How do broken plurals work in Afro-Asiatic languages?

In many Afro-Asiatic languages, particularly within the Semitic branch, nouns are categorized into two types of plural formation: "sound" plurals and "broken" plurals.1

While "sound" plurals are formed by adding a suffix to the singular stem, broken plurals are indicated directly through changes to the internal pattern of the stem itself. This process involves modifying the vowels or consonants within the word rather than relying on external endings.1

Key characteristics of broken plurals include:

• Systematic Patterns: Some broken plurals follow predictable configurations. For example, in Arabic, the pattern CaCāCiC- is used for many stems containing four consonants, such as thaʿlab (fox) becoming thaʿālib (foxes).1
• Isolated Stems: Other broken plurals function as independent collective nouns, where the plural form may appear quite distinct from the singular, such as ḥimār (donkey) and ḥamīr (donkeys).1
• Variability: Many nouns may have more than one possible plural form, and different plural patterns can sometimes distinguish between homonymous singular stems.1
• Broad Distribution: While most prominent in Arabic, Ethiopic, and South Arabian languages, broken plurals are also found in other Afro-Asiatic branches, often involving internal inflection or combinations of vowel changes and consonant repetition.12