I'm reading Collar and Daniell's First Year in Latin right now and they mention that Latin has no articles such as "a", "an", and "the". Is this true? I have heard the book be inaccurate before.

  • 8
    Note that Latin is hardly special in this regard. Most Slavic languages (like Polish or Russian) have no articles either, for example. – Sebastian Koppehel Mar 30 '20 at 6:28
  • 1
    A.A. Milne's book "Winnie the Pooh" has been translated into Latin under the title "Winnie Ille Pu". – A.k.a. Demic Mar 30 '20 at 11:26
  • 1
    ... and other languages only have a definite article ("the"), and no indefinite article ("a/an"), e.g., Hebrew. – Stephan Kolassa Mar 30 '20 at 12:00
  • 1
    @StephanKolassa: And, conversely, some have indefinite articles but no definite articles. The World Atlas of Linguistic Structures, Chapter 37 includes 45 such languages, including Japanese, Persian, and Cantonese. (Chapter 38 includes 98 languages with definite articles but no indefinite articles, including Icelandic, Welsh, and various forms of Arabic.) – ruakh Mar 31 '20 at 0:40
  • @A.k.a.Demic In that context ille has a meaning something like 'the well-known', rather than the plain definite article. – dbmag9 Apr 1 '20 at 14:57

The book is correct. There is no equivalent to "the" in Classical Latin.

In Vulgar Latin, the demonstrative ille (which means "that" in Classical Latin) got bleached into a definite article, with a meaning similar to English "the". That's where forms like Spanish el, Italian il, French le, and so on come from. But that wasn't good Classical style.

  • 2
    Another interesting case is Romanian, where various forms of ille/illa morphed into enclitic definite articles, e.g. om (man) → omul (the man), muiere (woman) → muierea (the woman) etc. – Sebastian Koppehel Mar 31 '20 at 20:33

As discussed here:

When did the word "ly" enter the Latin language and where did it come from?

the word "ly" is occasionally used as a definite article in mediaeval Church Latin.


There is no article in Latin. You just don't translate it. "The girl is the servant of the lady" = "puella est ancilla matronae".

However, there are some cases where an article may be translated with various words: this happens when you're not really using "the" or "a" as an article, but to mean "the famous", "that famous", "a certain". For instance: "the teacher is wise" = "magister savius est", but "the famous teacher is wise" = "ille magister savius est". "A certain Flavius" may be rendered as "Flavius aliquis", literally "someone Flavius". This is especially true with proper names: "ille Cato, censor perclarus". This in particular explains the "Winnie ille Pu" example.


There is one Romance language, namely Sardinian, in which it was not the demonstrative "ille" to get bleached into the definite article but where the definite article derives from the Latin word ipse (the same), and so it became su (masc.), sa (fem.)

  • 1
    This doesn't really answer the question, but it's very nice to know. I knew about the "sa" article in Sardinian, but I had no idea it came from "ipsa". – Simone Mar 31 '20 at 12:27
  • 1
    Eh, it answers it for a dialect of vulgar Latin which developed into Sardinian. It's as valid in its way as the classical bias pretending lower-class Latin didn't use ille and intensifiers for clarity. – lly Mar 31 '20 at 18:48
  • Not only Sardinian. Some dialects of Catalan did, too. Nowadays the articles es/sa, from ipsus, are restricted to the Balearic Islands and a few locations in North Eastern Catalonia, but thew used to be common in a larger area. – Pere Mar 31 '20 at 19:49

The other answers already covered the most important part of your question but since you did mention

..."a", "an"...

you do know those are just archaic forms of "one" and the way to express that is unus, una, unum, right? Similar to Chinese, though, Classical Latin can express this idea but usually just doesn't bother to. Use unus &c. in places where you might say "one" or really want to emphasize the "a" (eg, when you'd pronounce it /aɪ/ instead of /ə/). More often, rather than repeat unum every other sentence, Classical Latin depended on context or synonyms like ullus/a/um ("any") and certus/a/um ("a certain...").

In lower class/vulgar/medieval Latin, of course, unus/a/um became more and more common, which is where you get French un(e), Spanish uno/a, &c.


You can use "is, ea, id." It's used sort of like an adjective; its a softer demonstrative and is often used to mean "he, her, it," on its own. But you could say "is frater" to mean "THE brother," although the common translation is "this/that brother."


The theoretical proto-Indo-European language that is at the root of all Western (except Basque), all Near Eastern and some Middle (eg Farsi, Sumerian) and South Asian (eg Hindi) languages had no articles so they tended to adopt them from the native cultures the speakers conquered. Latin, as stated above, had to form its own, possibly from the influence of their neighbours whom, as non-native speakers, would try to look for an equivalent.

  • 7
    This is rather inaccurate. PIE is not the ancestor of Sumerian, nor of Near Eastern languages except Anatolian. And though many IE languages developed articles, they didn't do so by adopting them from other languages, and Latin never did so at all. – TKR Mar 30 '20 at 22:12
  • 1
    @TKR: Aside from the inaccuracies with respect to Proto-Indo-European, perhaps what Lucky Day means is that Late Latin and/or its daughters adopted ille as a (proto-)article under influence from Germanic. But Germanic is also Indo-European. So I wonder which "neighbours" he has in mind. Greek would seem less likely. – Cerberus Mar 31 '20 at 1:54
  • Greek is also Indo-European. There are several mistakes Lucky should just clean up, if they'd like this to be a good answer. They'd need to start with a scholarly cite for the idea that definite articles were adopted from outside the language, which doesn't seem to have been the case. It generally developed from demonstrative adjs like this or that within the languages. – lly Mar 31 '20 at 18:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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