take it all for granted, and that's quite lame in my opinion.

In this series of posts, I intend to try to rectify this by blogging about making a language from scratch, including its phonology, syntax, grammar, vocabulary, writing system and any other part that might come up in the process.

To do this, I'll first do a proto-language (which is basically an acestor of a language, i.e. Latin is a proto-language of all modern romance language, or how Proto-Germanic is a proto-language of English, German, Dutch and Nordic languages) and then evolve it by applying

Sound shifts
This is a situation when a pronunciation of one sound or a group of sounds changes — that's why we have accents!
Semantic shifts
This is when the meaning of a word changes, i.e. how the meaning of "very" changed from "literal" — my own very eyes — to an intensifier — very good, — or how the word "literally" is now fairly commonly used to mean "figuratively".
Syntactic shifts
Which is when grammar rules change, like when English went from using an "I verb not" — for example, I know not — for negations to using the negated form of do for that — I don't know.
Other parts of language evolution which we'll talk about later.

Phonology

But first, what is phonology and how can you choose one? Well, there are actually two definitions. The first one is that it's a science that studies sounds (from Ancient Greek phonema, which means sound), specifically, their organization. The second one, which is also the one we'll be using throughout the post, is that phonology is the system of sounds of a particular language (or dialect, or even an accent). As you can see, the two definitions are related, which is not unexpected.

"But why do we even care about sounds? Don't all languages represent sounds with writing?" I hear you ask, my hypothetical reader who asks just the right question at the right time. And the answer is a definite no. I'm sure all of you have heard that "ough" in enough, cough, through and thorough all make different sounds, despite being written the same in all of them.

What's the solution to that? Why, I'm glad you asked. The answer is called IPA, which stands for International Phonetic Alphabet. It's a system specifically designed to be able to represent all sounds in all languages in all of the world. It's also those funny-looking, rotated and flipped letters you might see next to words in dictionaries. It's also the system we'll be using thoughout the posts. If you don't know IPA, don't worry. We won't be using any complicated concepts or symbols. In fact, most symbols we'll be using look exactly like the letter in English that makes the same sound with just two exceptions in the consonants. Vowels might take a bit of work to grok, but you won't feel like they come completely out of the left field.

Consonants

Consonants are the first category of sounds. They're the kind of sounds that restrict airflow in various ways to sound in a particular way. I'll quickly go over important concepts right now.

Unfortunately, I have no real desire to explain the place of articulation, manner of articulation and voicing, which is why I'm going to recommend three excellent videos by Artifexian that briefly explain what those are:

  1. Place of Articulation (exactly 6 minutes long)
  2. Manner of Articulation (5 minutes and 8 seconds long)
  3. Voicing (6 minutes and 36 seconds long)

Here's a quick summary of those in case you just need a quick refresher and don't want to watch three whole videos:

Place of Articulation
What parts of your mouth are involved in producing a certain sound.
Manner of Articulation
The manner in which the air flows out of our mouth when producing a certain sound.
Voicing
Whether your vocal cords vibrate when making a sound (voiced) or not (voiceless).

Additionally, aspiration is that little puff of air you make after a certain consonant. Compare, for example, the word pin, which contains an aspirated p at the beginning, to the word spin, which does not contain an aspirated p.

Another important part to mention is the use of // vs [] for transcription. When // is used, it signifies a broad transcription, which means all symbols correspond to the way they actually sound, although some variation is allowed. When [] is used, however, they mean a narrow transcription, which means that all symbols inside are to be pronounced exactly as they're written, with no variation. We'll be using // for all transcriptions over here since we're not transcribing some complicated dialects or trying to write an an article for a linguistic journal (also, I have no specific narrow transcriptions in mind).

Finally, let us proceed to selecting the consonants. I have no desire to make my language too exotic, but I want it to have a characteristic of its own, which is why I selected pretty much all of the common sounds: nasals /m/ and /n/, both voiced and voiceless stops /b/ and /p/, /d/ and /t/, /k/ and /g/, fricatives /f/, /s/, /h/ as well as the approximant /l/.

However, selecting only common phonemes would lead to a fairly stable phonological inventory, which would limit the ways I can evolve the language. This is why I decided to allow all stops to become aspirated (denoted as a superscript h after the sound), as well as add the dark (or pharyngealized if you prefer more proper terminology) l (written as /ɫ/).

I can already hear your questions, "What is a dark l and how is it distinguished from a regular l?" Well, first of all, it's not actually alveolar: it's usually denti-alveolar, which means that the tip of your tongue is touching the back of your teeth instead of your alveolar ridge when making the sound. Second of all, unlike regular l, which only involves one place of articulation, making the dark l involves two: you make a regular l while touching the back of your teeth, but you also start producing sound with your pharynx. The result is a dark l.

Another feature you might notice is the absence of voiced fricatives (like /v/ or /z/). This is by design, as I'm planning to evolve them during the process of evolution, as well as show how languages that traditionally didn't use to make regular distinction between them started doing so.

Now, let me present the full consonant table to you:

Fig 1. Consonants
Labial Alveolar Palatal Velar Glottal
Voiced Voicless Voiced Voicless Voiced Voiced Voicless Voiceless
Plain Aspirated Plain Aspirated Plain Aspirated Plain Aspirated Plain Plain Aspirated Plain Aspirated Plain
Nasal m n
Plosive p b t d k g
Fricative f s h
Approximant l, ɫ j

Feel free to study the table and research the terms on your own. Also, try pronouncing symbols you recognise to get a feeling for the whole system.

Vowels

Vowels are the second part of what makes up sounds. They're an important part of speech that forms the basis for syllables (mostly, but let's just ignore the caveats for now).

Vowels have three main properties:

Height
In this context, height doesn't refer to tone, but rather to tongue position. During high vowels (i.e., the ee in beet, which stands for the vowel /i:/, where /i/ is the base form of the vowel and : is a modifier that tell us that the vowel is long), then tongue is very close to your palate (which means that it's positioned high), which makes it so there's little space left between them.
In constrast, when pronouncing low vowels (for example, the a in can, which is denoted as /æ/ or the a in the British Received Pronunciation accent pronunciation of the word bath, which stands for the /ɑ/ vowel).
Another, not-totally-correct, but decently accurate way of thinking about it is how widely your mouth is open when pronouncing a vowel. If it's wide open, then it's a low vowel. If it's small and narrow, then it's a high vowel.
Basically, just try pronouncing all vowels you know and try to note their height. Another useful resource is This Wikipedia page which contains all vowels recognised as distinct by the IPA. Try listening to audio samples underneath of them and remember that high vowels like /i/ or /u/ are on the top and low vowels like /æ/ or /ɑ/ are on the bottom.
Backness
Backness is the second characteristic of every vowel. In simple terms, it's how far back your tongue is when making the sound. Compare, for example, the ee in beet (which is, as I told before, /i:/) and the oo in boot (which makes an /u:/) sound. If you pay close attention to your tongue, you'll see that when you say beet, your tongue is touching your teeth and might even exert a bit of force onto them. Compare that to your tongue when you're saying boot. It's either barely touching your teeth (because the /u:/ in most English dialects is slightly advanced to the front instead of being a true back vowel) or isn't touching them at all (if you're one of the speaker of one of those dialects that does have a true /u/).
Roundedness
Roundedness is the third and final property that all vowels posess. All vowels are either rounded or unrounded. These terms mean the shape of your lips. An example of a rounded vowel is the o in not (denoted as /ɔ:/ in IPA). Notice how your lips form an almost perfect circle when saying it. On the other hand, /i:/ (as I mentioned previously, it's the ee in beet) is unrounded. Your lips form a thin line when saying it, which is about as far away from a circle as possible.

Now, let's select the vowels. They go as follows:

/i/
Pronounced as ee in beet, but shorter.
/u/
Pronounced as oo in boot, also shorter.
/a/
Pronounced as a in hat in Received Pronunciation or the a in the General American pronunciation of bra.
/o/
pronounced as the aw in yawn in Autralian, New Zealand and some variations of Received Pronunciation, or, more accurately, the o in poso in Spanish or sol in Latin.
/e/
Its closest equivalent is the e of bed in Australian English, as well as e of Seele of Standard German or peso of Spanish or é of beauté in French.
/y/
This is probably the most complicated vowel to describe since it doesn't exist in any dialect of English. However, it does exist in German and French. In German, it's the ü of über, while in French, it's the u of tu. If you're not familiar with either of those languages, you pronounce this sound as follows: start saying /i/, but put your lips (and lips only, don't move your tongue) in the same shape as when you say /u/. Try practicing this! Also, I suggest you do go on the Wikipedia page I linked earlier and listening to the sample there.

And now, here's the vowels presented in a standard table form:

Fig 2. Vowels
Front Central Back
Close i y u
Mid e o
Open a

Phonotactics

While phonology specifies what kinds of sound may appear, phonotactics specify the manner in which they can appear. For example, the word ngob or bregb don't feel like English words, while strusk or spoolth do, even though I made all four of them up. Why is that?

The answer is that while the latter two words follow the phonotactic rules of English, the first two don't (English prohibits the consonant cluster ng in the beginning of words, as well as the cluster gb anywhere).

In my opinion, it's important to define the phonotactics for the language you're making because it helps you to create a stronger identity for it. Not wanting to make something too crazy, however, I decided to set the syllable structure as C(C)(i̯)V(C).

Now, what does that even mean? Well, first of all, C means consonant and V means vowel, while parentheses mean that the component is optional. So, C(C)(i̯)V(C) means that a syllable may consist of one or two consonants, followed by a single vowel, which may optionally be preceded by the semivocalic equivalent of /i/, and then optionally followed by a single consonant. This means that syllables like ston or tel are allowed, while syllables like en, o or jomp are disallowed.

Now, it's important to mention what semivocalic means. Basically, it means a consonant that, in a lot of cases, can be treated or transition to a vowel under certain circumstances, which is why sounds like those are also called semivowels. /i̯/ is the semivocalic equivalent of /i/, and can also be written as /j/, which is the sound the letter y makes in the words yawn, yoghurt and yacht.

One final thing, I need to explain what onset, nucleus and coda are:

Onset
Onset is the first cluster of consonants in a vowel. For example, in the syllable strengths, the cluster str is the onset. Some languages (but not ours) allow the onset to be empty (the syllable on has an empty onset).
Nucleus
Nucleus is the vowel (or a syllabic consonant, but we'll get to those in a later post) around which a syllable forms. In the syllable strengths, the vowel e is the onset.
Coda
Coda is the cluster of consonants which occurs after the nucleus. In addition, all languages allow the coda to be empty. As an example, the syllable strengths contains the coda ngths, while the syllable bra has an empty coda.

And now, without further ado, here are the phonotactic rules of Connic:

Onset

  • Any single consonant except /j/
  • Any plosive, followed by /f/, /h/, /l/, /ɫ/
  • /f/, /h/, followed by /l/, /ɫ/, /n/
  • /m/, followed by /l/, /ɫ/, /s/, /h/
  • /f/, followed by /t(h)/, /k(h)/, /d(h)/, /g(h)/, /l/, /ɫ/, /s/
  • /s/, followed by any plosive, /l/, /ɫ/, /n/, /m/

Nucleus

  • Any monophthong
  • /i̯/, followed by any short vowel except /y/

Coda

  • Empty
  • /l/, /ɫ/, /m/, /n/, /j/, /h/

Stress

Stress is the final component we'll discuss in this post. It's an important part of a language as well, but first, we'll have to distinguish two kinds of stress: lexical and prosodic.

Lexical Stress

Lexical stress is the kind of stress which changes definitions of words. English and Russian, for example, have lexical stresswhich can change the meanings of words if placed on a different syllable (compare the meanings of the words desert, incline, insult, project or reject when they have their stress placed on the first syllable to their meanings when it's placed on the last syllable).

On the other hand, languages like French, Finnish or Polish don't have lexical stress in that manner. They always consistently place their stress on the same syllable (first in the case of Finnish, final in the case of French and second-to-last in the case of Finnish; those three are the most common locations to place stress on).

My language is going to have the second kind of lexical stress - the one that's consistently placed on one vowel. I've chosen to place the stress on the first vowel since I like the aesthetic of it.

Prosodic Stress

Prosodic stress (also sometimes called sentence stress) is applied to a whole word and doesn't change the meaning, but only the connotation. As an example, compare the following two sentences: "I didn't say anything" (but someone else did) to "I didn't say anything" (but I did imply it in some other way).

All languages in the world have prosodic stress in one way or another, so I can't just choose not to have it. Although, even if I could, I wouldn't since it's a very cool feature.