English is rule-based. I repeat: Like all languages, English is rule-based.
A tweet recently appeared in my Twitter feed in which Merriam-Webster, the dictionary, touted the nonsense that English rejects order.
“Because English rejects order. It actively opposes it. It lures order into an empty house and sets a series of Home Alone-style traps to hit it with paint cans and light its hair on fire. #AskMW”
Because English rejects order. It actively opposes it. It lures order into an empty house and sets a series of Home Alone-style traps to hit it with paint cans and light its hair on fire. #AskMW https://t.co/2wYGbG6yyu
— Merriam-Webster (@MerriamWebster) April 30, 2020
The absurd claim came in reply to a question about the plurals of <thief>, <knife>, and <chief>. One thief, two thieves. One knife, two knives. One chief, two chiefs. The noun <chief> affixes the <s> suffix to form the plural, which is a prototypical plural in Modern English. One cat, two cats. One hug, two hugs. One chocolate, two chocolates. The nouns <thief> and <knife> experience a consonant change plus the affixation of the <s> suffix to form the plural. Why?
At first glance, the plural forms of <thief>, <knife>, and <chief> might appear disorderly. But the truth is easy to uncover.
Using Etymonline, I find that <thief> comes from Old English <þeof> meaning “thief, robber” and <knife> comes from late Old English <cnif>. The letter <þ> was later rewritten as <th> and <cn> as the digraph <kn>. I also find that <chief> comes from Old French <chief> around 1300.
Even if you know nothing about the history of the English language, you can easily see that the histories of the three words in question differ. <thief> and <knife> are older, developing from Old English forms. The year 1066, when the Normans invaded English, marks the end of the Old English period (although languages exist on a continuum with no definitive start or end dates between periods). <chief> is younger, arriving in English from Old French around 1300.
Words borrowed more recently tend to use prototypical morphology. For example, the noun <fjord> comes from Norwegian <fjord> from Old Norse <fjörðr> meaning “an inlet, estuary” around 1670s. Rather than using the Norwegian plurals <fjorder, fjordene, fjordar, fjordane>, the English <fjord> affixes the <s> suffix to form the plural <fjords>. The younger <chief> likewise affixes the <s> suffix to form the plural <chiefs>.
<thief> and <knife> are older words. Understanding the plural forms requires delving into Old English phonology. Relevant to <thief> and <knife> is the knowledge of Old English allophones. Allophones are a set of speech sounds that represent a single phoneme. For example, in Modern English, the alveolar flap [ɾ] is an allophone of /t/ and /d/. Both [mɛtəl] and [mɛɾəl] are possible pronunciations of <metal>, and both [mɛdəl] and [mɛɾəl] are possible pronunciations of <medal>. The [ɾ] is an allophone, which is why <metal> and <medal> can be homophonic. Is [mɛɾəl mɛɾəl] a medal made of metal (<metal medal>) or metal for medals (<medal metal>)?
In Old English, the fricatives /f/ had the voiced allophone [v] between vowels or voiced consonants. The nominative plural of <þeof> was <þeofas>. In the singular, the <f> spelled [f]. In the plural, the /f/ occurred between two vowels and thus spelled [v]. Likewise, the nominative plural of <cnif> was <cnifas>. In the singular, the <f> again spelled [f]. In the plural, the /f/ occurred between two vowels and thus also spelled [v]. As Old English developed into Middle English, the plurals of <þeof> and <cnif> remained. [v] stopped being an allophone of /f/. The singular forms were thus spelled <thief> and <knife> and the plural <thieves> and <knives> because, although the primary function of English spelling is to represent meaning, sounds do matter.
Other words from Old English similar to <thief~thieves> and <knife~knives> include <wif~wifian/wife~wive~wives>, <lif~lifian/life~live~lives>, <leaf~leafa/leaf~leaves~leaved>, <cealf~cealfian/calf~calve~calves>, and <wulf~wulfas/wolf~wolves>.
Nouns that end in <f> that take the <s> suffix in the plural often came into English later. Regularization, which refers to the language change in which irregular forms are replaced by regular forms, can also occur. For example, <eage> meant “eye” in Old English. The plural form was <eagan>. Old English <eage> developed into Middle English <eie> whose plural form was <eien>. Over time, regularization occurred, and the Modern English plural of <eye> is <eyes> rather than <*eyen>. Regularization, however, has not affected the plural of <ox>: <oxa~oxan> -> <oxe~oxen> -> <ox~oxen>.
That the plurals of <thief> and <knife> are <thieves> and <knives> while the plural of <chief> is <chiefs> does not mean English rejects order. Rather, the words entered the language at different times. Both <thief> and <knife> come from Old English prior to 1066 when [f] and [v] were allophones of /f/. <chief> is younger, arriving around 1300 from Old French.
Always remember that English is rule-based. If you think a form is a disorderly, irregular exception, you have not investigated the word enough yet.