How the Hebrew Verb Works

An English verb is a word you learn and then bolt endings onto: write, writes, writing, wrote, written. A Hebrew verb is not a word at all until three separate things have been assembled — a root, a בניין, and a tense slot — and each of them contributes a different part of what you finally say. This page shows the assembly line. Nothing here is a paradigm to memorise; it is the map you need before any paradigm makes sense.

One root, six words

Here is כ־ת־ב, the root for writing, run through five verb forms and one noun. Look only at the consonants: כ, ת, ב, in that order, in every single row.

Five of those are verb forms; the sixth is a thing you put in an envelope. Everything separating them is vowels and affixes, and the root כ־ת־ב is neutral between all of it.

The three layers

Every finite Hebrew verb is these three things stacked, and each layer answers a different question.

The middle layer is the one with no English counterpart. A בניין (binyán, "building") is a mould that decides not only how the verb sounds but what it means grammatically: כָּתַב is "wrote", נִכְתַּב is "was written", and the difference is carried entirely by the mould. Hebrew has seven of them, laid out on the seven binyanim.

Three tenses, one infinitive, one imperative — that is the whole list

Hebrew has exactly three finite tense slots: past, present, future. There is no perfect, no pluperfect, no progressive, no conditional tense. Add the infinitive and the imperative and you have the complete inventory of verb forms in the language.

That is not a simplification for beginners; it is the actual size of the system. English I wrote, I have written, I had written and I was writing all land on the single Hebrew form כתבתי, and context does the rest of the work.

No auxiliaries anywhere

English builds most of its verb forms out of two words: do you write, I am writing, I have written, it is written. Hebrew builds all of its forms out of one. There is no do-support in questions or negatives, no progressive "be", and no perfect "have".

The one thing that looks like an auxiliary is היה + present participle — היה כותב, "used to write". It is real, but it is a habitual and counterfactual construction, not a tense of the ordinary kind, and it never translates "was writing". It has its own treatment on the three tenses.

The present tense marks gender and number, never person

The Hebrew present is really the בינוני, the participle, and a participle agrees the way an adjective does: it has exactly four forms, and none of them tells you who.

כותב serves I write, you write and he writes alike, so long as the subject is masculine and singular. Two consequences follow immediately, and both surprise learners who met the past tense first:

• The present has to be told who its subject is. אני and אתה can be dropped in the past and the future, where the verb itself says who; drop them in the present and nothing is left to say it. Casual speech does leave out a first-person אני — הולך לישון, "off to bed" — but that shortcut can only ever mean I, because there is no other reading available.
• Gender agreement is not optional politeness. A woman saying אני כותב has made the same kind of error as she are.

Recovering a verb from an unpointed page

Israeli text prints no vowels, so the layers you have just met are invisible. This is the skill that replaces looking things up alphabetically.

הזמנתי is six letters. Take off the ־תי, which is the past-tense ending for "I". Take off the ה־ at the front, which belongs to the binyan הפעיל. What is left is ז־מ־ן — and that root gives you זמן "time", הזמנה "an invitation, an order", מזמין "invites" and מוזמן "invited". The verb was never filed under ה.

Now notice what made that hard. Six letters went in and only three came out, and nothing on the page marked which three. Hebrew's affix letters — א, ה, ו, י, מ, נ, ת — are also perfectly ordinary root letters, so a printed word gives you no way to tell a prefix from a radical without knowing the system. On top of that, the vowels that would have identified the binyan are simply not written.

This is why a Hebrew dictionary cannot be scanned the way an English one can. It is organised by root, so you cannot look up הזמנתי under ה, and you cannot even look it up under ז until you have done the stripping yourself — see looking a word up.

Common Mistakes

1. Putting the infinitive where a finite verb belongs. Apps and wordlists often gloss verbs by their infinitive (לכתוב "to write"), and learners then plug that form straight into a sentence.

2. Expecting the present tense to say who. Hebrew really does drop pronouns — but only in the past and the future, where the verb marks person. A bare participle can be stretched to cover a first-person I in casual speech (הולך לישון), and nothing else: it can never be made to mean you or she.

3. Leaving the participle unagreed. This is the single most frequent error in beginner Hebrew, because the English participle never agrees with anything.

4. Counting affix letters into the root. A vav or a hey inside a word is as likely to belong to the mould as to the root, and guessing wrong sends you to the wrong dictionary entry.

Key Takeaways

• A Hebrew verb is root + בניין + tense slot. The root gives the meaning, the binyan gives the vowels and the voice, the tense slot gives the person prefixes and suffixes.
• One root generates the whole family: כתב, כותב, יכתוב, לכתוב, נכתב — and the noun מכתב on top.
• The system has three finite tenses, an infinitive and an imperative. That is all of it; English's wrote / have written / had written / was writing all collapse into כתבתי.
• No auxiliaries. No do-support, no progressive "be", no perfect "have" — and no present-tense form of להיות to insert before a participle.
• The present marks gender and number but never person, so כותב covers I, you and he for a male subject — which is why the present, unlike the past and the future, has to be given its subject.
• Unpointed text hides both the vowels and the boundary between root and affix, which is why stripping a verb to its root is a skill you practise rather than a lookup you perform.