This section gives technical details that weren't covered in the Intermediate section, about sounds that were covered.
Gemination is the lengthening of consonants. Some languages —for example Italian and Classical Arabic— distinguish between a single consonant and a double "geminate" consonant. Geminate consonants mostly occur in intervocalic (between vowels) position. (Swedish is unusual for its final geminates). Most languages only allow some of their consonants to be geminated.
Although English doesn't directly provide geminate consonants in its sound inventory, gemination can still occur in a compound word, or two consecutive words, when the first ends with the same consonant that the second begins with; for example, in cattail (with a geminate plosive), misspell (geminate fricative), roommate (geminate nasal).
Diphthongs are complex sequences of vowels which are made of two vowels pronounced next to each other. They are contrasted from "flat" vowels in many languages. They are made of an on-glide or an off-glide vowel as well as an obligatory nucleus whereas a "flat" "pure" vowel would only be made of a nucleus. Most common on-glides/off-glides are the vowels [i] (realized as [j] when it's a glide) and [u] (realized as [w] when it's a glide). Occasionally, some languages feature triphthongs made of both an on-glide and an off-glide (examples: [jaj] or [jew]). They must not be distinguished from hiati which are sequence of vowels separated by a syllable break (example: [ej] would be a diphthong whereas [e.i] would be a hiatus, French distinguishes <paye> [pej] "pay" from <pays> [pe.i] "country").
Suprasegmentals are distinctions that apply not to one phoneme but to a whole syllable or word. Notable suprasegmental features are:
-stress (applies to a specific syllable of a whole word)
-intonation (Few languages make complex intonation distinction however some languages make simple intonation distinction such as Spanish and French which use it to distinguish statements from questions)
-juncture (where a syllable break occurs, often distinguishes multiple words from one, example in English are <a name> [a.nejm] and <an aim> [an.ejm]) -tone
-vowel and consonant harmony, the latter being relatively rare (a separation in groups of different phonemes and the exclusive use of one of those group to form a word, for more information see the section "Vowel Harmony" below)
-mora (the use of a distinctive syllable's weight which determines stress, timing)
A morphome is a coherent part of the form of a word. The difference between a morphome and a morpheme is that morphemes are defined primarily by their meanings, but morphomes are form without necessarily any meaning. The term morphome was coined for use in word-and-paradigm morphology, an approach to morphology that does not, generally, acknowledge that meaning can be properly attached to any unit of language smaller than a word. So, no morphemes; but words are still sometimes evidently made up of smaller parts, hence the term morphome.
For example, in English there are a number of words ending with /n/ (or in some dialects, /@n/):
- mitten smitten bitten kitten happen cotton passion harden garden ...
There's no single meaning associated with the ending in all these words, so it isn't a morpheme; but it does seem to be a recurring word-piece with some sort of formal (though not semantic) identity. So it's a morphome.
There is a fair amount of confusion surrounding the linguistic term morphome. If an author starts talking about the meaning of a morphome, they've misunderstood the term and probably should be using morpheme (since morphemes have meaning while morphomes do not; and yes, prominent authors have been known to make this mistake). The confusion runs deeper than that, though, because word-and-paradigm morphology is not a single approach to morphology, but a family of approaches: different linguists using a word-and-paradigm approach may be doing somewhat different things, and as a result, there may be treacherous differences in how they use the term morphome. The general sense of it, though, should still be that of a coherent part of the form of a word. Note that, on the face of it, even if you believe in morphemes, not only are some morphomes not morphemes (having no meaning), but some linguists view a morpheme as a rule or process rather than a word-piece as such — for example, a disfix which adds meaning while deleting part of the word form — so that morphemes might not be morphomes, either.
Morphomes can be a useful way of giving a naturalistic feel to a conlang: certain patterns of form, regardless of meaning, occurring more often than random chance, can give a realistic-feeling shape to the phonaesthetics of a conlang.
This feature is common in agglutinative languages, i.e. in languages where one affix represents one property; such as Turkic, Mongolian, Finnish, Hungarian. All vowels are divided into, typically, two groups, such as back vowels and front vowels, and only vowels from one group can be used in the word; therefore all affixes have two forms.
For example, in Tatar language the groups are following:
First group (Back vowels): a, ı, í, o, u Second group (Front vowels): ä, e, i, ö, ü
Plural suffix has forms lar and lär. The word for friend is dus, so friends is duslar; the word for house is öy, therefore, houses is öylär.
Sometimes there are neutral vowels that belong to both groups, like i and e in Finnish:
First group (Back vowels): a, o, u, e, i Second group (Front vowels): ä, ö, y, e, i
Usually, group is selected by the kind of vowels in the root of the word. But in the Chukchi language, for example, there are dominant and recessive group, and all vowels from the recessive group are replaced with their dominant counterparts when a suffix with a dominant vowel is used.