The Hebrew of the late centuries BCE and early centuries of the Common Era had a system with five phonemiclong vowels and five short vowels. In the later dialects of the 1st millennium CE, phonemic vowel length disappeared, and instead was automatically determined by the context, with vowels pronounced long in open syllables and short in closed ones. However, the previous vowel phonemes merged in various ways that differed from dialect to dialect:
In Tiberian Hebrew, which underlies the written system of vowels, short became ; long became ; while and both merged into an in-between sound , which was indicated by qamatz.
In the Babylonian vocalization, however, short and long variants simply merged, with and becoming , while and became ; and this system underlies the pronunciation of Modern Hebrew.
The result is that in Modern Hebrew, the vowel written with qamatz might be pronounced as either or , depending on historical origin. It is often said that the two sounds can be distinguished by context:
The qamatz sound of, known as Qamatz Qaṭan occurs in a "closed syllable", i.e. one which ends in a consonant marked with a shwa nakh or with a dagesh ḥazaq ;
The qamatz sound of, known as Qamatz Gadol occurs in an "open syllable", i.e. any other circumstance: one which ends in a consonant followed by a normal vowel, a consonant at the end of a word and with no vowel marking, or a consonant marked with a shwa na.
Unfortunately, the two varieties of shwa are written identically, and pronounced identically in Modern Hebrew; as a result, there is no reliable way to distinguish the two varieties of qamatz when followed by a vowel marked with a shwa. An example of the qamatz qatan is the Modern Hebrew word . According to the standard Hebrew spelling rules as published by the Academy of the Hebrew Language, words which have a qamatz qatan in their base form must be written without a vav, hence the standard vowel-less spelling of is. In practice, however, Modern Hebrew words containing a qamatz qatan do add a vav to indicate the pronunciation; hence the "nonstandard" spelling is common in newspapers and is even used in several dictionaries, for example Rav Milim. Words, which in their base form have a ḥolam that changes to qamatz qaṭan in declension, retain the vav in vowel-less spelling: the noun is spelled in vowel-less texts; the adjective is spelled in vowel-less text, despite the use of qamatz qatan, both according to the standard spelling and in common practice. Some books print the qamatz qaṭan differently, although it is not consistent. For example, in siddur Rinat Yisrael the vertical line of qamatz qatan is longer. In a book of Psalms used by some Breslov hassidim the qamatz qatan is bolder. In the popular niqqud textbook Niqqud halakha le-maase by Nisan Netser, the qamatz qatan is printed as an encircled qamatz for didactic purposes. Unicode defines the code point, although its usage is not required.
Ḥaṭaf Qamatz
Ḥaṭaf Qamatz is a "reduced qamatz". Like qamatz qatan, it is pronounced, but the rationale for its usage is different: it replaces the shva on letters which require a shva according to the grammar, but where the traditional pronunciation is. This mostly happens with gutturals, for example in , but occasionally also on other letters, for example and .
The following table contains the pronunciation and transliteration of the different qamatzes in reconstructed historical forms and dialects using the. The transcription in IPA is above and the transliteration is below. The letters bet and heth used in this table are only for demonstration. Any letter can be used.
These vowel lengths are not manifested in Modern Hebrew. The short o and long a have the same niqqud. Because of this, the short o is usually promoted to a long o in Israeli writing, written as a vav, for the sake of disambiguation. By adding two vertical dots the vowel is made very short.
Unicode encoding
Note: the glyph for QAMATS QATAN may appear empty or incorrect if one applies a font that cannot handle the glyph necessary to represent Unicode character U+05C7. Usually this Unicode character isn't used and is substituted with the similar looking QAMATS.
