Most of the differences in how Quenya nouns are inflected depend on whether the noun ends in a vowel or consonant. Thus, Quenya nouns can be divided into two broad classes: vocalic nouns and consonantal nouns. Most people who study Quenya add a third noun class, e-nouns, for vocalic nouns that end in -e. Tolkien himself acknowledged the special behavior of e-nouns: in the famous Plotz letter, he gave declensions for both a vocalic noun (cirya “ship”) and an e-noun (lasse “leaf”).

Conceptual Development: In the Early Qenya Grammar of the 1920s (EQG) Tolkien gave the main noun classes as vocalic and consonantal: “Nouns. Are divided into two classes: (A) vocalic; (B) consonantal (PE14/42)”. In the Declension of Nouns from the early 1930s, he had a much more elaborate system of noun classes, with numerous variations based on (a) the final vowel or (b) the final consonant (PE21/2-3). He broke nouns up into the following classes and subclasses:

• Vocalic:
  • Nouns originally ending -ā.
  • Nouns originally ending -ē.
  • Nouns originally ending -ō.
• Semivocalic:
  • Nouns originally ending -ī.
  • Nouns originally ending -ū.
• Consonantal:
  • Polysyllabic nouns ending in a consonant.
  • Monosyllabic nouns ending in a consonant.
  • Various irregular monosyllables, mostly involving lost or vocalized final consonants.

Too add to this complexity, many of the consonantal nouns allowed both long suffixes (e.g. -unta/-ullo/-esse) and short suffixes (e.g. -ta/-lo/-se), and the short suffixes had various specialized assimilated forms that depended on the final consonant of the noun stem. Some of the details of these specialized forms are discussed in the entries on specific noun classes and noun cases.

In Tolkien’s later writing, there is no sign of a system with this level of complexity in the published corpus, but there are a few later examples that indicate Tolkien played with more complex inflections in his later writing. See the assimilated locative for one example of this.