When nothing means something:
In morpheme-based morphology, a null morpheme is a morpheme that is realized by a phonologically null affix (an empty string of phonological segments). In simpler terms, a null morpheme is an “invisible” affix. [Wikipedia]
Wikipedia says that examples in English include hiatus and co-operation. I didn’t quite get how they have null morphemes. (Well, it is invisible, how am I supposed to see?)
I know that hiatus is the slight pause between two vowels coming together but not in the same syllable. Hiatus itself contains hiatus also. So, is that slight pause a null morpheme? But why do we need to explain with a null morpheme?
Another odd example in English is the invisible notion of singularity and plurality:
- sheep = sheep + -Ø = ROOT (“sheep”) + SINGULAR
- sheep = sheep + -Ø = ROOT (“sheep”) + PLURAL
I know that the plural form sheep was leveled with the singular in Old English. But why is the need of null morpheme here?
Can you please shed a light on this? Also, is null morpheme a well-accepted concept?
The Wikipedia examples are not relevant. They are examples of insertion of a glide to prevent two vowels from being adjacent: specifically, a /j/ in hiatus and a /w/ in co-operation. That doesn’t have anything to do with null morphemes.
As for the main question, “is null morpheme a well-accepted concept?”, yes, it is, but it depends on your theory of inflectional morphology.
In incremental theories of morphology, inflectional morphology is information-increasing: each morpheme adds features to the stem. In this kind of approach, the feature PLURAL could be seen as the function of a morpheme added to a stem that is unspecified for number. For many words, the morpheme is realized as /s/ or /z/, e.g. cats, dogs. For others, the “morpheme” is some kind of floating vowel quality effect, as in mouse ~ mice. For examples like sheep, the morpheme is phonologically null. This is a necessary step if you believe that plurality can only be the result of adding a morpheme.
In realizational theories of morphology, on the other hand, lexical stems themselves are associated with a set of morphosyntactic features, which license the addition of certain affixes. In this kind of approach, there’s no need for a phonologically null morpheme (though you could have one if you wanted).
Most popular theories of inflectional morphology these days are realizational rather than incremental, e.g. Distributed Morphology, A-morphous Morphology, Paradigm Function Morphology. Hope that helps, and let me know if anything needs clarification!