I previously asked about indicating a question in writing, and I got interested in how questions were recognized in the first place. To know if a question mark would be used after a question, one should probably know whether or not such sentences were recognized in the first place. Certainly the concept of a question was known to the Romans, but I want to know whether they made a formal distinction between a declarative and interrogative sentence.

When did grammarians first recognize a question as a type of a sentence? Was it discussed from the very beginning of grammatical literature, or was such classification of sentences a later development?

I am interested in Latin, but if there is something relevant in Greek literature, feel free to share it.

  • 2
    When you say "question," do you mean a "formal distinction between a declarative and interrogative sentence"? I assume so, because the idea of a "question" per se is as primitive as it gets.
    – brianpck
    Mar 9, 2017 at 22:11
  • In line with @brianpck I find it difficult to prove what seems most plausible: nearly as soon as there were grammarians, there must have been questions classed as such, because 1) there is an apparent difference in grammatical structure and 2) there is common vocabulary to name questions ([inter]rog-)
    – Rafael
    Mar 10, 2017 at 2:08
  • 1
    I don't think I have enough information to make this into a full-fledged answer, but have you read this ? en.wikipedia.org/wiki/Five_Ws#History
    – blagae
    Mar 10, 2017 at 8:39
  • 2
    @Rafael, though that assumption does seem plausible, it's possible that the earliest classical grammars were concerned with morphology to the exclusion of other areas like syntax; the grammar of Dionysius Thrax, which if genuine (there is much debate on this point) is our earliest extant grammar, from the second century BC, says nothing about sentence types, as far as I can find.
    – TKR
    Mar 11, 2017 at 1:16
  • 1
    "He asked a question." Is that not implicitly and inherently classifying types of sentences? You don't ask a command, right? Unless you're looking for a grammatical treatise on the subject...
    – cmw
    Mar 15, 2017 at 1:03

2 Answers 2


In the history of linguistics, Protagoras (5th century BCE) is assumed to be the first scholar ever who classified sentences into sentence types. Naturally, we don't have any textual evidence except Diogenes Laertius who says the following:

διεῖλέ τε τὸν λόγον πρῶτος εἰς τέτταρα, εὐχωλήν, ἐρώτησιν, ἀπόκρισιν, ἐντολήν: 8 [54] ῾οἱ <:>ὲ εἰς ἑπτά, διήγησιν, ἐρώτησιν, ἀπόκρισιν, ἐντολήν, ἀπαγγελίαν, εὐχωλήν, κλῆσιν̓, οὓς καὶ πυθμένας εἶπε λόγων. Ἀλκιδάμας δὲ τέτταρας λόγους φησί, φάσιν, ἀπόφασιν, ἐρώτησιν, προσαγόρευσιν.


"He was the first to mark off the parts of discourse into four, namely, wish, question, answer, command; others divide into seven parts, narration, question, answer, command, rehearsal, wish, summoning ; these he called the basic forms of speech. Alcidamas made discourse fourfold, affirmation, negation, question, address."


The general rule for distinguishing a question in Latin is the suffix "-ne". For example, LicetNE mihi ire ad latrinam roughly translates to Permission for me to go to the bathroom?

  • 1
    You are right, a question is often marked with -ne or a question word of some kind. But this is not what I asked about; I wanted to know if the Romans made a distinction between questions and statements (and maybe something else) at the level of grammatical theory.
    – Joonas Ilmavirta
    Apr 1, 2017 at 5:42

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.