Why isn't "giraves" the plural of "giraffe" like "wolves" is for "wolf"?

Question

The plural of giraffe, according to Merriam Webster and some other dictionaries I checked, is "giraffes".

Normally when the final sound of an English word is F, its plural ends in V sound. But not always.

Examples:

• knife → knives
• wife → wives
• life → lives
• wolf → wolves
• leaf → leaves.

But giraffe → giraffes?

My best guess would be that the former words are all native English words while "giraffe" is not.

I searched "wolf" in Etymology Dictionary and it says:

from Old English wulf

For "life" it says:

Old English life (dative lif)

For "wife" it says:

Middle English wif, wyf, from Old English wif

For "knife" it says:

late Old English cnif

BUT for "giraffe" the entry is quite different:

long-necked ruminant animal of Africa, 1590s, giraffa, from Italian giraffa, from Arabic zarafa, probably from an African language. Earlier Middle English spellings varied wildly, depending on the foreign source, and included jarraf, ziraph, and gerfauntz, some apparently directly from Arabic, the last reflecting some confusion with olifaunt "elephant.

The modern form of the English word is attested by c. 1600 and is via French girafe (13c.). Replaced earlier camelopard (from Latin camelopardalis), which was the basis form the name of the "giraffe" constellation Camelopardalis, among those added to the map 1590s by Flemish cartographer Petrus Plancius.

So why is the plural of giraffe different than the other similar words? Why is it not "giraves"?

Answer 1

You're on the right track.

Old English had a phonetic property called intervocalic voicing, where a fricative consonant—f, þ, or s—became voiced when it occurred alone between two vowels: sē stæf (“the staff”) was pronounced [seː ˈstæf], but þā stafas (“the staves”) was [θaː ˈsta.vas]. This was, in fact, the only way to get a v, z, or voiced th sound in English. The liquids (l and r) did not cancel this voicing, so þā wulfas (“the wolves”) was spoken as [θaː ˈwul.vas] as if the l were not there.

Now, OE nouns were heavily inflected, and many nouns gained extra syllables on the end in some of their forms, changing the phonetics of the final consonant of the stem to give us þæt wīf ([θæt 'wifː], “the wife”) in the nominative case but þæs wīfes ([θæs 'wiː.ves], “the wife’s”) in the genitive. However, this was only sometimes true of the plural form: the plural of þæt wīf was actually just þā wīf.

How, then, did the intervocalic /v/ make its way into the plural form of the word? For that, you can blame Middle English, which, as a part of its many, many grammatical simplifications, made -es into the standard pluralizing suffix for all “strong” nouns. This caused the f at the end of many words to become voiced. Although the Great Vowel Shift has since silenced the /e/, the /v/ survives to this day in many words. If a word ends with an f sound but doesn’t have a -ves plural, this could mean any of a number of things:

• The word might have been adopted into the English language after it had already lost universal intervocalic voicing, as is the case for your giraffes.
• The word might have had a -ves plural at one point, but have since lost it. When was the last time you encountered the word turves?
• The word's ending might not have been subject to intervocalic voicing in the first place. This is true of dwarfs, which, in its OE form, never contained an f to begin with. (The spelling “dwarves” was popularized by Tolkien—he liked it for its similarities with “elves”, but called it “a piece of private bad grammar”.)

Also interesting to note: since intervocalic voicing applied to þ as well as f, many modern English words ending in an unvoiced -th also have plural forms in which their final consonant sound becomes voiced before the -s. You can't tell by the spelling, since there is no voiced counterpart to th, and it varies by dialect, but it's there: mouths, moths, oaths, baths, paths...