I w, II weak, and irregular verbs: verbs which do not follow the strong paradigm

So far in our review of weak verbs, we have seen that verbs whose first or third consonant is ʾ (aleph) essentially follow the same paradigm as strong verbs, with some phonological modifications. In this final page we will survey a few varieties of verb that obey their own rules.

When you have read this page, you might like to test your understanding by trying Cuneiform exercise 8.

I w ("first w") verbs

Verbs whose first consonant is w don't behave quite like strong verbs or like other I weak verbs. The following rules are particular to I w verbs:

• In the G stem and Š stem the vowel just before the w often changes and/or lengthens, for instance, ēkallam uššab (*iwaššab), "he dwells in the palace". See the paradigm table below and the Sound rules page for full details of all the changes.
• But in the N stem and D stem the w behaves like a strong consonant, for instance D-stem nišī ina ālim uwaššib, "he settled (lit. made dwell) the people in the city".
• The vowel class is i/a for verbs denoting actions and i/i for verbs that denote states of being.
• In forms where the middle consonant isn't doubled, suffixes beginning with a vowel can cause the immediately preceding syllable to collapse. The vowel before the collapsed syllable also shortens, for instance, ūbil (*uwbil), "he brought", but ublam (*uwbilam), "he brought here".

Here is a short paradigm table for the four main stems, showing just three of the most common persons. The paradigm verb here is walādum, "to give birth (to)" (so all of its persons are necessarily feminine in the G stem). In the N stem it means "to be born", and in the D "to act as midwife" (literally "to make give birth"). The Š stem means "to breed (animals)".

Note how the tense vowel is squeezed out of some forms with suffixes, as explained on the Sound rules page.

Look out for the doubly weak verb waṣûm, "to go out", which obeys I w rules at its front end and III weak rules at its rear end (e.g., G-stem preterite ūṣi (*iwṣiʾ), "he went out"; and Š-stem preterite ušēṣiam (*ušawṣiʾam, "he brought out", literally "he caused to go out here").

II weak ("second weak") verbs

There are two sorts of II weak verbs. In one variety, the aleph behaves like a strong consonant, so we can treat those just like strong verbs (e.g., ibaʾʾiš, "it is stinking", from baʾāšum, "to stink" (i/i)). The second variety is much more interesting. We need to put aside most of what we know about the morphology of strong verbs and treat them on their own merits.

These more interesting II weak verbs are sometimes called hollow verbs because it is a useful approximation to say they have a vowel in the middle of the root instead of a consonant. In other words, hollow roots consist of two consonants either side of a long vowel. For instance:

• dūk, "kill"
• kūn, "permanence"
• mūt, "die"
• rīš, "rejoice"
• šām, "purchase"
• šīm, "determine (fate)"

These are roots, not words. In order to turn them into verbal forms, we need some morphological rules. Most of them are beautifully simple. Here are the rules for the G-stem:

• The verb's subject is marked with the same prefixes and suffixes as any other verb.
• The preterite is formed simply by adding the subject marker to the hollow root (e.g., kalbam idūk, "he killed the dog").
• The perfect is formed by adding a t immediately after the first root consonant of the preterite (e.g., kalbum imtūt, "the dog has died").
• The present is formed by adding a present-tense a immediately before the final root consonant of the preterite (e.g., arīaš, "I am rejoicing"). Except when the root vowel is ī, the two vowels often contract to â, as in išâm (*išāam, "he is buying".
• When any suffix starting with a vowel is added to present-tense forms, the final root consonant doubles and the root vowel becomes short. The present-tense a, if there is one, disappears, for instance in iriššū (*irīašū), "they are rejoicing". This is the only tricky detail of II weak verbs.
• In the infinitive, the root vowel shortens before a long ā (e.g., šiāmum, "to determine (fate)"). Except when the root vowel is i, the two vowels contract usually contract to â, as in šâmum (*šaāmum), "to purchase". It is most useful to learn infinitives in their uncontracted form. This way you automatically memorise the root vowel and prevent confusion between fixing fate and going shopping!

Here, then, is a short paradigm table for II weak verbs in the G stem, giving the two most common persons for all three possible root vowels, using šâmum, "to purchase", šiāmum, "to determine (fate)", and duākum, "to kill".

As you can infer from the table, in transliteration and in cuneiform writing it is often difficult to distinguish between present and preterite forms of hollow verbs with root vowel a. Is i-ša-am a writing of išām, "he bought" or išâm, "he is buying"? It is impossible to know without further context.

The N, D and Š stems of II weak verbs are also formed through elegantly minimalist rules:

• N-stem verbs add an n immediately after the subject prefix of the G-stem form , for instance in iddâk (*indūak), "he shall be killed".
• D-stem verbs replace all subject-prefix vowels with u, double the first root consonant, and replace the root vowel with a long tense vowel: a for the present, i for the perfect and preterite, as in išdī ēkallim ukīn, "he established the palace's foundations".
• Š-stem verbs also replace all subject-prefix vowels with u and of course they add š immediately after the subject prefix> They also replace the root vowel with a long tense vowel: a for the present, i for the perfect and preterite, for example in awīlam ušdīk, "he had the man killed".
• In all three stems and in all three tenses, when any suffix starting with a vowel is added, the final root consonant doubles and the root (or tense) vowel becomes short. The present-tense a, if there is one, disappears, as in iddakkū (*indūakū), "they shall be killed".

Here is a short paradigm table for II weak verbs, giving third-person forms in the N, D, and Š stems. We use nadūkum, "to be killed" (the N stem of duākum), kunnum, "to establish" (the D stem of kuānum), and šurūšum, "to have (someone) rejoice" (the Š stem of riāšum).

For a more technical explanation of II weak verbs, which explains their morphology in terms of aleph, and for paradigms of the derived stems, you will need to consult a good Akkadian grammar.

Irregular verbs

There are a few verbs, such as edûm, "to know" (i/i), and išûm, "to have", that exist only in the preterite. This makes them relatively easy to deal with—and we can translated them with whatever tense makes most sense in the context. They take the form īdi (*iʾdiʾ), "he knew", in the singular and īdû (*iʾdiʾū), "they knew", in the plural. (Similarly īši, "he had", and īšû, "they had".)

The verb alākum, "to go", does not behave in quite the same way as other first weak verbs. Essentially, the aleph assimilates to the following consonant rather than lengthening the preceding vowel as expected. Like I w verbs it has class vowels i/a. Here is a short paradigm of its main stems and tenses, in the third person.

The verb izuzzum, "to stand", is also irregular. It uses tense vowels and doubles the final root consonant before a suffix beginning with a vowel. It occurs mostly in the G and Š stems (with two variants):

For the behaviour of other types of irregular verbs, various sorts of doubly weak verbs (with two weak consonants), and verbs with more than three root consonants you will need to consult a good textbook of Akkadian.

Content last modified on 02 Apr 2024.

Eleanor Robson

Eleanor Robson, 'I w, II weak, and irregular verbs: verbs which do not follow the strong paradigm', Knowledge and Power, Higher Education Academy, 2024 [http://oracc.museum.upenn.edu/saao/knpp/cuneiformrevealed/akkadianlanguage/iwiiweakandirregularverbs/]