I learned once that ancient texts (for example in Latin) did not separate words. Was that always true or only in specific kinds of documents and writings.

Since I have been a bit interested in wordplays, like holorimes and phonological ambiguity, I started wondering whether the lack of separation between words could produce sentences that could have different meanings depending on how they were cut into words.

Is there any knowledge of ancient texts doing that, or similar wordplays.

According to wikipedia, Homer was adept to wordplay, but they are not more specific.

  • It's still true in many modern languages. Though I don't have any examples offhand, I can't imagine there not being any back then as I know I've seen some in modern works that do a similar play in spite of the spaces being there. Commented Mar 2, 2015 at 2:09
  • Asking if there are examples of wordplay from ambiguous word breaks is an interesting question, but the exact way this question is phrased brings in a lot of other sub questions and makes it too broad. I'd suggest editing it to focus it on that single question.
    – curiousdannii
    Commented Mar 2, 2015 at 5:24
  • It should be noted that some languages still do not put spaces between words. Chinese and Japanese are two examples. Korean has spaces, but only between phrases. I know that a lot of Chinese culture is based on wordplay, but I don't know how much Chinese wordplay has been based on the lack of spacing between words.
    – Zgialor
    Commented Mar 3, 2015 at 19:49
  • @Zgialor: In ancient texts, the hanzi are largely one character, one word, so there wasn't much room for such wordplay (I can't think of any at the moment) - but there definitely was wordplay based on lack of punctuation (though still not as important as e.g. puns, which are of extreme importance in Chinese culture). In modern Chinese this kind of wordplay is more common; for example, the sentence 獅子山下體現香港精神 could mean 'Embodying the Hong Kong spirit in Lion Rock', or it could also mean... something inappropriate, depending on whether you segment the words as [下][體現] or [下體][現]. Commented Jul 11, 2018 at 7:21
  • Thai, which uses an alphabet, doesn't space between words. Of course a lot of Thai letters are silent, which adds a piquant note to games based on written Thai.
    – jlawler
    Commented Jan 9, 2019 at 16:24

5 Answers 5


Since you are asking for any knowledge, I assume that the one example I can think of should be an answer. In the Ancient Greek joke collection Philogelos, there is a joke where a man calles his wife "He Hera mou", 'the Hera mine' = my Hera; she calls him "Ozeus mou", 'stinker mine', my stinker, which at that time or in that dialect was homophonous with "Ho Zeus mou", 'the Zeus mine' = my Zeus. Yes, it's hilarious.

(But since the joke was obviously meant to be told aloud, I don't know whether it's the kind of thing you were asking for. On the other hand, all ancient literature was meant to be read aloud.)


Aristophanes (Knights 21–26), much earlier than the Philogelos, punned on repeating molōmen auto, molōmen auto "let us go, that" ending up sounding like the taboo automolōmen "let us desert".

Remember, Aristophanes in Frogs mocks Euripides for adopting new-fangled notions from the sophists, like "word". As Willi's monograph on Aristophanes' language points out (can't find it, but it'll be around p. 58), only Euripides uses the word lexis "word" in the play. The old-fashioned epos "utterance" was good enough for Aristophanes' hero in the play, Aeschylus; and an utterance could be as short as a "huh" and as long as... an epic.

So ignoring word boundaries in word play was the status quo ante.


I believe the first recorded occurrence of a pun in written text was around 2100 BC, in the language of the Sumerians.

In the epic of Gilgamesh when Utnapishtim warns the ruler of his city about the flood, he does so by saying that the sky will rain "kibtu (corn)" and "kukku (sound of kernels hitting the ground)" as a pun on "kibitu (misery)" and "kukkû (suffering)" in Sumerian.

In the Penguin Classics English translation, he instead hides his warning as though he is forecasting a bountiful harvest.

  • This certainly seems like wordplay, but I think the question was specifically about wordplay based on ambiguous separation of text into words
    – b a
    Commented Jan 10, 2019 at 10:31

One piece of ancient wordplay is the statement attributed to the oracle at Dodona:

Ibis redibis nunquam per bella peribis

which, depending on how you group the words, can be taken to bear either of two meanings:

  • You will go, you will return, never will you die in battle

  • You will go, you will never return, you will die in battle

  • I think the question was more about division of text into different words rather than punctuation of them
    – b a
    Commented Jan 10, 2019 at 10:32
  • @ba True, but it does give context to the question, by showing that the time was not adverse to wordplay in general. This is also a case of division, but grammatical rather than lexical. I think the distinction between lexical and grammatical took time to be clarified ... though I do not know when it actually was better perceived. The initial question may be more complex to answer than I thought when I asked.
    – babou
    Commented Jan 10, 2019 at 13:00

This is purely a fun/informative answer based on the question title, since you've already got an answer (from an older text), and the nature of the wordplay below doesn't involve ambiguous syntactic boundaries.

The Odyssey

Here are a couple in The Odyssey (this is Stephen Mitchell's translation and I'm just reporting his notes).

When Odysseus blinds Polyphemus, he knows that the cyclops' cries for help will be ignored since he's given his name as "No-man" (outis), so that Polyphemus shouts to his friends, "No-man has tried to kill me."

And they went away, and I laughed to myself at how
my cunning little cognomen had fooled them all.

Mitchell: The Greek for "cunning/shrewdness" is mētis, a homonym of mē tis, which is a variant of outis. ("No-man/cognomen" replicates the pun.)

Later, Odysseus is recounting the story of his birth, and reports this speech of his maternal grandmother's:

Autólycus said to them, "Daughter and son-in-law,
here is the name you should give him. Since I have been
odious to so many women and men
throughout the wide world, let him be named Odysseus."

Mitchell: The Greek for "odious" here is odussamenos (similar sound to Oduseus), from a verb meaning "to be angry with"; it can be parsed here as a passive participle: "after having been the object of many people's anger" (citing R.B. Rutherford's 1992 commentary of books XIX and XX).

The Hebrew Bible

While we're at it, this kind of after-the-fact folk etymology of names is also common in the Hebrew Bible, leading to wordplay in a similarly broad sense. Especially with the often hard-to-trace names in Genesis 1-11, there are various attempts by the author or editor to explain a name, such as "Noah" explained by Lamekh's use of נחם n-h-m "to comfort" (despite clearly lacking the third consonant).

The Bible also plays with words for the sake of symbolic imagery. In 1 Kings 18:35-37, for example, Elijah has just asked that the altar he wants God to light on fire be soaked with pails of water. The water "runs around" סבב the base of the altar. He then prays that God will "turn around" the hearts of Israel, using סבב again.

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.