I was just looking at a Zulu word entry in Wiktionary that implied it was made from a prefix and a suffix, but there was nothing between them.

Now this could just be sloppy editing of Wiktionary but either way it got me thinking about whether this is a valid word-formation strategy used by any languages.

When I first asked this question I didn't word it in a way that clarified I was asking about word formation. It turns out there are words in several languages which can be analysed on the furface into only affixes, but the examples so far were not originally formed that way, but my more complex processes of adding and dropping morphemes at various stages.

So is this affix-only manner of word formation known to occur? Is it widespread amongst languages?

  • 2
    Ever shopped at a Superette? – jlawler May 23 '13 at 13:19
  • 3
    @jlawler: No - is that a word? Before I look it up though I would assume it has a more complex history like market -> supermaket -> superette or somesuch ... – hippietrail May 23 '13 at 13:46
  • 1
    @jlawler super- is root here. (even if prefix-derived). There is even separate word "super" meaning "excellent, exceptional". – Anixx May 23 '13 at 17:32
  • 2
    Too bad you didn't tell us what the Zulu word is. If you had said "affixes" (only) then I'd say yes, certain demonstratives in a number of Bantu languages. But prefix is defined relative to the root, likewise suffix, so anything with a prefix or a suffix has a root. – user6726 Sep 23 '16 at 2:08
  • 2
    @Anixx: No. In THIS word, the root is "mercado", and "minisuper" is a shortened form of "minisupermercado" – Flimzy Sep 23 '16 at 14:53

Yes. One such word is Russian word вынуть "to take out". Here the вы- is prefix, -ну- is suffix and -ть is ending.

The old form of this word was вынять which had the root -ня-, but later the root was re-analyzed as suffix by analogy with other words (сунуть, дунуть).

  • This is interesting but I wonder how I can amend the wording of my question best to clarify that I don't want to ask about indirect derivations and re analyses but just words actually derived directly from combination only of affixes in the first place. – hippietrail May 23 '13 at 17:27
  • Again, that depends on what you mean by "derived directly from". If you can specify what tests could distinguish "direct derivation" from something else, that would be helpful. – jlawler May 23 '13 at 23:48
  • I'm very happy to improve the wording of my questions with the help of others in the comments. Sometimes the idea for a question is clear but getting it into words and keeping it sharp is not so easy for me. I guess tests such as active word formation known to involve just adding prefix(es) and suffix(es) or known etymologies that didn't go through intermediate steps like the superette and вынуть so far suggested. We can analyse both as currently consisting of morphemes which are usually used as adfixes but we know they didn't come about just by adding two adfixes. – hippietrail May 24 '13 at 5:20
  • 1
    @hippietrail in superette super is root. – Anixx May 24 '13 at 10:25
  • @Anixx: Exactly. Thanks for analysing rightly (-: – hippietrail May 24 '13 at 10:34

In Esperanto there are some words of this kind, e.g., malina "male" composed of mal- "negation, opposite of" and -ina "feminine"

More examples can be found in this answer: https://esperanto.stackexchange.com/a/407/7

  • 1
    Idiotic idea. In any natural language it would mean "feminine nay-sayer". If -ina means feminine, then it should mean feminine person everywhere. – Anixx Sep 23 '16 at 17:48
  • 1
    If -ina does not mean a feminine person, it is used as root in this case. – Anixx Sep 23 '16 at 17:58
  • @Anixx: I think I have zo explain the mal- prefix better:as "opposite of". -in- is a Movierungs-Suffix for creating femal versions of male nouns (like and derived from the German suffix -in with the same meaning) – jk - Reinstate Monica Sep 25 '16 at 7:40
  • @Anixx: It's not helpful for this community to dismiss what is a fact as idiotic. I wish I could downvote your comment. – jogloran Sep 25 '16 at 9:16

No. It is possible for an affix to become a root, as in the mentioned case of super, and then to derive new words by attaching prefixes and/or suffixes to it, but that's a different case.


How about antis ("those opposed to something")? Unlike the example of "superette" in the comments, where the prefix is shortened from "supermarket," with "antis" the prefix isn't taken from any particular word.

  • It is derived from PIE root a̯ent- "end". PIE used roots as "prefixes" in most cases (or one can consider such words having two roots). In this case we simply have the root with a suffix. – Anixx Sep 23 '16 at 17:52

Mono-ward is a great example of affixes realising roots where either its prefix or suffix figures as root-to.

  • 1
    Welcome to Linguistics! Please add some more detail to your answer so that it's not just a one-line comment. If that's not possible, I can convert your answer to a comment until you have enough reputation to do so yourself. – Alenanno Apr 24 '20 at 2:25

A bit of a cheap shot, but what about 'unable'? Prefix and Suffix with nothing inbetween.

  • 4
    'able' is not a suffix, but a root - As in "an able mechanic" (slightly archaic now, but still..) – robert Jul 15 '15 at 15:18
  • @robert interestiongly, this word can be analyzed differently depending on stress. If the stress is on the first syllable, the root is "un". Quite artificial, though. – Anixx Sep 23 '16 at 17:55
  • I don't think it has any relation to stress. Portuguese and other Romance languages are full of words in which the stress is on suffixes - and even desinences. We even have a word for that - arrizotônica, ie, literally, "stress not in the root". – Luís Henrique Sep 25 '16 at 12:43

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