In many Indo-European languages, you use the verb "to have" to describe the past. For example:

"I have been",

"J'ai été", (French)

"He estado" (Spanish)

"Jag har varit". (Swedish)

What is the reason behind this? Why do we use the verb "Have" (or to be in some cases), and not any other verbs?

  • 2
    It isn't always the past. The Romance future tenses are formed by conjugating have as a suffix on the infinitive, from an earlier auxiliary verb. The real question is why Indo-European repeatedly uses a verb meaning possession as an auxiliary.
    – jlawler
    Commented Jun 6, 2021 at 21:50

3 Answers 3


The periphrastic 'have'-perfect isn't a feature common to the Indo-European languages, but rather one that's part of the Standard Average European Sprachbund, which developed as the various European languages came into closer contact, first through the Roman Empire and then through the various migrations and the general wider trade of the Middle Ages.

The 'have'-perfect starts showing up sporadically in Vulgar Latin in the 2nd century CE (Classical Latin didn't have it), and became grammaticalised in the subsequent centuries. It's not clear if it originally came from some other language, but after it took root in Vulgar Latin it was inherited by the Romance languages, and from there it spread further (e.g. High German developed its 'have'-perfect in the 8th century).

Not all SAE languages have it, but it exists in all Romance and Germanic languages (Romance and the West-Germanic languages are generally considered to be the core of the Sprachbund), plus Albanian, modern Greek (but again, not Classical Greek), and a few Slavic languages. The meaning of the construction isn't always exactly the same—in some languages it's a normal perfective past, in others, including English, the meaning is distinctly present-anterior—but it's the same borrowed construction.


This is a feature of Standard Average European (SAE), a Sprachbund across much of Europe. We know it's not an Indo-European genetic feature, since it's not generally found in Indo-European languages outside SAE (including e.g. Classical Latin).

The most prominent theory I've seen claims that this feature evolved in Latin, and spread from there.

Classical Latin marks both tense and aspect on verbs, with a slightly defective system: "I ate" (past aoristic) and "I have eaten" (present perfective) are both indicated by the same form. Classical Latin also has a very common past participle, which was often used in sentences like this:

Habeō scriptās litterās.
I have written letters. (That is, I am holding letters that have been written.)

Over time, the habeō here started to lose its semantic meaning, and could be used even when you weren't literally "holding" anything:

Habeō scriptās litterās.
I have written letters. (That is, I previously wrote some letters.)

This filled in the gap in the system, allowing Latin-speakers to distinguish between "I ate" and "I have eaten". According to some textbooks, this development may have happened as early as the first century BCE and shows up in Vitruvius; according to others, it was a later, Vulgar development.

This then evolved into various Romance forms: Latin habēre scriptum "to have written" became French avoir écrit, for example. And the feature spread from there into various Germanic languages, even though English "have", German haben, etc, aren't cognate to Latin habēre. It's now found in many of the SAE languages, to the point that this feature is often used as part of the definition of the Sprachbund.

  • 3
    It’s not only an SAE thing either. In Mandarin and other Chineses, for example, the negated form of 有 yǒu ‘have’, 没(有) méi(you) (or equivalent forms), is used to negate past-tense verbs, and the past-tense form of affirmative-negative questions uses 有没有 yǒu méiyou. Commented Jun 7, 2021 at 0:13

The pattern "to have + Participle" is the result of a long process:

  1. it needs the invention of the copula "to be"
  2. it needs the creation of the Passive-Resultative pattern: to be + participle, as in "the food is cooked"
  3. it needs the invention of a verb "to have". Originally, IEan languages used to say "to me (is) a house".
  4. it needs the creation of a kind of Stative pattern: I have (at hand) cooked food
  5. then, final step, I (have cooked) food, a true Perfect tense.
    Obviously all this process took a lot of time, possibly more than one thousand years.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.