Part forty-six of a tour through Greek inflectional morphology to help get students thinking more systematically about the word forms they see (and maybe teach a bit of general linguistics along the way).

We saw in Part 42 that the aorist middle endings were:

  • -σθαι
  • -μην
  • -σο (often with loss of sigma and subsequent contraction)
  • -το
  • -μεθα
  • -σθε
  • -ντο


  • preceded by alpha
  • preceded by a theme vowel ε/ο
  • affixed directly

which correspond to our classes:

  • alphathematic (first aorists)
  • thematic (second aorists)
  • root

Note again what we said in the previous post: this is just a classification based on the distinguisher paradigms and there are other ways of categorizing aorist forms.

Does this cover all aorist middle indicatives and infinitives in SBLGNT? Are there any words in more than one class or with more than one stem? And what are the counts for the different classes and dominant lemmas or forms within each class?

We’ll cover that here.

There is one form that doesn’t match our distinguisher patterns and that’s ξυρᾶσθαι in 1 Cor 11.6. This seems to be an error in the MorphGNT tagging, though. It’s clearly a present (imperfective) infinitive not an aorist (perfective) infinitive and so is not relevant here.

Now in terms of multiple stems, we do have an augmentation difference with ἐργάζομαι. We find both ἠργασ- and εἰργασ-.

And in terms of words that seem to appear in more than one class, we have these forms of ἐξαιρέω:

  • ἐξελέσθαι, which is clearly thematic
  • ἐξειλάμην and ἐξείλατο, which are clearly alphathematic

We also have these forms of ἀποδίδωμι (which we’ve brought up before):

  • ἀπέδετο
  • ἀπέδοσθε
  • ἀπέδοντο

We would expect the forms to follow the root pattern. ἀπέδοσθε is unambiguously root, ἀπέδετο is unambiguously thematic. ἀπέδοντο could be taken to be root or thematic. For the purposes of the counts below, we’ll take ἀπέδοντο to be root.

Note that ἐκδίδωμι only appears as ἐξέδετο which is also thematic so there’s definitely some reanalysis going on with the δίδωμι compounds.

Here are the total counts across classes for aorist middles in SBLGNT:

class # lemmas # tokens # hapakes
alphathematic 109 393 49
thematic 20 320 10
root 14 39 3

(yes, “hapakes” is a running joke, equivalent to calling them “the onces”)

ἀπο-δο is the only root ending with ο and the rest of the root endings are θε and compounds:

  • θε
  • ἐκ-θε
  • ἀπο-θε
  • ἐπι-θε
  • δια-θε
  • κατα-θε
  • παρα-θε
  • προσ-θε
  • προ-θε
  • συν-θε
  • ἀνα-θε
  • προσ-ανα-θε
  • συν-επι-θε

The thematics come from ten familes and are:

  • γεν (γίνομαι)
  • παρα-γεν
  • πυθ (πυνθάνομαι)
  • ἐξ-ελ (-αἰρέω)
  • συμ-βαλ (-βάλλω)
  • περι-βαλ
  • ἀνα-βαλ
  • ἀνα-σχ (-έχομαι)
  • ἀπ-ολ (-όλλυμι)
  • συν-απ-ολ
  • ἐκ-δο (-δίδωμι)
  • ἀπο-δο
  • συλ-λαβ (-λαμβάνω)
  • κατα-λαβ
  • ἐπι-λαβ
  • προσ-λαβ
  • ἀντι-λαβ
  • ἀφ-ικ (-ικνέομαι)
  • ἐφ-ικ
  • ἐπι-λαθ (-λανθάνομαι)

But note that γίνομαι alone makes up 269 out of the 320 tokens!

Now by person/number:

  alphathematic thematic root
INF 63 45 3
1SG 23 16 3
2SG 8 2 1
3SG 208 227 16
1PL 12 0 0
2PL 22 4 2
3PL 57 26 14

Because μι verbs have root forms throughout the middle (not just in the infinitive like in the active in Hellenistic Greek) we don’t get the disproportionately high INF root counts that we did in the active.

The 3SG expectedly dominates. This is particularly true in the thematic, in large part due to ἐγένετο. But in addition to dominance of the 3SG by ἐγένετο, all the 2SG and 2PL thematic aorist middles are γεν and most of the INF, 1SG and 3PL are too, as show here in our table showing dominant forms:

  alphathematic thematic root
INF   γενέσθαι 36/45 καταθέσθαι 2/3 ἀποθέσθαι 1/3
1SG   ἐγενόμην 12/16 προεθέμην 1/3 προσανεθέμην 1/3 ἀνεθέμην 1/3
2SG κατηρτίσω 2/8 ἠρνήσω 2/8 ἐγένου 2/2 ἔθου 1/1
3SG   ἐγένετο 201/227 ἔθετο 7/16
2PL   ἐγένεσθε 4/4 ἔθεσθε 1/2 ἀπέδοσθε 1/2
3PL ἤρξαντο 19/57 ἐγένοντο 14/26 ἔθεντο 4/14

As with the actives, there is greater lexical variety amongst the alphathematics than amongst the thematics.

class token-lemma ratio % hapakes
alphathematic 3.61 45.0 %
thematic 16.00 50.0 %
root 2.79 21.4 %

The top 5% of alphathematic lemmas make up 32.1% of the tokens whereas the top 5% of thematic lemmas make up a whopping 84.1% of tokens. For the actives, recall the numbers were 44.1% and 60.7% respectively.

In the next couple of posts we’ll look at the (θ)η aorists (often misleadingly called aorist “passives”).