What Each Binyan Means — and When It Means Nothing

Every Hebrew course teaches the same table: pa'al is basic, pi'el is intensive, hif'il is causative, nif'al is passive, hitpa'el is reflexive. The table is not wrong. It is just far less predictive than it looks, and knowing exactly how much weight it will bear is worth more than the table itself.

Where the system works

Start with the cases that make the semantics look like a rule, because they are real and they are the reason the framework exists.

A dash means the root has no everyday verb in that binyan — the ordinary state of affairs, not a hole.

ל־ב־שׁ is the showcase root: three binyanim, three argument structures, and the semantics fall out exactly as advertised — you wear something, you dress someone else, you dress yourself. ל־מ־ד fills the causative slot with pi'el instead of hif'il and behaves just as neatly: לָמַד "learned", לִימֵּד "taught". (Beware of filling ל־ב־שׁ's empty pi'el cell: לְיַבֵּשׁ "to dry" is not a pi'el of לבש at all but the unrelated root י־ב־שׁ, of יָבֵשׁ "dry".)

Where it stops working

Now the verbs you meet every day. Each of these sits in a binyan whose textbook meaning it flatly contradicts.

These are not obscure verbs dredged up to embarrass the rule. הגיע, החליט, הרגיש, דיבר, חיפש and נכנס are among the first fifty verbs any learner needs, and none of them can be derived from its binyan's meaning by any procedure.

So what is a binyan actually telling you?

State the working principle plainly, because half-believing the semantic rule costs more than not knowing it:

Which binyan a given root sits in is a lexical fact, learned with the verb. The semantics are a decoding aid — they help you make sense of a verb you have already half-heard — and they are not a generating rule. They will never let you produce a verb you have not met and be confident it exists.

What the binyan does tell you reliably is grammatical rather than semantic, and it is worth more:

• How many arguments the verb takes. נשבר cannot take a direct object; שבר must have one. That much is dependable.
• Which paradigm to conjugate it in. This is the binyan's actual job.
• What the related forms will look like — the verbal noun (דיבור from pi'el, הסבר and הַסְבָּרָה from hif'il, הִתְלַבְּשׁוּת from hitpa'el), the passive participle, the passive counterpart.

Which binyanim still take new verbs

This is the part of the picture that most references skip, and it is the part that tells you where the language is actually going. The findings below are Bat-El's (2013, Encyclopedia of Hebrew Language and Linguistics, Brill, 704–707) and Bolozky's (1978, Afroasiatic Linguistics 5; 2007).

Why pi'el wins. Bat-El's account is structural, and it explains the pattern rather than just recording it. Of the binyanim you can actually choose — the active ones — pi'el is the only one that adds no derivational consonant to the root: no נ, no ה, no הת. (pu'al adds none either, but pu'al is never a choice; it is pi'el's automatic passive.) So a four-consonant root fits inside pi'el without anything being displaced. Its consonant frame also stays put across the paradigm: past סִימֵּס, future יְסַמֵּס, infinitive לְסַמֵּס — s-m-s intact and in order in all three, with only the vowels moving. And it is neutral as to transitivity, so it can host a new verb whatever its argument structure. Between them, those three properties make pi'el the obvious mould for anything new — especially the loans and acronyms that arrive with four consonants and would not fit anywhere else.

Why hif'il is not dead. It is frequently described as closed. It is not: it still takes inchoatives — הִשְׁחִיר "turned black", הִבְרִיא "got better", הִרְזָה "lost weight" — and loans whose consonant cluster survives better in the hif'il frame, הִשְׁפְּרִיץ and הִסְנִיף. What has happened is narrower and more interesting: Bolozky (2007) finds hif'il's retreat to be register-conditioned, still preferred in literary Hebrew where colloquial Hebrew has switched to pi'el.

Why pa'al is not quite dead either. Say essentially closed rather than dead, because that is what the source says: Bolozky (1978) reports knowing of only two recent denominative pa'al verbs — gamáz, and kharáp "to nap", from Yiddish khropn "snore" — and concludes that pa'al's productivity is very restricted rather than extinct. Bat-El independently cites the same kharáp as one of the very few denominal pa'al verbs. Two verbs is not zero; it is also not a pattern you can join.

Bolozky adds a knock-on finding that explains a gap learners notice: nif'al is closed too, and because nif'al is the binyan pa'al pairs with, no recent inchoatives are realised in it. New inchoatives and ingressives go to hitpa'el instead — which is exactly why the modern language keeps producing התייאש, התאהב, התמסטל and not their nif'al equivalents.

pu'al and huf'al are not choices

The two remaining binyanim are not options you pick between. pu'al is the passive of pi'el and huf'al the passive of hif'il, mechanically: סִידֵּר → סוּדַּר, הִזְמִין → הוּזְמַן. And each has a structural gap that follows from that role — no infinitive and no imperative, because you cannot order someone to undergo something, and there is no independent verb to point an infinitive at. In practice you meet both far more often as participial adjectives (מְסוּדָּר "tidy", מוּכָן "ready") than as finite verbs. The pairings are worked out on binyan pairs.

Common Mistakes

1. Giving a hitpa'el a direct object. The reflexive already contains its object.

2. Giving a nif'al a direct object. נשבר describes what happened to the subject; it has no room for anything else.

3. Assuming a hif'il must be transitive. Several of the commonest hif'il verbs take no object at all.

4. Adding a reflexive pronoun to a hitpa'el. The binyan already carries the reflexive; saying it twice is not emphasis, it is ungrammatical.

Key Takeaways

• The textbook semantics are real but partial: ל־ב־שׁ, שׁ־ב־ר and ל־מ־ד behave exactly as advertised, and הגיע, החליט, הרגיש, דיבר, חיפש, נכנס and נשאר do not.
• The binyan of a root is a lexical fact, learned with the verb. The semantics decode; they do not generate.
• What a binyan does predict dependably is argument structure, the paradigm, and the shape of the related forms.
• pi'el and hitpa'el take most new verbs — pi'el when agentive, hitpa'el when not. Pi'el wins because it is the one active binyan that adds no consonant to the root, keeps the consonant frame intact across the whole paradigm, and is neutral as to transitivity, so quadriliterals and loans fit straight into it.
• hif'il is not closed. It still takes inchoatives (השחיר, הבריא, הרזה) and cluster-preserving loans (השפריץ, הסניף); its retreat is register-conditioned, holding on in literary Hebrew where speech has moved to pi'el (Bolozky 2007).
• pa'al is essentially closed — Bolozky (1978) knows of two recent coinages — and nif'al with it, which is why new inchoatives surface in hitpa'el instead.
• pu'al and huf'al are not choices: they are the automatic passives of pi'el and hif'il, and they have no infinitive and no imperative.