In a similar light to How do conlangs/natlangs have prefixes suffixes and not get them jumbled up? , what is the use or purpose of deriving words from smaller parts in a conlang, generally speaking? It appears that natural languages evolve to do this out of necessity, reusing parts and attaching things to parts to extend them ("ex-" extend, extra, extravagant, "inter-", etc.). But is it really necessary? What advantages does this provide us in language?

The only advantages I can see is that it occasionally can give use some clue as to the words meaning. But it will never give us an accurate definition of the word's meaning. So why even bother is my main question?

For example, you might have bi- meaning two, so you have biannual (2x per year) and biweekly (every two weeks (why not twice a week? another story lol)). So given you know "weekly" already, you might be able to guess that biweekly means twice a week or every two weeks or something. But even with bi-, you might not have an accurate guess. For example, bible, biography, billing, all start with bi, yet don't mean two.

Which brings me to my second point. Often times you have words which have the same spelling as the prefix (assuming we're talking about English), but which don't have any relation. Beneath, Benefit. Bell- (war), belligerent, bellyache, bell itself. There are better examples out there. So this seems like, more than adding a degree of helpfulness, it actually adds more to the confusion overall. In the end you need to memorize the meaning of every word, and then you can backtrace/recover the meaning of the "etymological parts" from the definition of the whole. But then it's like, if we're building a conlang, what is the purpose of deriving words from smaller word parts? Other than to resemble natural languages and their evolutionary dirtiness perhaps, I guess.

It seems that you would be better off just picking random words to mean different things. Especially for foundational concepts. So instead of "inter-x" like "internet" and "interconnected" and "interwoven", you would just have "foo", "bar", and "baz" (no ba- prefix lol). So there is no relation derived from simpler parts, thus removing the confusion. But does this make it harder to learn a language? That is, do the prefixes/suffixes (as in these English examples) actually aid in understanding/memory in some way or sort? Maybe there is some benefit after all, I am not sure.

How important/useful is it to derive words from smaller parts in a conlang?

Sandhi in Sanskrit (for combining words) seems related to this question in some way, is it necessary to do such a thing in a conlang though?

  • 1
    Spelling is sometimes problematic where pronunciation is not: be-neath and bene-fit are different in that respect, similar to uni-form and un-interesting. In spoken English they are not confused, only in written. Jan 4, 2022 at 9:11

1 Answer 1


All languages have some way to combine atomic units to make more complex or elaborate meanings. Sometimes the result is considered a single word; sometimes it's not. But the underlying principles are the same either way. The advantage is that it lets you express concepts that you don't have atomic words for.

The problem you're seeing is, most of your examples aren't formed within English. They're formed within other languages and then borrowed into English. "Belligerent" is a perfectly regular formation in Latin, from words meaning "war" and "carry". But in English it's pretty much opaque and has to be learned as an indivisible unit. If you look at a regular formation within English, like "uninteresting", you probably have no difficulty at all realizing it means "not interesting", and "interesting" means that it "interests" you. (The fact that "interest" is then made up of "inter" and "est" isn't an English thing, it's a Latin thing.)

English is also fairly unusual among the world's languages in that it doesn't like using regular English compounds for new technical terms (and will instead apply Latin or Greek compounding rules and then borrow the result into English). Most languages are more transparent in these sorts of compounds.

  • 1
    "English is also fairly unusual among the world's languages in that it doesn't like using regular English compounds" Only reading this did it occur to me that I really have internalized a bizarre thing as normal. Without very specific resources, it would seem almost impracticable, let alone desirable, for people to go find foreign words which their fellow speakers would not use whenever they wanted to make new words for their fellow speakers to learn.
    – Vir
    Jan 14, 2022 at 0:53

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.