A dictionary defines words in a language, in terms of other words in that same language. An English dictionary is not the same as a Spanish dictionary, simply because the sets of English words and Spanish words are not the same. Dictionaries are revised over time, new words are added and old words are discarded, and we may have two dictionaries of the same language which are not identical to each other, the differences roughly proportional to distance in time and space. So a particular dictionary is distinguished enough to identify not only its language but in some sense even itself.

Does a dictionary completely describe its language? Is it possible for a person to learn a language by reading only a dictionary of that language, without any other sources of information? Initially I guess it would be something like solving a cryptogram but can learning progress from that point onward?

  • Question for you. Is music the notes?
    – user4894
    Commented Jun 14, 2016 at 17:32

6 Answers 6


A "complete description" of a target language would in one sense require enumerating all the possible sentences one could form in that language. I presume you don't intend us to be quite this strict since clearly a dictionary doesn't (explicitly) enumerate all possible sentences.

One point to consider here is how vacuous a dictionary would be without any form of external reference. It could only be used to decode another language once you had some other means of translation; it's nothing but a map of words to other words.

On the other hand, consider a similar but distinct hypothetical situation: suppose we only have access to an encyclopedia in the target language. I submit that a reasonably complete encyclopedia would come closer to answering to your terms of a "complete description" of the language, since it embeds a significant amount of context the dictionary lacks; it maps words to things, concepts, places, etc.

Finally, there is certainly at least a sense in which a dictionary formulates a 'description' of a language, inasmuch as a dictionary is by definition a standard for usage and pronunciation. (However, even in this sense standard dictionaries perhaps fall short, as they usually don't include a complete grammar.)


No given dictionary for a language completely describes that language.

One of the elements of a language is its lexicon: the collection of words that are composed to make up strings of that language. Dictionaries are always behind the lexicon as it changes more quickly than lexicographers can keep up with.

Even if we change the question to ask about the lexicon (together with the meanings of the lexical elements), that will not determine a language. There is also the grammar of the language: the rules by which lexical items can be composed to yield strings of that language.

It is at least conceptually possible that two distinct languages with very different grammars could have been developed using the same lexicon and associating the same meanings to each lexical element.

  • The dictionary gives statistical information about both the lexicon and the grammar, can't we learn something from those statistics? Commented Jun 13, 2011 at 3:50
  • @Dan No dictionary is comprehensive for the lexicon of its language, so that settles that languages are not determined by their dictionaries. As for the grammar, any decent dictionary will tell you grammatical categories (noun, verb, etc.) The OED even provides a substantial corpus of examples of the language in use. So, it might be that the OED would provide what you need to discern the grammar of the language. But the first point still seems decisive against the hope that dictionaries determine their languages.
    – vanden
    Commented Jun 13, 2011 at 3:56
  • I think the determination of a language can be ambiguous, and a dictionary has determination as its purpose, but as you say it can't possibly be comprehensive by including every word or every grammatically-correct instance. I'm thinking of "language" as a knowledge. If I only know English, can I know Spanish by reading a Spanish dictionary? Commented Jun 13, 2011 at 4:36
  1. Current languages can certainly be identified from their dictionaries, in the same sense that I can be identified from my social security number. This aspect certainly does not in any sense justify an affirmative answer to the title quesiton Is a language a dictionary?

  2. A dictionary does not completely describe a language because the actual usage is subtle and involved a lot of cultural and social information and intonation and nonverbal cues. I would concede that it is possible in principle that a dictionary could describe the actual usage at a certain point in time and space, but this is already on practically impossible because many aspects of natural languages are just not explicitly understood.

  3. Is it possible to learn a language from a (modern) dictionary? (where we restrict "learning" to the typical level of learning a foreign language, as in: We want to decipher these hieroglyphs). This is an interesting question and I would say that the answer is very likely yes with some basic assumptions about human cultures and the interconnectedness of the given dictionary.


Surely a dictionary does not define an entire language. Apart from the fact that new words (beginning as slang, sometimes making their way into formal language) are being invented all the time, so are new meanings for existing words in new contexts.

In addition, it could be argued that combinations of the words together (ie. literature) form part of the language itself. The dictionary on its own describes the words in Shakespeare, Chaucer, Betjeman, Tolkien, etc. but it does not contain the prose. Without the literary works in a language, it can be reasonably argued that you haven't captured the whole language.


Suppose that we are not talking about any current dictionaries but one a bit more complete, with more examples of usage and meaning, something more precise about the context each word can be used in, something with phrases and idioms with specific meanings. This isn't terribly idealistic, just probably currently out of reach because of resources and desire.

First presume that we are given such a dictionary as a foreigner (the dictionary is written in our foreign language to help us know how to understand English). Then with the assumptions above I think it can be said that this augmented dictionary is the English language for the foreigner (yes, there are difficulties with translation, but not impossibilities and I'll presume that the assumptions above take care of these difficulties).

But instead, let's presume that it is self-defining, that the dictionary is in English, it is a definition of English in English. I don't see how this could be a definition of the language, because an understanding of English must be assumed to be able to interpret the definitions. There might be room for the 'cryptogram' idea, in the sense that the foreigner might be able to extract meaning out of the English definitions recursively and somehow extract meaning out of the circularity.

Those are just supposing we can solve the limitations of a dictionary.

But then also, even if individual words have every nuance explained, those words (or even phrases or idioms) are not the entirety of language. There are all the extra-lexical parts of a language: syntax and semantics.

And then there are all the extra-linguistic things, those not part of the spoken utterances but that surround them, the culture, the day-to-day interactions, and the economic and material climate, that restrict what is appropriate to say and what not. There's the anecdote about the professor who visits a country to give a highly technical lecture in full fluency, but can't even open his mouth to order in a cafe or talk to a child.

A dictionary is certainly a necessary component of a process to 'reconstruct' a language. And one could communicate passably in a language just spouting a set of words picked from a dictionary. But there is a lot more needed to define a language.


You don't write a dictionary first and then tell people to learn it. You begin to communicate and then refine that communication (language). A dictionary is used to define and refine the language. Therefore, the dictionary is continuously catching up with the spoken language. Simply compare the 1812 Websters with the 2011 Websters, the definitions are vastly different.

To answer your question, a language is not its dictionary.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .