Last updated 2025-09-29 21:06:43

A Note On Grammatical Terminology

Most works on Romanian grammar distinguish between NOM/ACC and GEN/DAT when there are only two distinctions (such as noun declensions). To save on space, I will be using NOM and OBL (oblique, as in, "the other form"). In some cases, when there is still a distinction between more cases, such as personal pronouns, the individual case will still be used (ACC, DAT, ...).

Pronouns

Personal Pronouns

Dumneavoastră is the polite pronoun, similar to vous in French. However, there is a separate pronoun, voi, for 2PL that does not overtly indicate politeness. There is also the pronoun dumneata that sits between tu and dumneavoastră in politeness and uses the 2SG form of verbs, etc. The polite pronouns extend to 3rd person domain as well. For 3SG, in increasing politeness: dânsul/dânsa, dumnealui/dumneaei, domnia sa. For 3PL: dânșii/dânsele, dumnealor, domniile lor.

Ele is used like elles in French, when the group being referred to is only women (or female nouns).

There is a long form of ACC, which uses the preposition pe. In all but 1SG and 2SG, it uses the NOM pronoun. In the case of 1SG, it is pe mine and 2SG it is pe tine. For reflexive verbs, the verb is marked with se, meaning "himself" or "herself". That is, one does not use îl or o with reflexive verbs to mean himself or herself, it is always se.

There are three DAT forms listed. The first (e.g. "mie") is the stressed form and can often be dropped. The second form (e.g. "îmi") is the unstressed form and can be commonly found in phrases like îmi place:I like and îmi pare rău:I'm sorry. The third is the shortened version of the second form.

"Eu" may be "io" in Transylvania (and Italy, of course).

Possessive Pronouns

Possessive pronouns differ based on possesor, gender, and number of the noun being possessed. To find the right form, first find the correct table for the possessor, e.g. "yours" would be 2SG. Then, find the gender of the noun being possessed, and then the plurality. For example, "my book" is carta mea – mea because I is 1SG, "book" is F, and there is one of them (SG).

3SG.M: lui
3SG.F: ei
3PL: lor

Some trends you can observe:

• 1SG and 2SG in PL share the same endings as the definite article.
• the M.SG form always ends in -u; M.PL, -i; F.SG -a/ă; F.PL -e

Nouns

Plurals

Some languages make plurals easy, some make them difficult. Romanian is one that chose the more difficult path. As with most things, explicit studying of the plural, along with the singular, is probably most beneficial. However, it is not all logic-less and with enough exposure you'll pick up on the phonetic patterns that dictate what the plural of most common words should be.

As you get started, here are some general rules. On the whole, if you are unsure, adding i or changing the last vowel to i will probably get the point across.

• Masculine nouns tend to either add i (if the noun ends in a consonant, ex. elev → elevi) or change the final vowel to i (câine → câini). Many will experience sound changes as a result: perete → pereți.
• Feminine nouns have more rules based on ending: -ă may become -e, -i, or -uri. -a becomes -i (note that in the case of V + ie, it will be V + i). -a adds le, whereas -ea becomes -ele (ex. pijama → pijamele and cafea → cafele ). -i will stay -i.
• Neuter nouns follow a similar breakdown as feminine nouns, but tend towards -e: C, -u, or -i will be -uri or -e. -iu will become -ii and -e will stay -e.

There are, of course, exceptions. ou:egg → ouă:eggs for example.

Articles

Definite

Unlike English, the definite article (i.e. "the") is suffixed to the noun, rather than as a standalone word. The suffix changes based on the case, gender, and number of the noun.

There is some nuance with the endings:

1. Add -le for M if the word ends in -e, e.g. perete > peretele.
2. The ending for F changes to a except in two cases where the -(u)a is added: endings in -e have a added (marea) and endings of -a and -i add ua (cafeaua, ziua).
3. Days will change -i to ea, except for joi:Thursday, which takes a (vinerea vs. joia).

The definite article is also used with the POSS pronoun, ex. tatăl meu:my father.

Indefinite

The indefinite article in Romanian is "o" for feminine nouns, and "un" for masculine/neuter. Unlike the definite article, it is a standalone word that comes before the noun, ex. o carte:a book

Verbs

Verbs are negated with nu. This can appear before verbs starting with vowels as "n-".

Infinitive

The infinitive is indicated by a {verb}. When verbs are used serially, the a becomes să. For example, Îmi place să citesc, să fac ciclism ⧸ I like to read and to go biking.

Present Tense

The Romanian present tense is not particularly regular. By one categorization, such as on dexonline, derived from Lombard and Gădei (1981), there are 667 different types of verbs. Guțu Romalo reduces that down to 10 main categories. Traditionally, there are 4 categories, but there are many variations within each category. Feldstein does something, but it's not very clear if this structure of Romanian verbs is useful to the learner.

The below is based on Guțu Romalo categorization, where the combination of infinitive ending, 1SG.PRS ending, and the participle are sufficient to uniquely identify a category. To use this, drop the vowel ending of the infinitive, given at the top of each column, and then add the appropriate ending for number and person. As you'll notice, some infinitive ending are duplicated, thus you will need to know at least the 1SG conjugation to know which column to use (cf. a-∅ and a-ez as a clear example). The final category from Guțu Romalo (rupe) is not given here, since the PRS endings do not change.

† this is somewhat less regular than implied: "ă" can become "a"
‡ in 3SG and 3PL, the "r" of the verbal root is dropped if there is one, e.g coborî to cobo+ară. This will seem more regular (and inline with other conjugations for 1PL and 2PL) if it is recalled that â is just î in the middle of words (and in prior spelling reforms, were not distinguished).
ª "ă" becomes "a" except for 1PL and 2PL.

Some trends to be aware of:

• a, i, e have variations where additional consonants are added (ez, esc, z, respectively)
• 1SG: the ending vowel is always dropped
• 2SG: always ends in -i
• 3SG: infintives ending in "a/î" end in "ă", otherwise, "e"
• 1PL: always ends in -m
• 2PL: always ends in -ți
• 3PL: will either match with 1SG or 3SG (but I'm not sure if this is helpful)

Past Tense

PST is formed with a specific declension of "to have", plus the participle.

Thus, for the verb a lua:to take, "I took" would be "am luat", where luat is the participle of "a lua". The participle should be memorized along with the infinitive and 1SG.PRS form.

3SG a may be o in regional dialects.

Future Tense

FUT is formed with the auxiliary verb seen below and the infinitive, without "a". For example, Noi vom mînca ⧸ We will eat.

Conditional Tense

COND is formed with the auxiliary verb seen below adn the infinitive, withou "a". For example, Eu aș dori... ⧸ I would like...

Reflextive

The verb will be marked in most dictionaries with se, meaning "himself" or "herself". That is, one does not use îl or o with reflexive verbs to mean himself or herself, it is always se.

Reflexive-Like Statements

I'm not sure how these are formally categorized, but there are some phrases, such as expressing hunger or thirst, that act like reflexive verbs, even though they are not verbs (e.g. in the case of hunger, hunger is a noun, not a verb meaning "to feel hungry").

The structure is: shortened DAT pronoun + i/e + the noun. As an example, Mi-i foame ⧸ I am hungry, or more literally: to me (it) is hunger. I/E in this case function as "it is" (este).

Irregular Verbs

Some common irregular verbs.

• a ști:to know, is știu in 1SG.PRS
• a pot:to be able to, can: pot, poți, poate, putem, puteți, pot

To Be (a fi)

The 3SG form is sometimes abbreviated "e". The particple of a fi is fost. It shares this participle with a merge:to go. Thus, eu ai fost could mean "I was" or "I went", depending on context. A merge also has a verb specific participle, mers, if you need to distinguish.

Simple Romanian, lesson 4 gives some Transylvanian regional variations:

• 1SG/3PL: îs
• 3SG: îi (which may appear as -i)
• 1PL/2PL: "u" may be pronounced as "â", i.e. sântem & sânteți (see note on spelling reform below)

In either case, îs and sunt may be shortened to -s in casual speech. For example, unde-s:where are they, astea-s:these are, etc.

Before the 1993 spelling reform, 1SG, 1PL, and 2PL would be spelled sînt, sîntem, and sînteți, which may be common with older speakers and speakers in Moldova. Pre-1953, î would be â. I look forward to another spelling reform in 2033.

Participles

Participle are generally formed by adding -t to the the infinitive. Ex. "a lua" → "luat". Based on the infinitive endings we have:

• a: -t
• ea: -ut (drop -ea, "părea" → "părut")
• e: -ut, -s, -t
• i: -t
• î: -t (spelling rules will change final î to â)

Possession

Something of Something

(rewrite so it's NOM and OBL)

the father of my wife: tatăl soției mele
tatăl: the father (NOM-def)
soției: the wife (DAT-def)
mele: mine (POSS-1SG-DAT-determiner)

the son of my friend: prietenul fiului meu
preiteunul: the friend (NOM-def)
fiului: the son (DAT-def)
meu: mine (POSS-1SG-DAT-def)

Proper Nouns

All masculine names use lui to mark possession: casa lui Bogdan – Bogdan's house. Feminine names use lui only when there is not a way to decline the name in OBL to mark the possession. In casual speech, lui may be used for all feminine names, too.

Comparisons

mai cald decît... warmer than
mai cald decît vara trecută: it is warmer than last summer
îmi place ... mai mult decît ...: I like ... more than ...

Numbers

• 1: unu
• 2: doi
• 3: trei
• 4: patru
• 5: cinci
• 6: șase
• 7: șapte
• 8: opt
• 9: nouă
• 10: zece

The teens are formed by combining the ones with ten (as sprezece, lit. "to ten"), with some small phonetic variation.

• 11: unsprezece
• 12: doisprezece
• 13: treisprezece
• 14: paisprezece
• 15: cincisprezece
• 16: șaisprezece
• 17: șaptesprezece
• 18: optsprezece
• 19: nouăzprezece

Tens are formed by the ones + zece, with two small phonetic variations:

• 20: douăzeci
• 60: șaizeci

Otherwise, it is a form like treizece:30. With ones, it is douăzece și unu:21.

Additional larger numbers (note the form changes for 200 and higher):

• 100: o sută
• 200: două sute
• 1000: o mie
• 1000000: un million

Oridinal Numbers

Original numbers, except for "first", which is primul, are formed with the structure al {num}lea. Thus:

• 2nd: al doilea
• 3rd: al treilea
• 4th: al patrulea
• 100th: al sutălea
• etc.

Prepositions

de

of, from

din

from a place

într (-o, -un)

inside

la vs în

As with so many languages, some prepositions don't seem to have any straightforward rules. Both la and în mean "in", among other meanings, but their use depends somewhat on what thing you're in:

la

places, restaurants

în

countries, restaurants, transportation

Neither are used with being at home, which is always acasă.

Sentence Structures

Liking Something

mamei mele îi place... : my mom, she likes...

Contractions

While not necessarily a "structure", they are very common in Romanian. Outside of perhaps the most basic introductory texts, you will encounter them immediately. You can find a table of contractions on Wikipedia, as well as some explanation on Reddit. It is easier to review that table and get a good sense of what the contractions are. In general, when a vowel is elided, it will be natural to elide the same vowel in casual speech.

One contraction that is not given in that table is -n, which is when în occurs after a word ending in a vowel.

Adjectives

Adjectives change according to the number and gender of the thing being described. M.SG is the default form and the other forms change based on its form. M.PL is used whenever there is at least one male in the group.

Ending Changes

Some endings have regular sound/form changes

-d or -t
M.PL is -zi and -ți, respectively (ex. flăminzi).

-ist
M.PL is -iști.

Miscellaneous Links on the Web

• Reddit post on slang