Did they have dictionaries in the ancient times?

I mean who used the dictionaries? Did authors use them to know how to write?
I don't think it worked this way. But when in the history dictionaries began to be really in use?

For instance, they give approximative dates in some sites, but no mean to know about the diffusion:

Was it a private project from the author? Did every civilizations have their dictionaries? Could it be that some texts have been lost?


The oldest known dictionaries were Akkadian Empire cuneiform tablets with bilingual Sumerian–Akkadian wordlists (...)
(early 2nd millennium BC)


For Latin:

The oldest Latin Dictionary was compiled by Solomon, bishop of Constance, about the year 1410.


1st cent. BC

Beginnings of ancient Latin lexicography, in works such as the lost Liber glossematorum of Lucius Ateius Philologus.

For Greek:

circa 300 BC:

Philitas of Cos and Simias (or Simmias) of Rhodes make the first extensive learned collections of glosses of ancient Greek epic and dialect words, initiating the Greek lexicographical tradition. Their work only survives in fragments.


  • Of course some texts are lost. Probably most ancient texts have been lost. This might be a better question for the History site.
    – curiousdannii
    Nov 20, 2019 at 3:44
  • Yes, but it's about lexicography. I mean we have very few dictionaries attestations because the texts were lost, and they had (for instance) a lot of them.
    – Quidam
    Nov 20, 2019 at 3:52
  • 1
    It really depends what you consider a "dictionary". Do you have a specific definition you want to use?
    – Draconis
    Nov 20, 2019 at 3:52
  • It's probably rather obvious, because I didn't find a question about "What is a dictionary" or "What is the definition for a dictionary" on this site. Do you mean whether it is a dictionary or words or expressions?
    – Quidam
    Nov 20, 2019 at 4:01
  • 2
    @Quidam "What does the word 'dictionary' mean" is most definitely off-topic on this site. But it's hard to say "when in the history dictionaries began to be really in use" without more clarification on what you would consider to be a dictionary. Does any list of words count? What about a bilingual list (words in language X with their translations in language Y)? A list of words in language X with synonyms in language X? A list of logograms mapped to their pronunciations? Etc etc. Projects along the lines of the OED didn't exist until very recently; there was nothing exactly like that before.
    – Draconis
    Nov 20, 2019 at 4:19

1 Answer 1


As OP clarified in the comments:

Bilingual list is a dictionary.

Therefore, the oldest dictionaries would be the cuneiform lexical lists. These are attested from close to 4000 BCE, and are extremely well-attested because they were used for practice at the edubba (scribal schools).

Much like modern Japanese, Akkadian cuneiform used a large inventory of logograms borrowed from Sumerian, sometimes augmented with phonetically-spelled endings. A typical lexical list would include a long series of logograms, their Sumerian pronunciations in phonetic spelling, and their Akkadian pronunciations in phonetic spelling. For example, one row might contain the logogram for "god" (𒀭), the Sumerian pronunciation diŋir, and the Akkadian pronunciation ilum. (To continue the Japanese analogy, it would be like listing out the logogram 神, the Mandarin pronunciation shén, and the Japanese pronunciation kami.)

These lists weren't widely used by the public, because in the third millennium BCE, most people weren't literate. But they were widely used in scribal education, with scribes-in-training copying them out over and over to learn by rote. Presumably if a student came across an unfamiliar logogram in an assignment, they could also find one of these lists to look it up.

(The use of such things, of course, varied widely across eras and civilizations since 4000BCE Mesopotamia. But this is almost certainly the oldest attestation.)

  • Cool, I didn't exactly realize there was this level of parallelism with the way Japanese works (down to equivalents of onyomi and kunyomi), although I knew some logograms were used.
    – LjL
    Nov 20, 2019 at 18:11
  • 2
    @LjL Indeed! It's a fascinating parallel, which even goes one step further in Hittite, where you can find the signs a-na-an-ši are meant to be pronounced siuni: ana is an Akkadian preposition "to", an is also the logogram for "god", ši can also be read as lim, so the Akkadian reading would be ana elim "to the god", which in Hittite is siun-i (god-DAT)! Thankfully, as far as I know, no other language has borrowed the Japanese writing system, so we don't see this third level of indirection (yet).
    – Draconis
    Nov 20, 2019 at 18:57
  • 1
    far from me to impart value judgments on languages or scripts, but I think I have to concur with the "thankfully" ;-)
    – LjL
    Nov 20, 2019 at 20:41
  • 1
    (Typo in my previous comment, and unfortunately I missed the editing window: the Akkadian should be ana ilim. Oops.)
    – Draconis
    Nov 20, 2019 at 20:57
  • oh well, at least elim made it more obvious to me how it was a word for god.
    – LjL
    Nov 20, 2019 at 20:59

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.