I’ve been looking for a long time, but I still don’t understand how to get the stem of an Ancient Greek noun that belongs to the first declension.

Almost every book says that for example -ᾱ, -η, -ᾰ (nom.sg.) are endings of first declension. And at the same time the first declension is called an α-declension because the stem of these nouns ends in -α. Which is not the same if I’m not mistaken.

But to get a stem for declension of a word we need to drop off the ending of gen.sg. How is it possible for stem to end with alpha if I drop off, for example, -ας in θύρας the stem is θύρ. So what is really a stem of first declension noun?

  • 1
    Don’t forget that the genitive form θύρᾱς has a long , even though the underlying ending is -ες (or -ος) – the two get mashed together into one vowel. Contractions like this can often make it tricky to untangle stems from endings. (Your question isn’t really about linguistics as such, but questions about Ancient Greek are considered on-topic over at Latin Language, so I’ve flagged it to be migrated there.) Commented Jul 17, 2023 at 20:19
  • 1
    Linguistically speaking this is a matter of opinion. τῑμή could be the stem, or it could be τῑμ, or maybe τῑμα. Each decision entails a different answer to the question "what is the case ending?". Your question is really not about the linguistic analysis of Greek, it is about the nomenclature of traditional Greek grammar.
    – user6726
    Commented Jul 17, 2023 at 21:58

2 Answers 2


It is certainly true that diachronically, the -ᾱ/-ᾰ(/-η) at the end of the nominative singular is part of the stem, and the nominative singular is endingless.

Pretty much all modern synchronic grammars of Ancient Greek will also tell you the former, but may disagree on the latter; the Cambridge Grammar of Classical Greek, for example, says:

4.1 The first declension is also known as the a-declension, since it consists of nouns with a stem ending in an a-sound (this sound is considered to be part of the endings, →2.4 n.1).

So the -ᾱ/-ᾰ is part of both the stem and the ending to them (while to others—including most university-level grammars I've seen—the nom. sg. has a zero ending). Presumably this is confusion brought over from Latin grammars, where the -a of the first declension disappears more often than in Greek and where the notion of contraction is more alien.


This is a great question! What is the stem of a noun? How do we define the stem of a noun?

Let's start with some vocabulary.

The base of a noun is the form we get when we drop the genitive singular ending. So the base of θύρᾶ, θύρᾶς is θύρ-.

The stem of a noun is the base of the noun plus its characteristic vowel. So the stem of θύρᾶ, θύρᾶς is θύρᾶ-, since the characteristic vowel is ᾶ.

These definitions of "base" and "stem" are consistent in both Latin and Ancient Greek.

We get the base of a noun by dropping its genitive ending, and we get the stem of a noun by adding its characteristic vowel to its base.

Now, in the first declension of Attic Greek, there are two large categories of nouns: ᾶ-stems and η-stems. θύρᾶ ("door") is an example of an ᾶ-stem noun. έπιστολή ("letter") is an example of an η-stem noun.

One of the nice advantages of having the word "stem" (in addition to the word "base") is that we can categorize nouns according to their stem. For example, some first declension nouns in Attic Greek are ᾶ-stem nouns, and some are η-stem nouns.

It's a little difficult to categorize nouns according to their base, isn't it? For example, the base of θύρᾶ is θύρ-, and the base of έπιστολή is έπιστολ-. But how can we create categories from these bases?

We can categorize nouns from vowel declensions according to their stem vowel. Some examples are ᾶ-stems and η-stems from the first declension, and ο-stems from the second declension.

Getting back to your question, I think that the key insight is understanding the difference between a base and a stem. The base of a noun is the form we get by dropping the genitive singular ending. The stem of a noun is the base of the noun plus its characteristic vowel. [1] [2]

In Attic Greek, first declension nouns often have stems ending in -ᾶ or -η. The base of θύρᾶ, θύρᾶς is θύρ-, which we get by dropping the genitive singular ending. The stem of θύρᾶ, θύρᾶς is θύρᾶ-, which we get by adding the characteristic vowel ᾶ to the base.


[1] According this definition of "stem", we can say that if a noun does not have a characteristic vowel (a stem vowel, that is) then its stem is identical to its base.

[2] Words like θύρᾶ, θύρᾶς have a stem vowel of α. Words like έπιστολή, έπιστολῆς have a stem vowel of η. What about words like θάλαττα, θαλάττῆς, which have a stem vowel of α in some forms and a stem vowel of η in other forms? We can say that words like θάλαττα, θαλάττῆς have a hybrid stem. One theory about the first declension in Attic Greek is that it started off with one set of endings (α endings) and then native speakers started to substitute η for α in the endings of some words because it was attractive to them, or convenient.

  • 1
    The two categories of first-declension nouns people tend to distinguish are the long-α stems (-ᾱ(ς), which includes -η(ς) because long α regularly becomes η in Attic except after ι, ε, and ρ; θύρα is in this category) and the short-α stems (-ᾰ; e.g. Μοῦσα). Either way, you cannot write ᾶ for ᾱ—the circumflex accent is not interchangeable with the macron.
    – Cairnarvon
    Commented Jul 20, 2023 at 7:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.