Usually classified as member of the Tibeto-Burman branch of the Sino-Tibetan linguistic family.

It is a mildly tonal language using two to four tones depending on dialect. It is described as primarily isolating but agglutinative in some degree.

It is spoken by approximately 6 million Tibetan people across the Tibetan Plateau as well as by approximately 150,000 exile speakers.

Tibetan is written with a Sanskrit-derived script.

By means of agglutination, the Tibetan language has developed a considerable grammatical system of word suffixes and is no longer strictly isolating in structure. Agglomerations of consonants are often met with as initials, giving the appearance of telescoped words -- an appearance which historical etymology often confirms. Many of these initial consonants are silent in the Gtsang dialects, or have been resolved into a simpler one of another character. The language is much ruled by laws of euphony, which have been strictly formulated by native grammarians.

Since at least around the 7th century when the Chinese came into contact with the Tibetans, phonetics and grammar of Tibetan have been studied and documented. Tibetans also studied their own language, mostly for translation purpose for diplomacy (with China) or religion (from Buddhism).

Verbs do not inflect for person or number. Morphologically there are up to four separate stem forms called by the Tibetan grammarians present (lda-ta), past ('das-pa), future (ma-'ongs-pa), and imperative (skul-tshigs), although the precise semantics of these stems is still controversial. Most verbs which describe uncontrollable action lack an imperative.

The classical written language has nine cases: the absolutive, (unmarked morphologically), the genitive (-gi, -gyi, -kyi, -'i, -yi), the ergative/instrumental (-gi, -gyi, -kyi, -'i, -yi), the locative (-na), allative (-la), terminative ( -ru, -su, -tu, -du, -r), comitative (-dang), the ablative (-nas), and the elative (-las). Case morphology is affixed to entire noun phrases, not to individual words.

There are no numeral auxiliaries or measure words used in counting, as in many languages of East Asia, though words expressive of a collective or integral are often used after the tens, sometimes after a smaller number.