Grammaticalization Paths in Modern Persian

Modern Persian grammar contains forms that once had greater lexical or discourse independence. را developed from an older relational postposition toward differential object marking; adverbial همی was reduced and bound as imperfective می‌; lexical داشتن، خواستن, and شدن acquired progressive, future, and passive functions in particular constructions.

This historical process is grammaticalization, but it is not a story of words becoming empty overnight. Older and newer functions coexist, intermediate uses persist, and different periods preserve different layers. A plausible semantic bridge explains how a change could occur; only dated texts, corpora, and comparative evidence show that it actually occurred.

A map of the five paths

FormEarlier or lexical sourceModern grammatical functionEvidence status
راolder relational postposition “for, with respect to”differential marking of direct objects; some topic-like remnantsdocumented across Old, Middle, and New Persian; details debated
همی ← می‌adverb with continuative/habitual forcebound imperfective prefixintermediate Early New Persian forms are textual evidence
داشتنpossessionprogressive auxiliary before a finite می‌-verblate and expanding Modern Persian innovation
خواستنvolition/intentionformal future; emerging prospective uses in speecholder formal auxiliary and newer colloquial pathway must be distinguished
شدنmotion/change of statechange-of-state light verb and passive auxiliarylayered development; earlier New Persian also used other passive strategies
💡
“Source meaning → grammatical meaning” is a historical hypothesis, not a license to translate the modern construction literally. Modern دارم می‌نویسم means “I am writing,” not “I possess writing.”

را: relation, topic, and direct object

Historical work relates modern را to Old Persian rādiy and Middle Persian rāy, a postposition with meanings such as “for, on behalf of, concerning.” Middle Persian rāy was not simply the same direct-object marker found in modern standard Persian. Across New Persian, the form extended toward core object marking while retaining traces of broader discourse-related functions.

Synchronically, را follows a direct object that the speaker treats as identifiable, specific, or discourse-linked. It does not mark every object.

گزارشی را که فرستادی با دقت خواندم.

gozâreshi râ ke ferestâdi bâ deqqat khândam.

I carefully read the report that you sent. (neutral standard; identifiable direct object)

در قطار چند گزارش قدیمی خواندم.

dar qatâr chand gozâresh-e qadimi khândam.

I read several old reports on the train. (neutral standard; nonspecific object without را)

Topic-like uses outside a canonical direct object survive in marked spoken or literary environments, but they do not justify teaching را as a general topic particle. Modern differential object marking is the productive center of the system.

💡
Historical breadth helps explain why را interacts with discourse status, but present-day learners should still determine whether the phrase is the direct object and how specific or identifiable it is.

همی to می‌: an adverb becomes a bound prefix

Early New Persian texts show همی or reduced می with durative, habitual, and continuative force. The older element had freer placement than today's prefix. Phonological reduction and tighter attachment produced modern می‌, written with a ZWNJ before the verb.

بچه‌ها عصرها در حیاط بازی می‌کردند.

bachche-hâ asr-hâ dar hayât bâzi mi-kardand.

The children used to play in the courtyard in the evenings. (neutral standard; past habitual imperfective)

وقتی رسیدم، باران آرام می‌بارید.

vaqti residam, bârân ârâm mi-bârid.

When I arrived, rain was falling gently. (neutral standard; ongoing background)

The two examples show why calling می‌ a “present prefix” is historically and synchronically inadequate: it occurs with a past stem in می‌کردند and می‌بارید. Its central contribution is imperfective viewpoint, which can support habitual, ongoing, generic, and other context-dependent readings.

Classical همی remains recognizable (literary/archaic), but inserting it into ordinary modern speech does not recreate neutral می‌. Register and period are part of the grammar.

داشتن: possession becomes a progressive auxiliary

Lexical داشتن introduces something possessed. Progressive داشتن precedes another finite verb already marked by می‌, and both verbs agree with the same subject.

برای نوشتن گزارش هنوز وقت دارم.

barâ-ye neveshtan-e gozâresh hanuz vaqt dâram.

I still have time to write the report. (neutral; lexical possession)

دارم آخرین بخش گزارش را می‌نویسم.

dâram âkharin bakhsh-e gozâresh râ mi-nevisam.

I am writing the final section of the report. (neutral spoken standard; progressive auxiliary)

Corpus-based research places the rise and expansion of this progressive in the modern period; descriptions identify it by the late nineteenth century, and twentieth-century literary data show increasing use. The construction is therefore much newer than imperfective می‌. With achievement predicates it can also support an imminent or prospective reading:

مواظب باش؛ لیوان دارد می‌افتد.

movâzeb bâsh; livân dârad mi-oftad.

Watch out—the glass is about to fall. (neutral spoken standard; prospective edge of the progressive)

This does not mean داشتن has become a general future auxiliary. The imminent reading depends on an event whose onset is visibly developing.

Volition naturally points forward: an intention concerns a not-yet-realized event. Persian has a grammaticalized formal future with inflected خواهـ followed by a bare past stem, alongside ordinary lexical خواستن + subjunctive.

می‌خواهم پیش از پاسخ با همکارانم مشورت کنم.

mi-khâham pish az pâsokh bâ hamkârân-am mashverat konam.

I want to consult my colleagues before replying. (neutral; lexical volition)

نتیجه‌ی بررسی در گزارش بعدی منتشر خواهد شد.

natije-ye barrasi dar gozâresh-e ba'di montasher khâhad shod.

The result of the review will be published in the next report. (formal written future)

The first does not entail that the consultation will occur; the second presents a future event without a meaning of desire. Modern conversation usually uses a futurate present instead of خواهد.

A newer colloquial construction with می‌خواد can signal an imminent development or speaker inference:

انگار می‌خواد برف بیاد؛ زودتر راه بیفتیم.

angâr mi-khâd barf biyâd; zudtar râh biyoftim.

It looks like it's going to snow; let's set off earlier. (informal, regional: Tehran; prospective inference)

This emerging use is not simply the colloquial pronunciation of formal برف خواهد بارید. It retains constructional links to appearance and imminence and has not displaced other future strategies.

شدن: motion, change, and passive

The history of شدن is layered. Older Iranian stages preserve motion-related meanings; New Persian developed “become” and a wide family of change-of-state predicates. Modern Persian also uses a past participle plus finite شدن for the periphrastic passive. Early New Persian sources show other passive resources, including constructions with آمدن, so the modern passive was not inherited as one unchanged template.

نزدیک غروب هوا ناگهان سرد شد.

nazdik-e ghorub havâ nâgahân sard shod.

Near sunset, the weather suddenly became cold. (neutral; lexical change of state)

نسخه‌ی نهایی قرارداد صبح امضا شد.

noskhe-ye nehâyi-ye qarârdâd sobh emzâ shod.

The final version of the contract was signed in the morning. (formal-neutral; conventional passive predicate)

در انبار بر اثر فشار باد باز شد.

dar-e anbâr bar asar-e feshâr-e bâd bâz shod.

The warehouse door opened because of wind pressure. (neutral; inchoative rather than necessarily passive)

The last two show a synchronic boundary: امضا شد is naturally linked to an unexpressed signer, while باز شد can simply present a change. Historical source meaning does not settle the modern parse.

💡
Grammaticalization creates layered constructions, not one universal meaning per verb. Diagnose شدن from the nonverbal element, participial form, participants, and context.

Synchronic layering in one passage

داریم نسخه‌ی نهایی را آماده می‌کنیم. اگر امروز تأیید شود، فردا منتشر خواهد شد.

dârim noskhe-ye nehâyi râ âmâde mi-konim. agar emruz ta'yid shavad, fardâ montasher khâhad shod.

We are preparing the final version. If it is approved today, it will be published tomorrow. (spoken-standard progressive followed by formal passive future)

This passage contains progressive داشتن, object را, imperfective می‌, passive شدن, and future خواهـ. Their sources do not compete; modern grammar can stack the resulting layers in a single discourse.

Evidence and uncertainty

The development of Early New Persian auxiliaries and particles is summarized from dated textual evidence in Ludwig Paul's scholarly account of Early New Persian. The extension of را across New Persian functions is documented in a diachronic semantic-map study. Modern داشتن progressives have been studied both as an auxiliation path and through a twentieth-century literary corpus.

These sources do not require one theoretical decomposition. Dates of first attestation are always provisional because an earlier example may be found, and textual frequency is not identical to spoken frequency. The page therefore distinguishes securely attested form sequences from explanatory semantic bridges.

Common Mistakes

1. Projecting the modern function unchanged into the past

❌ را در همه‌ی دوره‌های فارسی فقط نشانگر مفعول مستقیم بوده است.

Incorrect — its older relational functions and gradual extension are historically documented.

✅ نقش را در طول تاریخ فارسی تغییر کرده و گسترش یافته است.

naqsh-e râ dar tul-e târikh-e fârsi taghyir karde va gostaresh yâfte ast.

The role of râ has changed and expanded over the history of Persian. (academic statement)

2. Treating modern می‌ as an independent adverb

❌ بچه‌ها می در حیاط بازی کردند.

Incorrect modern syntax and orthography — می‌ is a bound prefix attached with a ZWNJ.

✅ بچه‌ها تا تاریک شدن هوا در حیاط بازی می‌کردند.

bachche-hâ tâ târik shodan-e havâ dar hayât bâzi mi-kardand.

The children played in the yard until it got dark.

3. Reading progressive داشتن as possession

❌ دارم مقاله می‌خوانم یعنی مقاله‌ای در مالکیت من است.

Incorrect — in this construction داشتن contributes progressive aspect, not ownership.

✅ دارم مقاله‌ی تازه را برای جلسه می‌خوانم.

dâram maqâle-ye tâze râ barâ-ye jalse mi-khânam.

I am reading the new article for the meeting.

4. Equating volition with future assertion

❌ می‌خواهم بروم همیشه همان معنای خواهم رفت را دارد.

Incorrect — wanting to go does not assert that the going will happen.

✅ می‌خواهم بروم، اما شاید فرصت نکنم.

mi-khâham be-ravam, ammâ shâyad forsat na-konam.

I want to go, but I may not get the chance.

5. Calling every شدن predicate passive

❌ در جمله‌ی «هوا گرم شد»، گرم شد یک مجهول است.

Incorrect — the sentence describes a change of state and does not imply an agent who warmed the weather.

✅ با طلوع خورشید هوا گرم‌تر شد.

bâ tolu'-e khorshid havâ garm-tar shod.

The weather became warmer after sunrise. (change of state)

Key Takeaways

  • Historical grammaticalization must be established from texts and corpora; semantic plausibility alone is not evidence.
  • را moved from broader relational functions toward differential object marking, while older discourse-sensitive traces help explain—but do not replace—its modern syntax.
  • Older همی became bound imperfective می‌ through reduction and tighter attachment.
  • Progressive داشتن is a comparatively recent innovation; formal future خواهـ and emerging colloquial می‌خواد belong to distinct layers.
  • شدن spans change-of-state and passive functions, and modern context—not etymology alone—determines the analysis.

Now practice Farsi

Reading grammar gets you part of the way. The exercises are where it sticks — free, no signup needed.

Start learning Farsi

Related Topics

  • Grammaticalization of AuxiliariesC1Trace how داشتن, خواستن, شدن, and بودن acquire progressive, future, passive, and perfect functions while retaining lexical uses.
  • Historical Expansion of the Light-Verb SystemC1Trace how Persian expanded noun- and adjective-plus-verb predicates through internal development, language contact, lexical replacement, and modern coinage.
  • Present Progressive with داشتنA2Build the Persian present progressive with finite داشتن plus a matching می‌ present, including its imminent reading and limits on negation.
  • Future with خواستنA2Build the explicit Persian future with conjugated خواه plus a bare past stem, including negation, compound predicates, and its formal register.
  • Differential Object Marking with راB1Use Persian را according to an object's identifiability, specificity, animacy, and discourse status rather than marking every direct object mechanically.