Pa'al Past: Regular Roots (כתבתי, כתבת, כתב…)

The pa'al past hangs a personal ending on the back of the stem. Nothing goes in front, nothing is inserted, and the endings are the same ones every other binyan uses — so this paradigm is the one that pays for itself most quickly. Learn it on כ־ת־ב and you have learned the endings for the whole language.

The full paradigm

Two counts worth doing yourself. Nine cells, eight distinct written strings — כָּתַבְתָּ and כָּתַבְתְּ are both כתבת on the page, and only the voice separates them. And there is no separate feminine plural: כָּתְבוּ covers הֵם and הֵן alike. The past simply does not make that distinction; the future and the imperative do, which is where the temptation to invent a form like *כתבנה comes from.

Three more verbs, same endings

Notice that ל־מ־ד behaves exactly like כ־ת־ב here. Its future is the odd one — אֶלְמַד with a patach where כתב has a cholam — but the past knows nothing about that split. Whatever a pa'al verb does in the future, its sound-root past is a–a.

The stem shifts, and the bare spelling hides it

Read down the הִיא and הֵם rows: כָּתְבָה katvá, כָּתְבוּ katvú. Not katáva, not *katávu. When a vowel-initial ending arrives, the second stem vowel is squeezed out and the two consonants collide: *kat-vá.

That is invisible in כתבה and כתבו, which look like כתב with a letter added. The niqqud is the only place the change is written down, and the pronunciation is the only place it is heard — which is why a learner who reads the paradigm silently ends up saying katáva for months.

The same reduction hits the second-person plural from the other side: כְּתַבְתֶּם loses its first vowel instead, giving ktavtém.

The verb already says who

Look at what ־תִּי and ־נוּ do that כותב cannot: they name the person. So כתבתי מכתב is a complete sentence with a specified subject, and אני כתבתי מכתב adds emphasis or contrast rather than information.

Both are entirely ordinary. Israelis use the pronoun constantly in the past — this is not Spanish, where the bare verb is the default and the pronoun is marked. Hebrew simply allows either, and drops the pronoun most readily in a continuing narrative, once the subject has been established.

The third person is different: כתב by itself says only "he wrote", and unless the man is already on the table you have to name him. That is why הוּא and הִיא survive in the past far better than אֲנִי and אֲנַחְנוּ do.

Two register facts the tables leave out

כְּתַבְתֶּן is receding, not gone. Colloquial Hebrew increasingly lets כתבתם address a group of any gender, and that is what you will hear most of the time. But do not file this with תכתובנה: אתן is a fully live pronoun, the ־תֶּן ending is transparent, and speakers addressing women only — a teacher to a class of girls, a woman to her friends — do produce למדתן, הלכתן, כתבתן. Learn it as a form you may well need to understand and can safely use; expect כתבתם to cover for it.

ktavtém is not how the word is said. The normative form stresses the ending and reduces the first vowel; nearly every Israeli restores the stem vowel and pulls the stress back: katávtem, shamártem, lamádtem. Both are spelled כתבתם, so the page never tells you which one you are looking at. Newsreaders and formal speech keep ktavtém.

The asymmetry that catches everyone

The past marks person; the present does not. So the licence you are given here stops at this tense.

The first clause needs no אני because ־תי supplies it. The second cannot manage without one, because עובד is a participle and participles have no person. The full argument is on the pa'al present.

Common Mistakes

1. Inventing a feminine plural. The future has תכתובנה and the imperative has כתובנה; the past has nothing of the kind.

2. Keeping the stem vowel before ־ָה and ־וּ. The spelling gives you no warning, so this one has to be learned by ear.

3. Reading כתבת aloud as the wrong person. One string, two words — and choosing wrongly misgenders the person you are speaking to.

4. Carrying the dropped pronoun into the present. This is the real cost of learning the past first.

Key Takeaways

• The past is a suffix conjugation: ־תִּי, ־תָּ, ־תְּ, zero, ־ָה, ־נוּ, ־תֶּם, ־תֶּן, ־וּ, identical across all seven binyanim.
• Nine cells, eight written strings — כתבת is both katávta and katávt, and only the voice separates them.
• There is no distinct feminine plural. כָּתְבוּ serves הם and הן; a form like *כתבנה does not exist in the past.
• The stem loses its second vowel before the vowel-initial endings: katvá, katvú — never katáva. ־תֶּם reduces from the other end: ktavtém.
• Where the stem loses a vowel, the ending takes the stress — katvá, katvú, and normatively ktavtém; where the stem keeps its vowels, the stress stays put: katávti.
• The verb names the person, so כתבתי is a complete sentence and אני כתבתי is emphasis. Israelis use both freely.
• כתבתן is receding, not dead: כתבתם increasingly covers any group, but speakers addressing women only still produce it. The normative ktavtém is heard as katávtem in ordinary speech.
• The pro-drop licence is specific to the past and the future. The present has no person marking and always keeps its pronoun.