There are cases where a word in another language means something else in another one. I do not mean cognates or loan words having close but not identical meanings, but two words in different languages with no etymological connection ending up looking or sounding the same. This can lead to misunderstandings, humor, or awkward situations.

For one example, Finns are instructed to avoid saying "katso!" ("look!") when in Italy. (I prefer not to give the corresponding Italian word explicitly in an attempt to keep my language clean.) For another example, here is a Finnish comic about "chat customer service", playing with the meaning of "chat" in French:


Are there similar recorded cases between Latin and other ancient languages? Surely there must be some similarities with unexpected words between Latin and, say, Greek.

(I am not happy with my wording of the title. Suggestions for something more concise are welcome. This is also a hard one for tagging.)

  • 1
    The general term for these, to my understanding, is "false cognates".
    – Draconis
    Commented Jun 9, 2018 at 17:09
  • 1
    @Draconis Hmm... I've never heard of that term. Does it cover cases where the two words are from distant languages and mean completely different things? Like when a Latin word is obscene in Finnish? I hope anyone wouldn't treat them as cognates to any extent, but just similar in written or spoken appearance.
    – Joonas Ilmavirta
    Commented Jun 9, 2018 at 17:15
  • 2
    They're also sometimes called "false friends" (calquing some foreign term), and are known for being annoying when learning languages: for instance, Latin mortalis does not mean "lethal" (as in English "a mortal wound") but rather "able to die". I would imagine it includes words that are originally completely unrelated, though the famous examples I know tend to actually have some etymological connection.
    – Draconis
    Commented Jun 9, 2018 at 17:22
  • 1
    I think the term "false cognates" applies to cases like Eng. "have" vs Sp. haver, or Eng. "day" vs Sp. dia, which at first sight appear to be cognates, but really aren't
    – varro
    Commented Jun 9, 2018 at 22:05
  • There must be such words! - If a dedicated article is not yet written on the matter, then only a very erudite person could come up with a significant collection!.... - Inspired by your Italo-Finish accident that must be avoided, I can only indicate a neo-Latin one (which is bound to happen): the Romanian word with the meaning that you discretely omitted is held proudly by a Sardinian town, and also by an Istrian one.
    – cipricus
    Commented May 30, 2023 at 17:02

6 Answers 6


One example that I think fits is L. Roma vs Gk. ῥώμη ("might"). (Presumably the latter in the Doric dialect would be ῥώμα, an even closer match.)


The narrowest and most interesting way to interpret your question is that you want anecdotes exemplifying language-based misunderstandings between ancient individuals; like, if Caesar said "Ave" and Cleopatra thought he was talking about birds. Sadly I'm not aware of any such anecdotes myself.

Interpreting the question broadly, there are literally hundreds of examples of scholars holding incorrect beliefs based on coincidental similarities of pronunciation between words in different languages. I'm thinking of false etymologies. The Wikipedia page on folk etymology lists many examples from all kinds of languages, but most notably

the word baceler or bacheler ... referred to a junior knight. It is attested from the eleventh century, though its ultimate origin is uncertain. By the late Middle Ages its meaning was extended to the holder of a university degree inferior to master or doctor. This was later re-spelled baccalaureus, probably reflecting a false derivation from bacca laurea "laurel berry" ...

Also, an intra-Latin example:

sincere is derived from the Latin sincerus ... from sin- (one) and crescere (to grow) ... An often repeated folk etymology proposes that sincere is derived from the Latin sine (without), cera (wax). According to one popular explanation, dishonest sculptors in Rome or Greece would cover flaws in their work with wax to deceive the viewer; therefore, a sculpture "without wax" would mean honesty in its perfection. ... Another explanation is that this etymology "is derived from a Greeks-bearing-gifts story of deceit and betrayal. For the feat of victory, the Romans demanded the handing over of obligatory tributes. Following bad advice, the Greeks resorted to some faux-marble statues made of wax, which they offered as tribute. These promptly melted in the warm Greek sun." The Oxford English Dictionary states, however, that "there is no probability in the old explanation from sine cera 'without wax'".

Finally, for a perennial case of cultural confusion between a Scandinavian-derived word and a similar-sounding Latin-derived word, look no further than niggardly.

  • 1
    Welcome to the site! These are interesting examples, but they answer a different question. I was wondering about similarities between two languages, not within a single one. We could have a separate question about that.
    – Joonas Ilmavirta
    Commented Jun 9, 2018 at 20:16

Greek κοίτη “bed, act of lying down, sexual intercourse” < κεῖμαι “to lie down”

Latin coitus “sexual intercourse” < co + ire “to go with”.


There might be a confusion, attested in Procopius (or maybe made just by him) between Greek píthēkos (monkey) and Latin (Vulgar, reconstructed) forms meaning "small", "little", or derivations of these. Cf. Vulgar Latin, 2000, Herman, Jâozsef, page 101: the word for “small,” paruus, was replaced everywhere by words that seem really to come from the nursery, such as pitinnus, pisinnus, putillus(teeny)".

Even Wiktionary reflects this: the page for πίθηκος (monkey), beside the figurative meaning "jackanapes/trickster", provides as a third entry the meaning "dwarf". But that seems attested only once, and that occurence may be based on the confusion with the Vulgar Latin word.

The Greek word seems in fact close to some later local variants: pittus > pittitus, pittinus, pitticus and piccus > piccinus, piccoccus, picculus.

These, in turn, have resulted in Romance words like Italian piccino, piccolo, Sardinian piccinu, picciocu, Spanish pequeño, Portuguese pequeno, Albanian picë, "little girl", Romanian pici (small boy), pic (a bit, a drop), French petit (and related), Milanese pitinu, Sardinian pithinnu, piticu, Romanian pitic ("dwarf") and piti, pitit ("to hide", "hidden", like in "make oneself small"). More scholarly references in links below.

As I mentioned in a comment, I don't have knowledge of a direct confusion as you asked for, but reading again those posts initially linked in the comment I have found something that might qualify.

My initial problem was to find the origin of the common Romanian word pitic (dwarf). The etymology in Romanian dictionaries was linking to the Greek πίθηκος - píthēkos (monkey) while many Romance languages and dialects had similar forms that pointed to a Latin origin. There was no "pitikkos" form in Latin, more or less related to the Greek word, and in the end I was convinced of the otherwise obvious Latin origin of the Romanian word, as argued here.

But I was kept for a while hesitating by the fact that Suda lexicon is defining píthēkos also as "what some call a short little man" and linked that meaning to Πίθηξ, "Used in the sense of "dwarf" by Procopius, Gothic War 4.24".

Looking at the sources provided under the Wiktionary page for πίθηκος the meaning is not attested in other sources, and its rarity pleads in favor of a confusion (or maybe a contamination) attested by the 6th century historian Procopius of Caesarea between the Greek word and a late vulgar Latin and early-Romance term meaning "small".

(Procopius' confusion ended up in Suda, was then quoted by various dictionaries, including Wiktionary, and thus became in Romanian dictionaries the basis for a wrong etymology of the word meaning "dwarf".)

Now, trying to identify that word in the text by searching in Procopious works online, I couldn't identify it with the meaning "dwarf". I couldn't find Πίθηξ in the Gothic Wars, nor the complete works. The only word containing the form πίθηκ is πίθηκες with the meaning "dwarf", and was found in Gothic Wars not at 4.24, but at he very end of 8.24:

ἐνταῦθα, ὥσπερ ἄνθρωποι πίθηκες γίνονται, οὕτω δή τινων ἵππων ἀγέλαι εἰσὶ τῶν προβατίων ὀλίγῳ μειζόνων.

On that island are found apes just like men, and there is also a breed of horses only a little larger than sheep.

(The exact same fragment seems present in Book 8 of The Wars of Justinian)

If the Suda entry is not verified by Procopius, that increases in fact the chances that πίθηκος with the meaning "dwarf" in Greek is due to a contamination from a proto-Romance word.


One that works on two levels is the Phoenician ḥawe, "to live" and the Latin avere, "to be well." An early borrowing from the Phoenician as have or ave got then re-analyzed as the imperative of avere, but it seems that's just a coincidence.

The difficulty has long been recognized, though, leading to e.g. Lewis and Short giving the two verbs separate entries without recognizing the underlying Phoenician.

See fdb's post on it and the follow-up discussion in the comments.


Some false cognates between Latin and Greek are "sapientia"-"sophia" (both meaning "wisdom") and "deus"-"theos" (both meaning "god").

  • 1
    What is false about such cognates?
    – Figulus
    Commented Jun 2, 2023 at 19:02
  • 1
    @Figulus Well, they are false cognates in the same way Latin "dies" and English "day" are: they are coincidentally similar, they don't come from the same root. Commented Jun 2, 2023 at 19:03
  • 1
    I see. But I am not so sure thay sound alike enough for one to be confused with the other. But maybe I'm not understanding the OP.
    – Figulus
    Commented Jun 2, 2023 at 19:08
  • 1
    @Figulus Draconis' comment under the OP explains a bit more. The first might be a bit of a stretch, but the second seems to fit well (unless I'm also misunderstanding something).
    – cmw
    Commented Jun 2, 2023 at 19:45

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.