In my budding conlang called Tune, I have "base" words which start and end with a consonant, either 3, 4, 5, or 6 sounds, with 1 or 2 vowels. Examples:

  • tun ("tune")
  • tunan ("tunahn")
  • torvan ("torvahn")
  • twin ("tween")

(where each Roman/Latin letter has a specific sound associated with it).

I was initially thinking of having these "bases" mean simultaneously a noun, a verb, and an adjective. Basically they would mean the full range of possibilities of that concept. Then you would append a -a to make it an explicit noun, append a -i to make it a verb, and a -u to make it an adjective. But the problem with this is that every word now is at least 2 syllables. I noticed that at least in English, most words are 1 syllable, if I were to guess, most words in regular English conversation are:

  • 1 syllable (50%)
  • 2 syllables (25%)
  • 3 syllables (15%)
  • 4+ (10%)

Just a feeling there, but the feeling is many words are simply 1 syllable. I would like to make most words one syllable if possible. But that means ditching my idea of the suffixes potentially.

So I'm wondering, if every word starts/ends with a consonant and is one or two syllables, how will you be able to tell which is a noun, verb, or adjective (or determiner/etc., if it's similar to English or Chinese in structure). What do I need to do to make sure I don't run into complex problems down the road when the language gets more vocabulary? It seems like some sort of fixed sentence word order with word "slots" would be required, but I'm not sure what the possibilities are. Can you outline some things I could do if I get rid of the suffix (or only use it in some cases) to make sure I can effectively include nouns, verbs, and adjectives without confusion? I am not sure what to imagine in trying to solve this.

If I make it so all words fit these patterns (start and end with consonant, 3-6 sounds long), and each word can be used as a noun, verb, or adjective "noun modifiers" (or adverb for verb modifiers), then I could do like buffalo buffalo buffalo... sentence, if there are no suffixes to distinguish in the common case. I am just not sure if I need to fix the word order or something like that.

It seems hard because:

The tree-y tree treeified treeily.
The [adjective] [noun] [verbed] [adverb]

If tree was written/pronounced dip ("deep"), then:

dip dip dip dip.

That could mean anything. So it seems like you need in certain context to add the suffixes, but I'm not sure what the possibilities are. The suffix system might have it be:

dipu dipa dipu dipi.

Which would make more sense, but then we are at 2-syllable words for everything again.

What I'm considering doing is the following. Making each "base" mean the most common frame of reference (noun vs. verb vs. adjective), and then the suffixes turn it into the other forms. So the most common reference for a "tree" (base), would be the tree as a noun, not treeify (verb). Then there is "walk" which is most commonly considered an action, but could also be a noun, so the default base means the verb, not the noun in that case. But some cases are in between like "calm" (to calm or the calm?), so those perhaps always distinguish with the suffixes -i or -u. But maybe you are allowed to add the extra information to specify if desired or to disambiguate, but otherwise you can use the base form to mean the most common form of the concept.

  • 1
    I know an English speaker who drinks drinks made of orange oranges. Some might also fish for fishy fish, or cut out a well cut cut. Somehow they get by. My favourite example from my native (Swedish) Swedish is the sentence "En bar barbarbarbarbar bar en bar barbarbarbarbar". Somehow most people manage to understand it if I say it slowly with the right intonation. I guess I'm just trying to say people can manage language tasks which seem impossible to an outsider, though I'm not at all sure how this is done in practice.
    – Edvin
    Commented Aug 21, 2022 at 17:15
  • As an aside, I suggest writing down a definition for each class of word nonetheless. For example, that does the root for tree mean when it is used in a verb? Is it "to stand like a tree" or "to grow"? Does the root for "eat" become the noun "food" or the noun "eater", or both depending on context? The choices you make here are where the personality of your language and conculture can shine :)
    – Domino
    Commented Sep 19, 2022 at 20:51
  • I guess what I was trying to say with my previous comment is: speakers of the language will think of these as different words -- derivational morphology as opposed to inflexion. And derivational morphology can afford to be irregular.
    – Domino
    Commented Sep 20, 2022 at 3:42

3 Answers 3


There are plenty of languages that require explicit marking on certain categories of word. In Latin, for example, all nouns need to be marked with case and number, and all verbs need to be marked with TAM and person. This means that all nouns and verbs are polysyllabic; Latin-speakers just spoke faster to compensate for the longer words. Esperanto famously works this way, with -o for nouns, -a for adjectives, -i (or some others) for verbs. Or you can change something inside the root to indicate the category; in Akkadian, parās- is a noun, pāris- is an adjective, purus is a verb, all from the stem P-R-S (along with a good dozen other patterns).

(Well, technically all Latin nouns and verbs are sometimes polysyllabic. Occasionally the ending combines with the stem and makes a single syllable: falc + -s = falx "sickle", da + = "I give". But forms with different endings still end up being multiple syllables: falcī "for a sickle", dare "to give".)

There are also plenty of languages where there's no explicit marking on these parts of speech, and you have to rely on syntax to know which is which. In English, pretty much any word can be used as a verb. And the fairly-rigid syntax means this is seldom an issue. "Index this file by tomorrow" is unambiguous that "index" is the verb and "file" is the noun, and vice versa for "file this index". Some languages require the verb to be the first phrase in the sentence, or the second, or the last. In toki pona, the first phrase in a construction is always the subject, then the predicate (separated by li), then the object (separated by e).


Apart from adding syllables, you could use word order in combination with with prosody (tone, emphasis, pausing...) and other pronunciation differences.

Some brainstorming examples for you to remix:

  • nasalize the vowel for an adjective; one tone (even without being a tonal language, like English uses tone for questions, sarcasm, emphasis, etc.) marks nouns; do neither for verbs
  • mark a verb by saying its first consonant distinctly: hold it for nasals, eject it for stops; do the same with the final consonant for nouns; mark an adjective by stressing it
  • words are nouns by default; the final word in an utterance is always the verb; a word with a short pause {before, after, within} it is an adjective
  • the noun is tun (unvoiced, unaspirated), the verb is dun (voiced, unaspirated), the adjective is tʰun (unvoiced, aspirated)
  • let one vowel never appear in nouns; make all adjectives verbs

In my own work I have "bases" like you had in mind. I know nouns come after declension particles and verbs come after conjugation particles. The only morphological change in the language is to mark an adjective: say the first vowel twice.


If you don't want to introduce another syllable, have your markers be final consonants, so you get e.g.,"tunant" is a noun, "tunans" is a verb., "tunansh" an adjective, etc. You may have to jigger things to keep the final cluster pronounceable, or have multiple final consonants that serve the same role (e,g., using "-d" or "-t" to match the voiced/unvoiced last vowel of the root).

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.