Unlike English, Ancient (e.g. Attic) Greek does not reorder words to formulate a question. The particle "ἆρα" does modify a statement into a question, but is not always present. In that case, I presume there would have been a way for native speakers to hear the difference, perhaps in a change in overall intonation.

What device did spoken Ancient Greek employ to signal a question when the particle ἆρα was not used?

Since there are no native speakers of Ancient Greek around, I imagine having a definite answer to this question is impossible. In that case I would be happy with an educated guess, or a pointer towards current scholarly debate.

  • 2
    I know very little about Attic Greek, but note that, if questions were marked intonationally, the intonation may not have been rising. While rising intonation for questions is a tendency across languages, it is not a universal. See this section of the Wiki article on intonation as well as my answer to another question about question intonation. Commented Sep 26, 2012 at 13:36
  • Since there are no native speakers around anymore, what evidence would there be of intonation? I don't know of any scripts that denote it, for instance. If everybody who had writing at the same time as the ancient greeks and who knew about the ancient greeks also had the same intonation, I doubt they would have bothered to record that fact. Treatises on dialect, perhaps.
    – kaleissin
    Commented Sep 27, 2012 at 6:19
  • 1
    @kaleissin It's not likely, yes but who knows... Besides this is a greatly interesting question for someone that loves Greek like me eheh :) By the way, if someone (with reference) wants to answer "Nothing is available to say that at the moment", that is still an answer. :)
    – Alenanno
    Commented Sep 27, 2012 at 8:19

3 Answers 3


As you say, reconstructing the sound of a language for which there are no longer any speakers can be very difficult, so it's unlikely that there is a definite answer to your question. However, you might start with A. M. Devine and Laurence D. Stephens, The Prosody of Greek Speech (Oxford, 1994). The authors use evidence from cross-linguistic studies to build a theory of prosody, and then apply that theory to classical Greek texts. It has a section on questions, and even if that doesn't help to answer your question, it also has a really good bibliography which might help you to track down more information.

[Edit by B. Crowell]

The relevant material in Devine seems to be at pp. 452-455. Aristotle's Poetics says that in acting, words are pronounced differently to show that they are commands or questions. Question words like τίς have accents that don't change to graves the way they do in most words, indicating that they had high pitch. In other tonal languages, it's common that the pitch pattern gets modified in questions.

  • What do the authors suggest for an answer to the OP's question?
    – minseong
    Commented Apr 5, 2021 at 15:27

Some help may come from the earliest punctuation systems, which used a upward rising line to indicate a question (an early form of our question mark); the shape of this line is supposed to be a quasi-music note and to indicate to the reader that (s)he needs to fluctuate the voice. Granted these come from well into the common era and are mostly confined to the Latin west, but one could look into similar punctuation systems in the Greek east; this may give you some further information.


A very interesting and intriguing question.

First one has to distinguish between open questions (Where have I put my car keys?) and yes-no-questions (Have I forgotten my car keys at your place?). In open questions the question is marked by interrogative adjectives, pronouns (τίς) or adverbs (ποῦ, πότε) which probably had a different pronunciation than their indefinite counter parts, as one could deduce from their different accentuation (τις, τινές / που,πού / ποτε, ποτέ).

In Russian -a languages that doesn’t mark questions by automatically inverting the inflected verb and the subject either- open questions usually don’t have a rising intonation, whereas yes-no-questions have; the question is marked by words like ‘Where? What? Who?’ at the beginning of the sentence.

One could assume that in Greek likewise the open questions were only marked by a different pronunciation of the interrogative word at the beginning of a sentence.

A different case are the yes-no-questions; in Russian these are either marked by the interrogative marker ли (but it sounds a bit elevated) or by simply pronouncing the sentence with a raised tone towards the end.

In Ancient Greek the interrogative particle ἆρα is often not present in sentences that are clearly intended as questions (Plato’s dialogues), like it isn’t in Russian. Another similarity between both languages is the rather free word order which is used to convey nuances of stress on particular information in a sentence or reported speech;

Therefore, I can hardly believe in Ancient Greek declarative sentences weren’t distinguished from yes-no-questions without ἆρα by a certain way of changing the melody/intonation in the question; there must have been a difference in melody/intonation to mark the spoken sentence as a question, mustn’t it?

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.