בניין נפעל: Passive and Middle

You meet nif'al long before anyone tells you it exists. נכנס, נמצא, נשאר, נראה and נגמר turn up in the first weeks of any Hebrew course as ordinary vocabulary — go in, be located, stay, look, be over — and only later does someone point out that they are one system. This page is that moment.

The hard part is not the meaning. It is that a nif'al verb does not look like the same word across the tenses.

The נ that comes and goes

In the past and the present the נ sits at the front where you can see it. In the future and the infinitive it is gone — and something else has appeared: the first root letter carries a dagesh and the prefix carries a chirik.

Nothing was lost. The נ of the binyan has no vowel of its own, and a vowelless consonant standing immediately before another consonant assimilates into it: yin-kanés → yikanés, and the doubling of the כ is the swallowed נ. That is the same assimilation the pa'al verbs נפל and נסע undergo in their own future (see פ״נ verbs); nif'al just does it with a prefix instead of a root letter.

In unpointed writing a chirik standing before a doubled letter takes a yod, which is where the two-yod spellings come from: ייכנס, ייגמר, להיכנס. The rule itself is on double ו and double י, and its consequences for this binyan on the nif'al future.

When the first root letter refuses a dagesh

Five letters — א, ה, ח, ע and ר — never take a dagesh. When one of them is the first root letter there is nowhere for the נ to go, so the prefix vowel lengthens instead: the chirik becomes a tsere. Grammarians call this compensation תשלום דגש, and the Academy of the Hebrew Language states it as a rule of the binyan: with a guttural first radical there is no dagesh, and a tsere stands where the chirik would have been.

Note the practical payoff in the future column: the unpointed spelling is the same two yods either way. Whether the vowel is a chirik (ייכנס) or a tsere standing in for one (ייראה), you write יי. So the two-yod opening is a reliable flag for a nif'al future even before you know which sub-class the verb belongs to.

What nif'al verbs actually do

Four jobs, and a given verb does one of them. You cannot read off which from the shape.

1. The passive of a pa'al verb. כָּתַב → נִכְתַּב, מָכַר → נִמְכַּר, בָּחַר → נִבְחַר, גָּנַב → נִגְנַב. This is the textbook description, and for these verbs it is exactly right: the object of the active sentence has become the subject.

2. Middle, or anticausative — something happens and nobody did it. נִשְׁבַּר, נִפְתַּח, נִסְגַּר, נִגְמַר. English uses the same verb for both sides (I broke the glass / the glass broke); Hebrew changes binyan, and the nif'al is the side with no agent anywhere in the sentence.

3. Stative and lexicalised — a verb that describes a state, or simply a word in its own right. נִמְצָא, נִשְׁאַר, נֶחְשַׁב, נִדְמֶה. Here the passive label has stopped meaning anything: the bank is located opposite the station is not something anyone did to the bank.

4. Reciprocal — two parties, one event. נִפְגַּשׁ, נִלְחַם, נִדְבַּר. The subject is plural (or joined by עם / את "with"), and each party is doing the thing to the other.

Why "the passive of pa'al" misparses the binyan

Take the five nif'al verbs a beginner meets first — נכנס, נשאר, נהנה, נזהר, נרדם — and look for the active verbs they are supposedly the passives of. There are none. No Israeli says כָּנַס for "he put in"; the causative is the hif'il *הכניס. Beside נשאר stands השאיר "leave behind", beside נרדם stands הרדים "put to sleep" — hif'il again, not pa'al.

So the pairing is not a grid you can fill in mechanically. It is a set of local relationships between individual verbs: some pa'al verbs happen to have a nif'al passive, and a large part of nif'al is not a passive of anything. The full picture of which binyanim pair with which is on binyan pairs.

Where the rest of it lives

This page is the orientation. The paradigms are on four pages of their own — present, past, future and infinitive and imperative — and the four jobs above each have a page: passive use, middle voice, statives and reciprocals.

Common Mistakes

1. Using a nif'al verb where Hebrew needs the hif'il causative. נשאר means stayed, not left behind — it has no room for an object.

2. Giving a reciprocal verb a direct object. English meet someone is transitive; Hebrew's נפגש already contains the "each other", so the other party arrives with עם or את "with".

3. Making the middle verb transitive. נגמר is what a film does, not what you do to it.

Key Takeaways

• nif'al shows its נ in the past and the present (נכנס, נשאר) and hides it in the future and the infinitive, where it has assimilated into the first root letter — leaving a dagesh there and a chirik in the prefix: יִכָּנֵס, לְהִכָּנֵס.
• Unpointed, that chirik takes a yod: ייכנס, להיכנס. Two yods at the front of a future verb is the binyan's best recognition cue.
• If the first root letter is א, ה, ח, ע or ר no dagesh is possible, and the prefix vowel lengthens to a tsere by תשלום דגש: יֵרָאֶה, יֵעָלֵם, יֵאָמֵר. The unpointed spelling still has two yods — ייראה, ייעלם.
• The binyan does four different jobs: passive (נכתב, נמכר), middle (נשבר, נגמר), stative (נמצא, נחשב) and reciprocal (נפגש, נלחם).
• "The passive of pa'al" is only part of it. נכנס, נשאר, נהנה, נזהר and נרדם have no active partner in use at all; where a causative is needed, Hebrew supplies a hif'il — הכניס, השאיר, הרדים.
• A nif'al verb takes no direct object, so the object-marker את never follows one. What it relates to arrives with a preposition: נכנסתי לחדר, נפגשתי עם יעל, נזהרתי מהכלב.