Stress in Pashto: free stress, aspect and plurals

At a glance

• Pashto stress is free. In a word of more than one syllable one syllable carries the heavy stress, and which one is a property of the word, not of a fixed rule. [1, 2]

• Some words differ only in stress: ṭánga, "pear tree", and ṭāngá, "cart"; áspa, "mare", and aspá, "spotted fever". [1, 4]

• The perfective is marked by stress. The particle wə́ takes the heavy stress, and in many verbs stress moving to the front of the verb is the only sign of the perfective. [1, 2]

• Many endings carry the stress themselves: the plural -úna is always stressed, so surúd, "song", becomes surudúna. [1, 3]

• The Pashto script does not mark stress, so a reader who knows the word supplies it. [1, 2]

Stress sits where the word puts it

Habibullah Tegey and Barbara Robson, whose 1996 reference grammar of Pashto sets it out plainly: the place of the heavy stress, "whether on the last syllable, the next-to-last, or one of the earlier syllables", cannot be predicted, and the spelling does not mark it. A learner has to ask a speaker how a word is said and listen for the stressed syllable. [1] The stressed syllable is said with more force and, they think, a slightly higher pitch. They add that stress and pitch in Pashto have been little studied, since such work needs long fieldwork. [1]

There is a loose rule. In native words the stress falls on the last syllable when it ends in a consonant, and on the next-to-last when the last syllable ends in a vowel: randzúr, "sick" (masculine), and randzúra, "sick" (feminine). Tegey and Robson add that the exceptions are about as many as the words that obey it, and their first exception is the name of the language, paxtó (پښتو), stressed on the last syllable although it ends in a vowel. [1] They also write that the stress in Pashtun is on the last syllable. [1]

Loanwords bring their own habits. Many Persian words ending in the long vowel ā keep the Persian final stress: abā, "daddy", atyā, "eighty", ashnā, "friend". Arabic words with the same ending are not stressed there: ámā, "however", búbalā, "monster". [1] Anne Boyle David's descriptive grammar of 2014 comes to the same view from the other side and says that word stress in Pashto is "assigned lexically", that is, listed word by word. She marks it with an acute accent in paradigms and where it helps the meaning, as in saṛáy, "man", and žə́ba (ژبه), "tongue". [2]

Georg Morgenstierne, whose survey of Pashto is printed in the Encyclopaedia Iranica, named the same thing in a sentence, calling "the role played by mobile stress" characteristic of the language. [3] Stress also shapes the sounds of a word. In his account an old stressed a normally became o, as in mor, "mother", and before a nasal u, as in nūm, "name". [3]

Pairs that differ only in stress

Tegey and Robson print six pairs "exactly alike except for stress". They say the number of borrowings and other factors produced them. [1] The transcriptions below are theirs.

• ṭánga, "pear tree", and ṭāngá, "cart"

• áspa, "mare" (اسپه), and aspá, "spotted fever" (اسپه)

• góra, "look!", and gorá, "fair-skinned"

• ghuṭá, "diver", and ghúṭa, "knot" (غوټه)

• joṛá, "pair", and jóṛa, "well" (feminine)

• kenastá, "he is sitting", and kénastə, "he sat"

The last pair is a verb, and it shows that stress can do the work of a tense ending. In each pair the Pashto script is the same for both words, which is why the pairs are told apart only by hearing them. [1] A crowd-edited dictionary, Wiktionary, lists áspa and aspá as two separate entries with the same spelling, and gives the same two meanings. [4]

David's grammar gives two further kinds of pair. The pronoun haɣə́ (هغه) means "he, she, it" and háɣə (هغه) means "that", and a few verbs differ between continuous and aorist forms only by stress. [2] The next sections take these in turn.

Stress marks the verb's aspect

Pashto verbs have two aspects. Tegey and Robson call them imperfective and perfective. David calls them continuous and aorist, and explains that she avoids "perfective" because it suggests completion and can be confused with the perfect. [1, 2] Whatever the name, stress is one of the ways Pashto marks the difference, and here the two grammars agree closely.

The particle wə́

The perfective particle wə́ (و) carries heavy stress. The negative particle nə́ carries it too, and when both occur, both seem to be stressed. Tegey and Robson give four forms of one verb: [1]

• daredə́m, "I was standing"

• wə́daredəm, "I stood"

• nə́daredəm, "I was not standing"

• wə́nə́daredəm, "I did not stand"

They give the same effect on a present verb: khwrí, "he is eating", against nə́ khwrí, "he is not eating". [1] Stress in a sentence can therefore shift the heavy stress off the word that would normally take it. [1]

David's first conjugation works the same way. These verbs make the aorist by adding the prefix wə́-, which carries its own stress, and the syllables after it are unstressed. [2] In Dzadrani, which David counts among the Middle dialects, the prefix has another form. Before a lip consonant the wə́- becomes ó-, as in ó-be-gərdzi, "we will walk". [2]

Verbs with a prefix

Many verbs have no wə- at all. They are built from a prefix and a stem, and the perfective is made by moving the stress to the prefix. Tegey and Robson call them doubly irregular and give two verbs that differ in no other way: [1]

• kenaw-, "plant" (imperfective present), and kénaw- (perfective present)

• past forms kenawál- (imperfective) and kénawal- (perfective)

They add that in many of these verbs the shift of stress is the only difference between the two aspects, that stress here is "not reflected in the spelling at all", and that other doubly irregular verbs also change stems, as with "take", imperfective byāy- and perfective bóz-. In perfective forms the stress is on the first part of the verb and in imperfective forms on the last or next-to-last syllable. [1] The pairs dar-kawál, "to give to you", and rā-kawal, "to give to me", show the prefixes in front of the same verb. [1]

David sorts verbs into three conjugation classes by how they form the aorist, which is a tidy way to see how much of Pashto aspect is stress. First conjugation verbs add wə́-. Second conjugation verbs, all of them prefix plus stem, "form the aorist by shifting the stress to their prefix", and if the prefix has more than one syllable the stress goes on the first. Third conjugation verbs are verbs made from a noun or adjective with kawəl, "to do", or kedəl, "to become"; they shift the stress from the verb to the noun or adjective and use an irregular aorist form of the verb. [2] David adds in a footnote that Tegey and Robson, though they are not explicit about it, also sort verbs into three groups by aorist formation, a wə- prefix or a stress shift, together with the structure of the verb. [2]

A small class of verbs that begin with a behaves differently. David says that unlike most Pashto verbs, their stress in the continuous aspect is variable, either initial or not, and that in the aorist the initial stress is obligatory. [2]

Stress on endings

Tegey and Robson describe a number of endings that always take the stress, so that a word with one is stressed on the ending and not on the root. One of the masculine plurals, -úna, is the clearest case. Five pairs from their book: [1]

• yarghál, "invasion", and yarghalúna

• surúd, "song", and surudúna

• afsún, "charm", and afsunúna

• anār, "pomegranate", and anārúna

• ormég, "neck", and ormegúna

The suffix -ég-, which makes verbs from nouns and adjectives, is also stressed and breaks the basic rule: rawānégəm, "I am starting out", and rawānégay, "you are starting out". [1]

Morgenstierne marked stress in his own examples with an apostrophe before the stressed vowel, and the plural endings he lists agree with Tegey and Robson. Shown with acute accents here, his plurals include ūxā́n, "camels", lāsū́na, "hands", špānə́, "shepherds" and saṛī́, "men", and for feminine nouns wə́ne, "trees" and lā́re, "roads". [3] His adjective tawd, "hot", moves its stress with the ending: tāwdə́ for the masculine oblique singular and tawdá for the feminine. [3] Morgenstierne notes that the linguist Dybo suggested this old pattern could go back to a Vedic accent, an idea Jerzy Kuryłowicz rejected, and he says Kuryłowicz's objections did not take account of all the Pashto facts. [3]

David describes the same thing as a property of word classes. Her adjectives of Class I are stressed on the stem and Class II adjectives take stressed endings, and which class an adjective belongs to "is a property of the lexeme and is not predictable". Some of her noun classes are split by whether the stress falls on the last or the next-to-last syllable. [2] A noun that ends in a stressed vowel keeps the stress on its inflected suffix, where an unstressed final vowel is dropped. [2]

Pronouns

Pashto has pronouns and demonstratives that are the same segments with different stress. David says the strong pronouns are stressed on the last syllable and the demonstratives on the first. Final stress marks a thing already mentioned, an anaphoric use, while initial stress introduces something new or points to it, which she attributes to Penzl and to Heston. [2] Her example of the final-stressed use is pas lə haɣə́, "after that" (پس له هغه). [2]

Stress decides where small words go

Tegey and Robson's chapter on sentences says that the particles kho ("but", "well", "then"), the future marker ba, the weak pronouns (me, "my" or "me", and the others) and the particle for "must" "are invariably positioned immediately after the first stressed element in a sentence or clause". That is why a word such as me, "my", can land away from the noun it belongs to. [1] Their example in script: [1]

د پلار کلا مې په لوګر کې ده.

de plār kalā me pa logár ke da.

My father's kala is in Logar.

The weak pronoun comes after kalā, the first stressed phrase, and not beside plār, "father", although it means "my father's". [1] What counts as the first stressed element can be a subject, an object, a prepositional phrase, an adverb or even the first part of a verb. [1] That makes stress a part of Pashto syntax as well as of its sound.

Stress is not written

Neither Tegey and Robson nor David finds a way to tell a reader where the stress is from the spelling. Tegey and Robson say the writing system "does not mark stress at all". [1] David puts Pashto on the scale of how easily a reader can get pronunciation from spelling "somewhere near English". [2] Short vowels are also unwritten, so the script leaves out both the vowels and the stress. For a pair such as áspa and aspá the reader works out the word from context. That is why grammars mark stress themselves, Tegey and Robson with an acute accent over the vowel, and why a learner needs recordings. [1, 2]

For the alphabet, the sounds and the other grammar, see The Pashto alphabet. The stress in wə́ and in the plurals is discussed there only in brief.

Stress in the dialects

Two things can be said. First, Wiktionary records the name of the language with a different consonant cluster in each dialect and the stress on the last syllable in all of them: Kandahar /pəʂˈto/, south-eastern /pəʃˈto/, central /paçˈto/, Kabul /pəxˈto/ and Peshawar /pʊxˈto/. [4] Tegey and Robson's Kabul-based form is paxtó. [1] The dialect split described in Pashto dialects is a difference of consonants and vowels, and the final stress stays the same.

Second, David writes up three conjugation classes and the aorist stress shift for General Pashto, her name for the four main dialects, and for the Middle dialects of Waziri and Dzadrani. She says second conjugation verbs in the Middle dialects "behave similarly" to those of General Pashto, with the Dzadrani ó- for wə́- before a lip consonant as one difference. [2] Waziri nouns follow the same rule that a stressed final vowel keeps the stress on the suffix. [2]

Stress in verse

Tegey and Robson state that stress patterns in words and sentences have not been studied much. [1] One fact is clear in David's grammar. The first person singular ending -ama (and its short form -am) is "heard most often in poetry, where meter determines" whether the one-syllable or two-syllable form is used. In Waziri and Dzadrani both forms occur in poetry and in ordinary speech. [2] For a poet, a syllable can therefore be added or dropped to fit the line, which is an example of how the shape of a word and the shape of a line interact.