Until the advent of free verse in the 20th century, this poetry was always quantitative—that is the lines were composed in various patterns of long and short syllables.
However, in recent years it has been recognized that for the most part Persian metres developed independently from those in Arabic, and there has been a movement to analyze them on their own terms.
In this article, the following scansion symbols are used: From a metrical point of view, classical Persian poetry can be divided into three main types.
[8] Such a quintet was known as a khamsa or khamse (from the Arabic خَمْسة xamsa '(group of) five'), and the practice was later imitated by other poets such as Amir Khosrow of Delhi and Khwaju Kermani of Shiraz.
Thus for example the rhythm of Ferdowsi's epic poem the Shahnameh (u – – | u – – | u – – | u –) was thought to be a modification of the Arabic metre mutaqārib, which is similar (u – x | u – x | u – x | u –).
Elwell-Sutton argued against the idea that Persian metres are simply an adaptation of Arabic ones, and on the whole his view has been accepted by subsequent scholars.
"[16] After examining the metres of over 20,000 Persian poems, Elwell-Sutton realised that the vast majority of them (well over 99%) could be analysed in terms of just five repeating patterns.
In an experiment, L. P. Elwell-Sutton recorded two well-educated Persian speakers reading a number of poems and measured the length of each syllable in hundredths of a second.
Some names, such as that of Yazdgerd III, the final king of the Sassanid empire, are exceptions to this trend in that they are almost always pronounced with a short anaptyctic vowel, either the nīm-fathe ('half-"a"') (i.e. Yazd-a-gerd) or mostly in the modern Iranian pronunciation, the nīm-kasreh ('half-"e"') (i.e. Yazd-e-gerd).
[27] The unwritten ezāfe suffix (as in Tork-e Šīrāzī 'Shirazi Turk') may be pronounced either long or short (-e or -ē), as the metre requires, and the word o 'and' similarly may be either o or ō.
[30] When the sound ī (written ی) is followed directly by another vowel in Persian words, as in بیا biyā 'come', it is pronounced short, and similarly with the sounds ey and ow when they are followed by a vowel; for example mey-ē bāqī 'the remaining wine', bešnow az ney 'listen to the flute', bīni ān Tork 'do you see that Turk?
There is also a preference for feet of four syllables, rather than three or five; thus the kāmil metre (common in Arabic) with its repeated five-syllable foot of u u – u – does not easily fit into the Persian metrical system and is almost never found.
[11] When the rhythm u u – is replaced by – – in the first half of the line, there is usually a phrase boundary or potential pause after the second long syllable.
[50] The following line from a ruba'i is typical: The biceps ending – u u – is also sometimes found in Arabic poetry, in the basīṭ metre, for example in poems written by the poet Abu Nuwas, who was of half-Persian origin.
[54] A feature of classical Persian pronunciation which is no longer observed in Iran today is the distinction between long ō and ū, and between ē and ī; for example, šēr 'lion' vs. šīr 'milk'.
The transliteration is based on that approved by the United Nations in 2012, which represents the current pronunciation of educated speakers in Iran, except that to make scansion easier, the long vowels are marked (ā, ē, ī, ō, ū).
These are pronounced longer than the usual long syllables The first pattern, based on the foot u – –, is known by the Arabic name mutaqārib (motaqāreb).
Examples include Fakhruddin Gurgani's Vis o Ramin, and Nezami's Khusrow o Shirin, which begins as follows: The same metre 2.1.11, or hazaj, was used from early times in popular poetry, such as the do-baytī, in which the opening iamb (u –) can sometimes be replaced by – – or – u.
With this metre there is frequently an internal rhyme at the mid-point of the line, as in the poem above, or in the following by Khwaju Kermani: The statues of Buddha from northern Afghanistan were proverbial for their beauty.
[77][78] The most famous of these was the Masnavi-e Ma'navī, or the "Spiritual Masnavi", completed in 1273, of Mowlana Jalal al-Din Rumi (better known in Iran as Mowlavī) of about 25,000 couplets, which begins: The same 15-syllable ramal metre, 2.4.15, was used in several poems by Hafez, including the following on a mystical theme: The same metre, 2.4.15, is also used in the following qasida by the 11th-century poet Manuchehri in praise of a beautiful minstrel.
In the example below by the 11th-century poet Manuchehri, the two short syllables are kept constant and not replaced by a single long: The poem is stanzaic, consisting of 35 stanzas of 3 couplets each.
When the singer Giti sings this poem to a modern tune, the music is in triple time with the downbeat on the final syllable of the above feet.
One such masnavi is the 12th-century Nezami's Haft Paykar (Seven Portraits or Seven Beauties), which begins as is customary with an address to God: As with other metres ending with (u u –), this is easily changed to (– –), as in the above couplet.
[90] The introduction to the Golestān includes a short 12-couplet masnavi of philosophical reflection, which begins as follows: It includes the famous advice: In the same metre 4.5.11 is this well-known poem also from the introduction to the Golestān: An example of 4.7.14 is a qasida, by the 10th/11th-century poet Farrokhi of Sistan, which begins as follows:[95] In the second line, the perfect suffix -e is short in tanīde 'spun', but lengthened in bāftē 'woven'.
[97][98] A well-known example is the following, playing on the words sūfī "Sufi" and sāfī 'clear': The metre 4.7.7(2) consists of the first seven syllables of 4.7.15 repeated.
Farzaad proposed the following, using a 5-syllable foot:[103] A shorter example of a masnavi in 5.1.10, consisting of just three couplets, is found in Saadi's Golestân.
The twelfth verse goes as follows: The internal rhymes -dast/-qast, -yāb/-tāb confirm Farzaad's claim that a foot division should be made after the first two syllables.
There is a break in the middle of each hemistich, but an overlong syllable may overlap the break, as in the first hemistich below, from a ghazal of Saadi:[106] It has been suggested that this metre is derived from 3.1.08(2) by the reversal (syncopation or anaclasis) of syllables 4 and 5:[62] In Ancient Greek poetry the rhythm (u u – u – u – –) is known as an anacreontic, which is named after Anacreon, a poet of Asia Minor (6th–5th century BC).
The following poem, for example, by the 18th-century poet Hatef Esfahani, is written in the kāmil metre, rare in Persian but common in Arabic.
It begins as follows: It is traditionally sung to a melody (gūše) called Čahārbāq, named after the well-known avenue Chaharbagh in Isfahan.