One moose, two moose, red moose, blue moose

On my post noting that Moose were Moose, though Elks were Elks Kathie commented:

"Moose" can be plural as well as singular. Ditto for "sheep," "deer," "fish," "caribou"/"reindeer" -- heck, even "elk"! I shall defer to your superior linguistic training, that you might expound further on such collective (defective?) nouns?

So I shall. Why not?

These aren't really defective nouns. A defective word is missing some part of its paradigm (as, for instance, modal auxiliaries such as must have no infinitive, no past tense, no participle, and no third-person singular). English does have defective nouns, mostly things like cattle and measles that have no singular.

But these nouns (such as deer, fish, and sheep) have a plural - it's one deer is, two deer are - so they aren't defective. (Note how you can't say one cattle.) They simply have a plural which is morphologically identical to the singular. This is called a zero plural (meaning 'zero affix') and is a direct descendent of the Old English neuter gender. For instance, only the form of the determiner shows that þæt wīf (the woman*) is singular whereas þā wīf (the women) is plural. So, too, þæt dēor and þā dēor, or þæt scaep and þā scaep.

Interesting note: in Old English it appears that neuter was often used for superordinates or broad categories. For instance, dēor meant wild animal, and fisc was the catchall for all kinds of fish. Note that this doesn't apply to most domestic animals (I suppose sheep have no subordinates plus there was ēowu, ewe, and ramm, ram). Also, because languages aren't mathematics, fugul which meant bird, fowl was masculine and had a plural of fugules...

This zero-ending pattern was extended in Middle English to most types of game animals, birds, fish - hence five pheasant or a lot of trout - and then in modern times to newly-named or -discovered animals such as moose or caribou.

*(Yes, woman was neuter. So was - for many, still is - child. Grammatical gender is not natural gender.)