Note 4: One of the main differences between Japanese and English grammars is the tendency of English to repeat information already disclosed in the context of the conversation vs. the tendency of Japanese of keeping it in mind and not repeating it again, i.e., English tends to be explicit while Japanese tends to be implicit. We go over examples in English, to get an idea of how and why Japanese uses implicit information, and to convince ourselves that it is not that different from what we sometimes do in English.

parts of a sentence

Let’s review some basic concepts so we can have a common ground for the examples; we’ll go over more formal definitions later, if we need them:

• word: a meaningful sequence of letters:
  ‘my’, ‘cousin’, ‘Maria’, ‘ate’
• phrase: a set of words that forms a meaningful unit:
  ‘[my cousin Maria]’
  ‘[my cake]’.
• clause: a meaningful unit that has a subject and a predicate:
  ‘my cousin Maria ate my cake’
  ‘Carlos says that my cousin Maria ate my cake’
  ‘she denies it’
• sentence: a full idea written as one or more clauses:
  ‘Carlos says that my cousin Maria ate my cake, but she denies it’

oops… we also need to define the subject and the predicate for the definition of ‘clause’ to make sense:

• subject: the person or thing that is or does something
• predicate: the part of the clause that is not the subject

We will write explicit phrases within brackets, [ ], if we want to emphasize them as units. For example, let’s consider the sentence

‘Carlos says that my cousin Maria ate my cake, but she denies it.’

If we want to emphazise its phrases it would be:

‘Carlos says that [my cousin Maria] ate [my cake], but she denies it’

Let’ go over the clauses of this sentence. One of its clauses is:

The clause above is nested within a larger clause:

and finally we have a third clause:

In the same way that we nested clauses above, we can nest phrases, e.g.,

‘[the dog of [my best friend David]]’

who ate my cake?

The mystery of the missing cake above started with the following conversation:

David: Who ate my cake?
Carlos: Maria.

We can identify that parts of David’s question:

That was easy. Now let’s see Carlos’ answer:

Wait… wait… a clause needs a subject and a predicate and ‘Maria’ is just a noun. How can it be a clause? On the other hand, in the context of the conversation, ‘Maria’ is a unit with full meaning, i.e., it means that Carlos thinks that ‘Maria’ is the one that ate David’s cake. As an isolated utterance, out of context, ‘Maria’ makes no sense; however, as a short way of saying ‘Maria did’ or ‘It was Maria’, it makes perfect sense as an answer to David’s question. The analysis with either ‘Maria did’ or ‘It was Maria’ is similar, so let’s consider one of the cases, e.g., that when Carlos said ‘Maria’ he meant ‘Maria did’.

Out of context, ‘Maria did’ doesn’t make sense because it doesn’t tell us what Maria did. Well… in the context of David’s question, ‘Maria did’ just means ‘Maria ate it’ but, again, ‘Maria ate it’ doesn’t stand on its own either because we wouldn’t know what is the ‘it’ that she ate. Thus, let’s specify it: ‘Maria ate your cake’. Argh… again, it doesn’t make sense out of context either because we don’t know who is the person that we mean by ‘your’. How about we replace ‘your’ to end up with ‘Maria ate David’s cake’. Ah! finally.

Our clause ‘Maria ate David’s cake’ out of context gives us the same information that ‘Maria’ gives us within the context of the conversation. Hence, is there really a point of having a sentence like ‘Maria ate David’s cake’ that spells out all the information if we already have that information from context? If we know the context of the conversation, do any of the following answers is more informative that any of the others about who ate David’s cake?

who ate my cake?

• Maria.
• Maria did.
• Maria ate it.
• Maria ate your cake.
• Maria ate David’s cake.

In the end, the succint ‘Maria’ gives exactly the same information as the wordy and redundant ‘Maria ate David’s cake.’. The difference is that ‘Maria’ relied on context, while ‘Maria ate David’s cake.’ didn’t but, in essence, all the answers are equivalent.

enter zero

If ‘Maria’ is equivalent to all the other answers then, from context, we should be able to infer the implicit information; after all, we are claiming the information is there but is simply invisible. For example, suppose that we want to infer the verb and predicate that is implicit in ‘Maria’ if ‘Maria’ was the subject of the clause; then we can write:

David: Who ate my cake?
Carlos: Maria ∅ ∅.

where the first ∅ stands for the implicit verb ‘ate’ and the second ∅ stands for ‘it’.

Then our sentence would be:

The symbol ∅, called zero, stands for a word or phrase that is not written explicitly in the sentence, but that we can infer if we so desire because the context gives us enough information to do so [wp]; we will place 0 words and phrases within curly brackets when we want to make clear that they are not actually written in the text.

In the example above, the first ∅ is a ‘∅ verb’ because is playing the role of the verb ‘ate’, while the second ∅ is a ‘∅ pronoun’ because is playing the role of the pronoun ‘it’.

If we had decided that ‘Maria’ was a short for ‘It was Maria’, with ‘Maria’ as part of the predicate of the clause instead of its subject,then we would have ended up with the dialog:

David: Who ate my cake?
Carlos: ∅ ∅ Maria.

and the sentence would have been:

Now, the first ∅ is a ‘∅ pronoun’ because is playing the role of the pronoun ‘it’; the first ∅ also happens to be a ‘∅ subject’ because, in this case, the implicit ‘it’ is the subject of the sentence; the second ∅ is a ‘∅ verb’ because it is playing the role of the verb ‘was’.

So which version is correct? Is ‘Maria’ short for ‘Maria {ate it}’, or for ‘{It was} Maria’? Within the given contex, both sentences give exactly the same information and, therefore, either version is a perfectly legitimate answer.

Insufficient information

If the two sentences ‘Maria {ate it}’ and ‘{It was} Maria’ had led David to different conclusions about who Carlos claims to have eaten the cake, then ‘Maria’ would not have been a valid answer that Carlos could have given to David. If the answer could have been misinterpreted, then Carlos would have needed to give more information with his answer. Let’s see an example:

David: Who ate my cake and drank my soda?
Carlos: Maria.

Given this answer, Carlos is saying that Maria did both. However, suppose that Carlos knew that Maria ate the cake but has no information about the soda. In that case, ‘Maria’ is not a correct answer becasue leads to the wrong conclusion; instead, Carlos would have had to answer something like:

David: Who ate my cake and drank my soda?
Carlos: Maria ate it.

which would be short for ‘Maria ate {David’s cake}’ or ‘{It was} Maria {who} ate it’.

If instead Carlos knew that Maria drank the soda but had no idea about the cake, then his answer would have had to be:

David: Who ate my cake and drank my soda?
Carlos: Maria drank it.

which would be short for ‘Maria drank {David’s soda}’ or ‘{It was} Maria {who} drank it’.

Thus, as long as the answer is correct within the context of the question, any possible interpretation will lead to the correct conclusion.

Is it possible to give the wrong impression with an answer that has too much implicit information? Of course! It happens all the time. When we realize that the implicit information in our clauses are leading to a wrong interpretation, we try to make more information explicit with followup sentences, or with corrections such as ‘What I meant to say is…’, or ‘Sorry, I didn’t mean that’, or similar.

Not true!

The reply of Maria to the accusation is:

David: Who ate my cake?
Carlos: Maria.
Maria: Not true!

Like the answer ‘Maria’, out of context, ‘Not true’, doesn’t make sense. However, given the context, ‘Not True!’ is a short form of ‘{that} {is} not true!’ so, with the help of ∅, we can analize it easily:

David: Who ate my cake?
Carlos: Maria.
Maria: ∅ ∅ Not true!

where the first ∅ is a ‘∅ subject’ and a ‘∅ pronoun’ because ∅ is playing the role of the pronoun ‘that’ and the subject of the sentence; the second ∅ is a ‘∅ verb’ because ∅ is playing the role of the verb ‘is’. Now we can make explicit our previously implicit sentence:

intro to zero words

The concept of the ‘∅ word’ has been used in linguistics for a few decades now; we can have ∅ subjects, ∅ verbs, ∅ nouns, ∅ pronouns, ∅ phrases, or anything else. Basically, if a part of the sentence is not written explicitly but we can infer that it is there, implicitly, then we can represent it with ∅ for the purposes of anaylizing its grammatical structure.

The missing-cake conversation shows how ∅ might make its appearance in English. The key to infering the implicit invisible ∅ words is to know the context; without context we cannot infer the ∅ words. We can infer the context in which we would say the clauses below and, therefore, we can infer the ∅ words:

These conversation and interjection examples might give the appearance that in English we use ∅ words much more often that we realize. Although that might be true, we really don’t use ∅ words in English nearly as often as we use them in Japanese. In English we tend to be very explitic about everything we say and write, while in Japanese we tend to be very implicit. For example, while in English we say ‘I am eating’ and ‘Are you eating?’, being explicit, in Japanese we say ‘∅ ∅ eating’ and ‘∅ ∅ eating?’, being implicit.

Japanese relies so heavily on the context of the conversation that is considered a context-based language and one of ways in which this manifests is in the broad use of ∅ words, specially of ∅ pronouns, ∅ subjects, and the ∅ particles that behave like English prepositions.

Summary

Out-of-context translations between English and Japanese are seldom correct because in English we tend to be explicit and avoid ∅ words, while in Japanese we tend to be implicit and embrace ∅ words. This lack-of-context is one of the reasons for the poor performance of automatic translation between English and Japanese. In general, translations of out-of-context implicit Japanese sentences into English become incomplete senseless sentences, while translations of full-of-context explicit English sentences into Japanese become redundant and self-centered sentences.

References

Every once in a while comes a person that is completely comfortable in both Japanese and English, and has the skill to do an insightful comparative analysis. In the past few decades, that person has been Prof. Jay Rubin, with his 1992 book ‘Making Sense of Japanese’; this book is not a text but a discussion of finer points. The book assumes that the reader already has enough knowledge of Japanese to be confused with its grammar, so its target readership is an intermediate-level Japanese student. Our advice is to get the book and re-read it every year or so. Unfortunatelly, the book’s insights get drown in the cacophony of Japanese textbooks, podcasts, videos, apps, and websites (like this one). Trying to not be part of the problem, we’d like to very much encourage you to get the book.

A few years ago a person that went by the pen-name of ‘Cure Dolly’ built upon Prof. Rubin’s insights; tragically, she passed away recently. From what we can gather, she was not a trained linguist; instead, she taught herself Japanese and moved to Japan where she lived her last years. What she might have lacked in formal training, she made up with an unbound love for Japan, a profound admiration for the Japanese language, a need to understand it at a deep level, and a desire to share what she had learned with people that were following in her steps. She hosted her Youtube videos with an avatar that hid her identity; the avatar’s look and voice turned many people away, though, which is unfortunate because her insights were solid. Her book ‘Unlocking Japanese’, published in 2016, her Youtube videos and her website – kawajapa – are aimed to beginner/intermediate students.

Disclaimer: I own both books; I derive no benefit of recommending either book.