Northwest Semitic
Concentrated in the Middle East
Linguistic classificationAfro-Asiatic

Northwest Semitic is a division of the Semitic languages comprising the indigenous languages of the Levant. It emerged from Proto-Semitic in the Early Bronze Age. It is first attested in proper names identified as Amorite in the Middle Bronze Age. The oldest coherent texts are in Ugaritic, dating to the Late Bronze Age, which by the time of the Bronze Age collapse are joined by Old Aramaic, and by the Iron Age by Sutean and the Canaanite languages (Phoenician/Punic, Edomite, Moabite and Hebrew).[1]

The term was coined by Carl Brockelmann in 1908,[2] who separated Fritz Hommel's 1883 classification of Semitic languages[2] into Northwest (Canaanite and Aramaic), East Semitic (Akkadian, its Assyrian and Babylonian dialects, Eblaite) and Southwest (Arabic, Old South Arabian languages and Abyssinian).[3]

Brockelmann's Canaanite sub-group includes Ugaritic, Phoenician and Hebrew. Some scholars now regard Ugaritic either as belonging to a separate branch of Northwest Semitic (alongside Canaanite) or a dialect of Amorite.

Central Semitic is a proposed intermediate group comprising Northwest Semitic and Arabic. Central Semitic is either a subgroup of West Semitic or a top-level division of Semitic alongside East Semitic and South Semitic.[4] SIL Ethnologue in its system of classification (of living languages only) eliminates Northwest Semitic entirely by joining Canaanite and Arabic in a "South-Central" group which together with Aramaic forms Central Semitic.[5] The Deir Alla Inscription and Samalian have been identified as language varieties falling outside Aramaic proper but with some similarities to it, possibly in an "Aramoid" or "Syrian" subgroup.[6][7]

It is clear that the Taymanitic script expressed a distinct linguistic variety that is not Arabic and not closely related to Hismaic or Safaitic, while it can tentatively be suggested that it was more closely related to Northwest Semitic.[8]

Historical development

Aramaic alphabets
Phoenician alphabets
Comparison of Northwest Semitic scripts, by Mark Lidzbarski in 1898
Charles Morton's 1759 updated version of Edward Bernard's "Orbis eruditi",[9] comparing all known alphabets as of 1689, including Northwest Semitic which is described as "Adami, Noachi, Nini, Abrahami, Phoenicum et Samaritarum ante Christe (5509) a nummis Iudaicis Africanisque Pentateucho Mosis"

The time period for the split of Northwest Semitic from Proto-Semitic or from other Semitic groups is uncertain. It has been recently suggested by Richard C. Steiner that the earliest attestation of Northwest Semitic is to be found in snake spells from the Egyptian Pyramid Texts, dating to the mid-third millennium BCE.[10] Amorite personal names and words in Akkadian and Egyptian texts from the late third millennium to the mid-second millennium BCE and the language of the Proto-Sinaitic inscriptions dated to the first half of the second millennium otherwise constitute the earliest traces of Northwest Semitic, the first Northwest Semitic language attested in full being Ugaritic in the 14th century BCE.[11]

During the early 1st millennium, the Phoenician language was spread throughout the Mediterranean by Phoenician colonists, most notably to Carthage in today's Tunisia. The Phoenician alphabet is of fundamental importance in human history as the source and ancestor of the Greek alphabet, the later Latin alphabet, the Aramaic (Square Hebrew), Syriac, and Arabic writing systems, Germanic runes, and ultimately Cyrillic.

From the 8th century BCE, the use of Imperial Aramaic by the Neo-Assyrian Empire (935–608 BC) and the succeeding Neo-Babylonian Empire (612–539 BC) and Achaemenid Empire (539–332 BC), a form of the Aramaic language, spread throughout the Northwest Semitic region of the Levant, northern regions of the Arabian peninsula and southern regions of Anatolia, and gradually drove most of the other Northwest Semitic languages to extinction. The ancient Judaeans adopted Aramaic for daily use, and parts of the Tanakh are written in it. Hebrew was preserved, however, as a Jewish liturgical language and language of scholarship, and resurrected in the 19th century, with modern adaptations, to become the Modern Hebrew language of the State of Israel.

After the Muslim conquests of the 7th century, Arabic began to gradually replace Aramaic throughout the region. Classical Syriac-Aramaic survives today as the liturgical language of the Assyrian Church of the East, Syriac Orthodox Church, Chaldean Catholic Church, and other churches of Syriac Christians. It is spoken in modern dialects with an estimated one million fluent speakers by endangered indigenous populations scattered throughout the Middle East, most commonly by the Assyrians, Gnostic Mandeans, the Arameans (Syriacs) of Maaloula and Jubb'adin, and Mizrahi Jews. There is also an Aramaic substratum in Levantine and Mesopotamian Arabic.


Sound changes

Phonologically, Ugaritic lost the sound *ṣ́, replacing it with /sˁ/ () (the same shift occurred in Canaanite and Akkadian). That this same sound became /ʕ/ in Aramaic (although in Ancient Aramaic, it was written with qoph), suggests that Ugaritic is not the parent language of the group. An example of this sound shift can be seen in the word for earth: Ugaritic /ʔarsˁ/ (’arṣ), Punic /ʔarsˁ/ (’arṣ), Tiberian Hebrew /ʔɛrɛsˁ/ (’ereṣ), Biblical Hebrew /ʔarsˁ/ (’arṣ) and Aramaic /ʔarʕaː/ (’ar‘ā’).

The vowel shift from *aː to /oː/ distinguishes Canaanite from Ugaritic. Also, in the Canaanite group, the series of Semitic interdental fricatives become sibilants: (), () and *θ̣ () became /z/, /ʃ/ (š) and /sˤ/ () respectively. The effect of this sound shift can be seen by comparing the following words:

shift Ugaritic Aramaic Biblical Hebrew translation
()→/z/ 𐎏𐎐𐎁
()→/ʃ/ (š) 𐎘𐎍𐎘
*θ̣ ()→/sˤ/ () 𐎑𐎍𐎍


Proto-Northwest Semitic had three contrastive vowel qualities and a length distinction, resulting in six vocalic phonemes: *a, *ā, *i, *ī, *u, and *ū. While *aw, *ay, *iw, *iy, *uw, and *uy are often referred to as diphthongs, they do not seem to have had a different status as such, rather being a normal sequence of a short vowel and a glide.


Proto-Northwest Semitic consonant phonemes
Type Manner Voicing Labial Interdental Alveolar Lateral Postalveolar Palatal Velar/Uvular Pharyngeal Glottal
Obstruent Stop voiceless *p [p] *t [t] *k [k]
emphatic *ṭ [] *q/ [] , ˀ [ʔ]
voiced *b [b] *d [d] *g [ɡ]
Fricative voiceless *ṯ [θ] *s [t͡s] [ɬ] [s~ʃ] *ḫ [x~χ] *ḥ [ħ] *h [h]
emphatic *ṯ̣/θ̣/ [t͡θˤ] *ṣ [t͡sˤ] *ṣ́/ḏ̣ [t͡ɬˤ]
voiced *ḏ [ð] *z [d͡z] /ǵ [ɣ~ʁ] , ˤ [ʕ]
Resonant Trill *r [r]
Approximant *w [w] *l [l] *y [j]
Nasal *m [m] *n [n]

Suchard proposes that: "*s, both from original *s and original *ṯ, then shifted further back to a postalveolar *š, while deaffrication of *ts and *dz to *s and *z gave these phonemes their Hebrew values, as well as merging original *dz with original *ḏ. In fact, original *s may have been realized as anything between [s] and [ʃ]; both values are attested in foreign transcriptions of early Northwest Semitic languages".[citation needed]


In Proto-Northwest Semitic the emphatics were articulated with pharyngealization. Its shift to backing (as opposed to Proto-Semitic glottalization of emphatics) has been considered a Central Semitic innovation.[12]

According to Faber, the assimilation *-ṣt->-ṣṭ- in the Dt stem in Hebrew (hiṣṭaddēḳ ‘he declared himself righteous’) suggests backing rather than glottalization. The same assimilation is attested in Aramaic (yiṣṭabba ‘he will be moistened’).



Three cases can be reconstructed for Proto-Northwest Semitic nouns (nominative, accusative, genitive), two genders (masculine, feminine) and three numbers (single, dual, plural).[13]

Number/case ‘dog(s)’ (m.) ‘bitch(es)’ (f.)
sg.nominative *kalbu(m) *kalbatu(m)
sg.genitive *kalbi(m) *kalbati(m)
sg.accusative *kalba(m) *kalbata(m)
du.nominative *kalbā(na) *kalbatā(na)
du.genitive/accusative *kalbay(na) *kalbatay(na)
pl.nominative *kalabū(na) *kalabātu(m)
pl.genitive/accusative *kalabī(na) *kalabāti(m)


Proto-Northwest Semitic pronouns had 2 genders and 3 grammatical cases.



nominative genitive accusative *ʔanāku/ *ʔana *-tu *-ī, *-ya -nī *ʔanta *-ka *-ta *ʔanti *-ki *-ti *hūʔa *-hu *-a *hīʔa *-hā *-at *naḥnu/ *naḥnā *-nā *ʔantum *-kum *-tum *ʔantin *-kin *-tin *hum(ū) *-hum *-ū *hin(na) *-hin *-ā


Reconstruction of Proto-Northwest Semitic numbers.

Number Masculine Feminine
One *ʔaḥadum *ʔaḥattum
Two *ṯnāna *ṯintāna
Three *ṯalāṯatum *ṯalāṯum
Four *ʔarbaʕatum *ʔarbaʕum
Five *ḫamisatum *ḫamisum
Six *siṯṯatum *siṯṯum
Seven *sabʕatum *sabʕum
Eight *ṯamāniyatum *ṯamāniyum
Nine *tisʕatum *tisʕum
Ten *ʕaśaratum *ʕaśrum


Paradigm of the strong verb (G-stem)
Person/gender/number Suffix conjugation (Perfect) Meaning Prefix conjugations (Imperfect) Meaning *qaṭal-a 'he has killed' *ya-qṭul(-u/-a) 'he will kill' *qaṭal-at 'she has killed' *ta-qṭul(-u/-a) 'she will kill' *qaṭal-ta 'you have killed' *ta-qṭul(-u/-a) 'you will kill' *qaṭal-ti 'you have killed' *ta-qṭul-ī(-na) 'you will kill'
1sg. *qaṭal-tu 'I have killed' *ʔa-qṭul(-u/-a) 'I will kill' *qaṭal-ū 'they have killed' *ya-qṭul-ū(-na) 'they will kill' *qaṭal-ā 'they have killed' *ta-qṭul-na 'they will kill' *qaṭal-tum 'you have killed' *ta-qṭul-ū(-na) 'you will kill' *qaṭal-tin 'you have killed' *ta-qṭul-na 'you will kill'
1pl. *qaṭal-nā 'we have killed' *na-qṭul(-u/-a) 'we will kill'

The G fientive or G-stem (Hebrew qal) is the basic, most common, unmarked stem. The G-stem expresses events. The vowel of the prefix of the prefix conjugations in Proto-Northwest Semitic was *-a- and the stem was *-qṭul- or *-qṭil-, as in *ya-qṭul-u 'he will kill', while the stem of the suffix conjugation had two *a vowels, as in *qaṭal-a 'he has killed'.

The G stative is like the fientive but expressing states instead of events. The prefix conjugation of stative roots, the vowel of the prefixes was *-i- and it contained and *a vowel, e.g. *yi-kbad-u 'he will become heavy', while the second vowel of the suffix conjugation was either *-i-, as in *kabid-a 'he is/was/will be heavy', or *-u-, as in *ʕamuq-a 'it is/was/will be deep'. Whether the G-stem stative suffix conjugation has *i or *u in the stem is lexically determined.

The N-stem (Hebrew nip̄ʕal) is marked by a prefixed *n(a)-. It is mediopassive which is a grammatical voice that subsumes the meanings of both the middle voice and the passive voice. In other words, it expresses a range of meanings where the subject is the patient of the verb, e.g. passive, medial, and reciprocal. The stem of the suffix conjugation is *naqṭaland the stem of the prefix conjugations is *-nqaṭil-; as is the case with stative Gstem verbs, the prefix vowel is *-i-, resulting in forms like *yi-nqaṭil-u 'he will be killed'.

The D-stem (Hebrew piʕel) is marked by gemination of the second radical in all forms. It has a range of different meanings, mostly transitive. The stem of the suffix conjugation is *qaṭṭil-, and the same stem is used for the prefix conjugations. It is not clear whether the Proto-Northwest-Semitic prefix vowel should be reconstructed as *-u-, the form inherited from Proto-Semitic (i.e. *yuqaṭṭil-u), or as *-a-, which is somewhat supported by evidence from Ugaritic and Hebrew (*yaqaṭṭil-u).

The C-stem (Hebrew hip̄ʕil) more often than not expresses a causative meaning. The most likely reconstructions are *haqṭil- (from older *saqṭil-) for the stem of the suffix conjugation and *-saqṭil- for the stem of the prefix conjugations. The reconstructed prefix vowel is the same as that of the D-stem, and similarly, the participle is to be reconstructed as *musaqṭilum.

All of the stems listed here, except the N-stem, could bring forth further derivation. The "internal passive stems" (Gp, Dp, and Cp; Hebrew passive qal, puʕal, and hɔp̄ʕal) aren't marked by affixes, but express their passivity through a different vowel pattern. The Gp prefix conjugation can be reconstructed as *yu-qṭal-u 'he will be killed'. Reflexive or reciprocal meanings can be expressed by the t-stems, formed with a *t which was either infixed after the first radical (Gt, Ct) or prefixed before it (tD).

The precise reconstruction are uncertain.

Proto-Northwest-Semitic verbal stems
G fientive G stative D C
perfect *qaṭal-a *kabid-a *qaṭṭil-a *ha-qṭil-a
imperfect *ya-qṭul-u *yi-kbad-u *yV-qaṭṭil-u *yVsa-qṭil-u
participle *qāṭil-um *kabid-um *mu-qaṭṭil-um *musa-qṭil-um
Gp N Dp Cp
perfect *quṭVl-a *na-qṭal-a *quṭṭVl-a *hu-qṭVl-a
imperfect *yu-qṭal-u *yin-qaṭil-u *yu-qVṭṭal-u *yusV-qṭal-u
participle *qaṭīl-um, *qaṭūl-um *na-qṭal-um or *mun-qaṭil-um? *mu-qVṭṭal-um *musV-qṭal-um
Gt tD Ct
perfect *qtaṭVl-a? *ta-qaṭṭVl-a *sta-qṭVl-a?
imperfect *yi-qtaṭVl-u *yVt-qaṭṭVl-u *yVsta-qṭVl-u
participle *mu-qtaṭVl-um *mut-qaṭṭVl-um *musta-qṭVl-um



  1. ^ Aaron D. Rubin (2008). "The subgrouping of the Semitic languages". Language and Linguistics Compass. 2 (1): 61–84. doi:10.1111/j.1749-818x.2007.00044.x.
  2. ^ a b The Semitic Languages: An International Handbook, Chapter V, page 425
  3. ^ Kurzgefasste vergleichende Grammatik der semitischen Sprachen, Elemente der Laut- und Formenlehre (1908), quote "Das Westsemitische gliedert sich in zwei Hauptgruppen, das Nord- und das Südwestsemitische... Das Nordwestsemitische umfaßt das Kanaanäische und das Aramäische...Das Südwest semitische umfaßt das Arabische und Abessinische."
  4. ^ Linguist List Central Semitic composite tree (with Aramaic and Canaanite grouped together in Northwest Semitic, and Arabic and Old South Arabian as sisters) Archived 2009-10-14 at the Wayback Machine
    Linguist List bibliography of sources for composite tree Archived 2011-07-23 at the Wayback Machine
    Rubin, Aaron D. 2007. The Subgrouping of the Semitic Languages, Language and Linguistics Compass, vol. 1.
    Huehnergard, John. 2004. "Afro-Asiatic," The Cambridge Encyclopedia of the World's Ancient Languages (Cambridge, pp. 138–159).
    Faber, Alice. 1997. "Genetic Subgrouping of the Semitic Languages," The Semitic Languages (Routledge, pp. 3–15)
    Huehnergard, John. 1991. "Remarks on the Classification of the Northwest Semitic Languages," The Balaam Text from Deir 'Alla Re-evaluated (Brill, pp. 282–293).
    Huehnergard, John. 1992. "Languages of the Ancient Near East," The Anchor Bible Dictionary, Volume 4, pp. 155–170.
    Voigt, Rainer M. 1987. "The Classification of Central Semitic," Journal of Semitic Studies 32:1–19.
    Goldenberg, Gideon. 1977. "The Semitic Languages of Ethiopia and Their Classification," Bulletin of the School of Oriental and African Studies 40:461–507.
    Ethnologue Central Semitic entry (with Arabic and Canaanite grouped together against Aramaic)
    The Ethnologue classification is based on Hetzron, Robert. 1987. "Semitic Languages," The World's Major Languages (Oxford, pp. 654–663).
    The older grouping of Arabic with South Semitic was "based on cultural and geographical principles", not on principles of empirical historical linguistics (Faber, 1997, pg. 5). "However, more recently, [Arabic] has been grouped instead with Canaanite and Aramaic, under the rubric Central Semitic..., and this classification is certainly more appropriate for Ancient North Arabian" (Macdonald, M.C.A. 2004. "Ancient North Arabian," The Cambridge Encyclopedia of the World's Ancient Languages Cambridge, pp. 488–533. Quote on pg. 489).
  5. ^ "Semitic". Ethnologue. SIL International. Retrieved 2014-06-02.
  6. ^ Huehnergard, John (1995). "What is Aramaic?". Aram (7): 282.
  7. ^ Kogan, Leonid (2015). Genealogical Classification of Semitic. de Gruyter. p. 601. doi:10.1515/9781614515494. ISBN 9781614517269.
  8. ^ Kootstra, Fokelien. "The Language of the Taymanitic Inscriptions and its Classification". ((cite journal)): Cite journal requires |journal= (help)
  9. ^ Edwin JEANS (1860). A Catalogue of Books, in all Branches of Literature, both Ancient & Modern ... on sale at E. Jeans's, bookseller ... Norwich. J. Fletcher. pp. 33–.
  10. ^ Steiner, Richard C. (2011). Early Northwest Semitic Serpent Spells in the Pyramid Texts. Eisenbrauns. ISBN 9789004369214.
  11. ^ Huehnergard, John & Pat-El, Na'ama (2019). The Semitic Languages. Routledge. pp. 11–12. ISBN 9780429655388.((cite book)): CS1 maint: multiple names: authors list (link)
  12. ^ Hetzron, Robert (2011). The Semitic Languages An International Handbook. De Gruyter Mouton. pp. ISBN 9783110186130.
  13. ^ Suchard, Benjamin (2019). The Development of the Biblical Hebrew Vowels: Including a Concise Historical Morphology. Brill. pp. 37–50. ISBN 978-90-04-39025-6.