How To: Phoneme Inventory
Deciding on phoneme inventories can be super daunting because it sets the tone for the entire language project, but there are certain restrictions you can put in place that make it easier.
I always aim to make natural languages - languages that could have been spoken by humans on this planet if things had turned out a little differently. This limits you to sounds within the bounds of human articulation and perception.
There are also universal tendencies that occur across natural languages, some of which are:
- all languages have consonants and vowels (this is an absolute universal)
-all languages have stops/plosives
- if a language has voiced stops they will have the voiceless equivalents
- alveolar/dental, bilabial, and velar tend to be the most common places of articulation
- languages that have fricatives generally have greater place of articulation contrasts than in stops
- of fricatives s is most common, followed by ʃ and then f, while θ/ð is actually quite rare (English speakers beware)
- vowel inventories seek to maximise distinction by spreading out across the vowel space, so you’re not going to get a clump of closely located vowels unless you have a large inventory
- the most minimal vowel inventory will generally have three: a high front vowel, a high back vowel, and a low central vowel as this defines the limits of the vowel space e.g. [i],[u] and [a]
- height distinctions outnumber backness distinctions, but the number of height distinctions in front vowels is equal or greater than height distinctions in back vowels due to the shape of the vowel space
- if there are rounded front vowels there will also be unrounded front vowels, but rounded back vowels are more common than unrounded ones
If you’re struggling with developing a phoneme inventory it might be useful to do some research on a language you like the sound of: Wikipedia usually has some pretty accessible tables of a language’s phonemes. Shop around and see what’s right for you.
