Morphemes

← Guide index · 中文 · fifth item in the navigation bar

The Morphemes page is one table for roots, affixes, clitics and particles. The division of labour with the lexicon: the lexicon stores lexemes (finished words with a part of speech, senses and inflected forms), while the Morphemes page stores the parts. A proto-language’s list of roots usually lives here.

The list works like the lexicon (see Common conventions); filterable columns are initial letter, type and tags (plus language when All languages is selected). Next to the title there is also a Statistics sub-page (click a distribution item to filter the list, or a ranking item to jump to it). It shows the number of morphemes; the share with a gloss / meaning / allomorphs / etymology; how many are used in corpus analyses, etymologies and paradigms; a list of unused morphemes; and duplicates with the same form and type. Distributions are by type, tag and initial letter.

Next to the title are two modes, View / Edit: View shows a read-only card (form, type, gloss, meaning, allomorphs, etymology, the entries that use it); Edit is the field-by-field form, just like in the lexicon. Double-click a row in the table to open it in the inspector directly in Edit mode.

1. Fields

The list only shows morphemes of the current language (with an extra Language column for All languages). By default it is sorted by the language’s custom alphabet.

2. Writing allomorphs

The environment is exactly the environment part of the rule syntax, where _ is the morpheme itself: a suffix looks at the end of the stem before it (V_ after a vowel), a prefix at the start of the stem after it (_d before d, _Vr before “vowel + r”). _CC is any two consonants, while _C1C1 is the same consonant doubled (the same number means the same sound). It doesn’t matter if the class is written in IPA (k in C) while the stem is spelled (c): the environment is compared against both the spelling and the pronunciation derived from the primary orthography. The ? next to Allomorphs and Syntax help next to the title are always there to look up the notation.

When a suffix has different forms in different environments (vowel harmony, consonant assimilation), don’t create several morphemes — write allomorphs under one morpheme:

Form: -lar
Allomorphs:
  -ler   / Front?_      ; when the preceding vowel is a front vowel
  -lar   / _            ; default

The first environment from the top that matches wins; write the default as _ and put it last.

3. A morpheme’s etymology

Morphemes have the same etymology chain as entries: type (root / compound / derivation / sound change / loanword / unknown or a custom type), sources, stages and notes. Once filled in, the morpheme is joined into the same relation graph as the entries that use it, so you can follow a word all the way back to earlier morpheme sources. For how the relation graph works (including splitting concatenated sources into nodes and jumping across languages), see Lexicon · Relation graph.

4. Referring to morphemes in paradigms

Generators on the Paradigms page can write @morpheme name (or @ followed by the morpheme’s form); the software picks an allomorph with the environment rules above and attaches it to the stem. That is better than hard-coding a literal suffix in the generator: change the morpheme and every paradigm follows, and the gloss abbreviations stay consistent.

5. Role in the corpus

When the Corpus page analyses sentences automatically, it not only looks words up in the lexicon but also strips affixes listed here, up to two layers deep: kasolarda → kaso-lar-da, each piece taking the morpheme’s gloss. Clitics are joined with =.

6. Referring to morphemes in sound-change rules

Sound-change, orthography and script mapping rules can all write @name, where the name is the morpheme’s gloss or its form without hyphens (@lar, @PL). It acts like a class whose members are all of the morpheme’s allomorphs, so alternations such as @lar > @ler / Front?_ can be written directly as rules; @name in the replacement corresponds by position to the class in the target.

7. Connections to the lexicon

• An entry’s etymology source can point to a morpheme (root derivation); the entry card then shows ← root ‘meaning’.
• Import → Import from CSV in the title bar opens the column-mapping wizard with Morphemes already selected as the target, and the inspector shows a live import preview (form, type, gloss, meaning, tags); switching the target to Morphemes in the Lexicon page’s CSV wizard does the same.
• The Input field of lexicon evolution can use the source language’s morpheme list, deriving a proto-language’s roots into a daughter language’s lexicon in one go.

8. Tips

• Two morphemes with the same form but different meanings (homonyms) can coexist in one language, as long as their glosses differ.
• Export in the title bar exports a morpheme CSV (for the current language); the Lexicon page’s export menu has it too.