6

Tim Ferris' blogpost How to Learn (But Not Master) Any Language in 1 Hour (Plus: A Favor) describes a "language deconstruction technique" to help people find out what a foreign is like before deciding whether they want to learn it. The technique relies on a small set of simple sentences that expose specific language features in the target language, for example:

  • does the language have a case system,
  • does it have verb conjugations,
  • how does it express certain basic things such as past, present and future,
  • how does it use pronouns,
  • what type of syntax does it mainly use (e.g. SVO order versus SOV order).

Below are the examples from the article:

  1. The apple is red.
  2. It is John’s apple.
  3. I give John the apple.
  4. We give him the apple.
  5. He gives it to John.
  6. She gives it to him.
  7. Is the apple red?
  8. The apples are red.
  9. I must give it to him.
  10. I want to give it to her.
  11. I’m going to know tomorrow.
  12. I can’t eat the apple.
  13. I have eaten the apple.

The idea is that you find a native speaker of the target language and discuss how these sentences would be translated into the target language. This should give you some clues about the language features listed above.

Some people have applied this to specific languages, e.g.

Ferris' list of example sentences is very short, so there may be important language features that don't get uncovered by means of this list. What sentences should be added to the list to cover language features that are typically learnt at CEFR levels A1 and A2?

In other words, any additional sentences should help uncover language features in foreign languages that the above sentences do not uncover yet (you should mention which feature and at least one language where it is used) and should be basic enough to be learnt at levels A1 and A2 of the CEFR (to remove the element of subjectivity). Proposals to replace sentences are OK. In total, there should not be more than 30 sentences.

The goal is not to cover all language features that are learnt at levels A1 and A2; I mention these levels only to exclude grammatical subtleties that you only need at higher levels. In other words, A2 represents the upper bound of what needs to be covered. The goal of language deconstruction is to get a rough idea of what the target language is like before you start investing time and energy into learning it.

  • I think the way you formulate it now, your question is a bit too open-ended. Maybe you could restrict it somehow? What is, roughly, the number of sentences that you want to have? A few dozen? A few hundred? A few thousand? Do you want to cover all languages in existence, or maybe only some most popular languages? – michau Oct 14 '16 at 14:50
  • In any case, if you have a good specification what you want, this might be a good candidate for a community wiki. – michau Oct 14 '16 at 14:52
  • @michau I have tweaked the question again, especially to cap the number of sentences. The number of target languages is not a criterion. We're only doing amateur contrastive analysis here ;-) As I mentioned before, the sentences don't need to uncover all language features in all foreign languages, but just enough features to help you decide whether you would like learning a specific foreign language. – Christophe Strobbe Oct 14 '16 at 15:12
  • Now it's much clearer. But I'm still not sure what the target language group is. It makes a big difference if you just want to cover, say, 30 most popular foreign languages, or all 6000 or so languages that currently exist. The former is much harder, obviously. – michau Oct 14 '16 at 15:31
  • @michau Can you please explain why it would be necessary to define what target languages I want to cover? It's just a bunch of sentences that should be translatable into any language. I see no reason to limit the number of target languages. Illustrations of target languages (in the responses) can pick any target language. – Christophe Strobbe Oct 14 '16 at 15:35
2

I don't think it's a workable approach.

A simple example from Spanish:

The apple is red.
La manzana es roja.

How about tomato?

The tomato is red.
El tomate es rojo.

Tomate happens to be masculine, and therefore requires the article el instead of la, and the adjective that describes it needs to end with -o, not -a. At the A1/A2 Spanish level you certainly need to know that.

Another example, from Polish:

The apple is red.
Jabłko jest czerwone.
I have eaten the apple.
Zjadłem jabłko.

Let's try with tomato again:

The tomato is red.
Pomidor jest czerwony.
I have eaten the tomato.
Zjadłem pomidora.

Jabłko is neuter and pomidor is masculine. If you just try sentences with the former, you may think that the accusative case is equal to the nominative case in Polish. Only if you choose something that happens to be masculine or feminine in Polish, you will see that it's not the case ("tomato" in nominative is pomidor, while in accusative it's pomidora).

Finally, consider simple Mandarin phrases, that even beginners need to know:

one day
一天 "one day"
one week
一個星期 "one 個 week"
one book
一本書 "one 本 book"

When you construct the phrase for "one day" you simply juxtapose the characters for "one" and for "day" (which happens to be a measure word, but let's leave that aside). When you combine 一 "one" with 星期 "week", you need to use the measure word 個, which cannot be translated into English in any way. Finally, when you combine 一 "one" with 書 "book", you need to use another untranslatable measure word, 本.

How to construct appropriate sentences in English? You can't. Of course, you can have English sentences with nouns of all possible genders in Spanish. And Polish. And Lithuanian. And with nouns that have all basic measure words of Mandarin. And Japanese. And Burmese. But you can never be sure if you cover all important distinctions in a new language you're trying to learn about, unless you cover whole vocabulary for the A1/A2 level, which probably means that you need more than 1000 sentences. And even then, I don't think it's possible to construct basic vocabulary in English that can simply be translated to other languages. In different languages, different things and actions are culturally important, and therefore the basic vocabulary should differ too.

I see the point of having some phrases translated to have a taste of the language, but I think that trying to cover the grammar of all languages at CEFR A1 and A2 is simply setting the bar too high.

  • Thanks for the answer. However, it's not English that needs to have all the grammatical features. That is obviously impossible The goal is to make the translations highlight certain grammatical features of the target language. For example, translating "I give John the apple." into German will show that German has a case system, while English doesn't. – Christophe Strobbe Oct 14 '16 at 14:11
  • @ChristopheStrobbe You're right, I edited it out, but that doesn't really change my main point: a small set of sentences will have arbitrary chosen vocabulary and you will never know if you cover all genders, noun classes and important measure words (among other things) in the target language. – michau Oct 14 '16 at 14:22
  • That's fine. You don't need to capture all grammar that needs to be covered in levels A1 and A2, but just get an idea of what a language is like without getting into subtleties that you learn only at a higher level. – Christophe Strobbe Oct 14 '16 at 14:29
  • "Of course, you can have English sentences with nouns of all possible genders in Spanish." That is not necessary. The goal of the sentence is to uncover whether the target language uses grammatical gender. If it does, further discussion with the native speaker can reveal how many genders the language has. – Christophe Strobbe Oct 14 '16 at 15:17
1

Well, for one adverbs. Adverbs are basically adjectives for verbs or other adjectives. An example sentence could be:

The apple is very shiny.
The man walks very fast.

"very" is used to help the reader know how shiny or in the second case, how fast. Yes, "very" is subject to opinion but there should be some similarity in the mental image provided by the sentence.

Past, Present, Future. All speakers should know these three tenses regardless of skill:

Past: I finished the dish.
Present: I am finishing the dish.
Future: I will finish the dish.

The usage of suffixes or extra verbs are also present (-ing for an action that is continuing i.e. progressive, -ed for past, etc.)

And clearly, all of the above are used in English.

  • Thanks for the answer. In your first example, only "very" is an adverb; it modifies "shiny", not the verb. "Shiny" (here) is an predicative adjective. In "He walks very fast", "fast" would be an adverb. – Christophe Strobbe Oct 14 '16 at 10:58
1

OK, so here's my another take at it. Instead of providing specific sentences, I'd suggest a way of discovering what the most common grammatical distinctions there are.

Since you seem to be equally interested in all languages, I would look at the language feature list on WALS, sort them by number of languages, and make sure that the sentences cover the features from the top of the list. Let's take a look at it:

  1. Order of Object and Verb
  2. Order of Subject and Verb
  3. Order of Subject, Object and Verb
  4. Order of Adjective and Noun
  5. Postverbal Negative Morphemes
  6. Preverbal Negative Morphemes
  7. Order of Negative Morpheme and Verb
  8. Minor morphological means of signaling negation
  9. Relationship between the Order of Object and Verb and the Order of Adjective and Noun
  10. Order of Genitive and Noun
  11. Order of Demonstrative and Noun
  12. ...

1-3 is already covered by your sentences with the indirect and direct object. For 4. we need a phrase such as "the red apple". For 5-8 we need negation, so we can say e.g. "I don't have the red apple", which also covers 9.

Now for 10 we need a genitive and noun, e.g. "John's apple". For 11: a demonstrative and a noun, e.g. "this apple". And so on...

As I look through this list, it seems to me that all the most common features are within A1/A2 level that you are interested in.

  • Thanks, but that is not the intent of my question. What you are suggesting would be more appropriate for Linguistics Stack Exchange. – Christophe Strobbe Oct 14 '16 at 16:24
  • @ChristopheStrobbe What's wrong with this approach? – michau Oct 14 '16 at 16:29
  • I'm only asking for example sentences (plus why they should be added), not for a list of language features. Finding common language features is one of the approaches that can be used to find suitable sentences, but that list should only be a means to an end, not the end itself. – Christophe Strobbe Oct 14 '16 at 16:35
  • @ChristopheStrobbe I did provide a sentence: "I don't have the red apple". Is it a problem that I also provided reasoning behind it and a way to build more such sentences? I don't get it. – michau Oct 14 '16 at 16:40
  • Sorry, you are right. I had apparently overlooked the example. – Christophe Strobbe Oct 14 '16 at 16:50

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.