Disambiguating English with Esperanto


Some problems

As a learner of English (who isn't?), I've come across a lot of inconsistent rules and exceptions in the language, and I can imagine they exist in most other natural languages. People just (have to) get used to them. In general:

  1. Grammar is complex (13+ tenses!) and verb modification is irregular.
  2. Ambiguous semantics: the same sentence can mean different things.
  3. Spelling does not reflect pronunciation; accents vary wildly.
  4. Vocabulary is vast and borrows from many other languages.
  5. Spoken vs written disconnect: slang, idioms, figures of speech.

What can be done to improve it? I'm not sure, but there have been been many efforts, mainly by simplifying sentence structure and restricting vocabulary.

Of course, there's no changing the way English is spoken in the mainstream, but ideas are ideas, so here's one.

Designing a new language from the ground up is hard, so in any case vocabulary should be kept constant (if not restricted to simpler words). This means points 3, 4, 5 are off-limits, there is no way to improve on those without modifying English words or how they are spelled (related). So, grammar has to change.

I've been thinking about how Esperanto grammar constructs (the few that I know) can be used to make English and similar languages to be more regular. This should address point 1 and, to a certain extent, point 2 as well. Specifically, this post explores that in the context of ambiguity in sentences, which is apparently an inherent property of all natural languages.

Grammar simplification

All we're taking from Esperanto are the suffixes and sentence structure.

Suffix Meaning
-o noun
-a adjective
-e adverb

The immediate problem are the words themselves. Most English words have grammatical meaning baked into their "root words", like 'often' or 'again'; modifying them with suffixes leads to non-obvious interpretations ('again-o': what would 'again' mean as a noun?) and should be avoided, but what if someone does it anyway? Exceptions regarding special word forms will have to remain.

A good practice might be to avoid special words and phrases altogether, and stick with simple suffix-friendly words, say, writing 'frequent-e' to mean 'frequently' instead of 'often'. But I don't know how 'again' could be broken down into a word and a suffix, so it might have to be a 'special word'.

The verb suffix system for Esperanto is well-defined:

Suffix Meaning Example Interpretation
-as present I do-as I do
-is past I do-is I did
-os future I do-os I will do
-u imperative do-u it-n! do it!
-us conditional I do-us I would do
-i infinitive do-i it to do it

These should cover most cases when forming simple sentences.

There are also more complex verb suffixes, which may be unnecessary:

Suffix Meaning Example Interpretation
-anta present active I be-as do-anta I am doing
-inta perfect active I be-as do-inta I have done
-onta future active I be-as do-onta I am about to do
-ata present passive I be-as do-ata I am being done
-ita perfect passive I be-as do-ita I have been done
-ota future passive I be-as do-ota I am about to be done

Going further

But what if we wanted to extend verbs such that they could be nouns, adjectives or adverbs?

Consider 'repeat-e'. It means 'repeat', as an adverb. But how exactly should it be interpreted?

  1. Repetitively
  2. Repeatedly

Both are adverbs and they overlap in meaning -- maybe they should not be different words in the first place. But the rule was to not modify English words more than necessary. Trying to get at meaning 1 leads to another issue:

  1. repetition-e
  2. repetitive-e
  3. repetitively-e

Which one is correct? Perhaps all of them, although 3 is redundant. We see that mixing English suffixes and (abused) Esperanto suffixes leads to even more choice.

So, abusing Esperanto's suffix system, words can become verbs, adjectives or adverbs, as appropriate. Let's say prepositions (over, on, at), articles (a, the), conjunctions (and, or) are not changed, and omitted if appropriate.

In this system, the words to which we attach suffixes are the English words, and those words don't have to be actual root words. It's not always possible and might lead to ambiguity (e.g. 'repeat-e').

Here's how the system might be used to disambiguate some sentences:

Syntactic ambiguity

Also known as structural ambiguity or grammatical ambiguity. This is when the structure of the sentence makes its meaning unclear.

A popular example is:

I saw a man on a hill with a telescope.

There are several interpretations:

  1. The hill has a telescope.
  2. The man has a telescope.
  3. I used a telescope to see the man.

In English, the sentence might be disambiguated by context, or a change in sentence structure:

1. I saw a man on a hill, which had a telescope.
2. I saw a man, who had a telescope, on a hill.
3. I saw, using a telescope, a man on a hill.

If it were "I saw a man with a woman with a telescope", disambiguation would be much more difficult. In speech, emphasis and pauses could be used.

A possible translation:

1. I see-is man-on on have-telescope-a hill-o.
2. I see-is have-telescope-a man-on on hill-o.
3. I use-telescope-e see-is man-on on hill-o.
  • the article 'a' is dropped.
  • -o marks a noun.
  • -on marks a noun that is the direct object of the verb (here the man is the object of seeing, so 'man-on').
  • -is turns a verb into past tense.
  • -e marks an adverb.
  • -a marks an adjective.

But using -e and -a to mark adverbial and adjectival clauses is probably not a good idea, since they might get more complex. What if instead of 'have telescope' we had 'had one telescope and now have two'? More extensible but less natural would be:

1. I see-is man-on on hill-o which have-as telescope-on.
2. I see-is man-on who have-as telescope-on on hill-o.
3. I see-is use-ante telescope-on man-on on hill-o.
  • -ant- marks continuity, and again, -e makes an adverb so 'use-ante telescope-on' is an adverbial clause that would describe an action as 'using a telescope', in this case the action is 'seeing'.
  • -as turns a verb into present tense.

Lexical ambiguity 1

She went to the bank.

What bank? Merriam-Webster lists 5 entries for 'bank', but the sentence structure rules out the verb definitions. Also, the 5th definition ('a group or series of objects arranged together in a row') seems quire rare. Here 'bank' is either:

  1. the bank of a river
  2. the bank to deposit your money in

Further context is needed, and often it is available. There is no good way to deal with this without adding an extra clause.

1. She go-is to the-bank-o of the-river-o.
2. She go-is to the-bank-o for store-i money-o.
  • -i makes a verb infinitive: 'store-i money-o' is 'to store money'.

Lexical ambiguity 2

I like understanding people.

This time the sentence structure doesn't help us, and the ambiguity arises from the fact that 'understanding' can be a verb or an adjective:

  1. I like people who are understanding.
  2. I like to understand people.

Disambiguation by inventing new English words (this is irregular, and not always possible):

1. I like understanders.
2. I like people-understanding.

Esperanto grammar is particularly well-suited for this. Changing the suffix changes the role of the word, and the structure is untouched:

1. I like-as understand-ajn person-ojn.
2. I like-as understand-i person-ojn.

In the first sentence:

  • -j, pronounced like 'y', marks plurality: 'people' becomes 'person-oj', then it becomes 'person-ojn' because it is the direct object of 'like'.
  • The adjective 'understand-a' describes 'person-ojn', so the former follows the latter's suffixes: 'understand-ajn'.

In the second sentence:

  • -i marks an infinitive: 'understand-i' is 'to understand'.
  • 'understand-i person-ojn' means 'people-understanding' as a noun, it is the direct object of 'like', so there is an -n.

Morphological ambiguity

Somewhat related to lexical ambiguity is morphological ambiguity, which (as I understand) is when adding prefixes and suffixes leads to (further) ambiguity.

This shoe is untieable.

Interpretations:

  1. This shoe can be untied.
  2. This shoe cannot be tied.

With punctuation (different emphasis in speech):

1. This shoe is untie-able.
2. This shoe is un-tieable.

With Esperanto grammar:

1. This-shoe-o can-as be-i not-tie-ita.
2. This-shoe-o not-can-as be-i tie-ita.
  • not- negates
  • 'can' is considered a verb
  • -ita means 'has been done': 'tie-ita' means 'tied' as in the adjective (note it ends with -a), e.g. tied shoes.
  • 'be-i tie-ita' means 'to be tied'
  • 'can-as be-i tie-ita' means 'be able to be tied'
  • 'can-as be-i not-tie-ita' means 'be able to be not tied'

The "comma problem"

I invited my parents, Jack and Jill to the party.

Interpretations:

  1. Two people are invited: my parents who are named Jack and Jill.
  2. Four people are invited: Jack, Jill and my parents.

Changing the sentence structure, and punctuation can disambiguate:

1. I invited my parents, Jack, and Jill to the party.
   I invited Jack, Jill, and my parents to the party.

2. I invited my parents (Jack and Jill) to the party.
   I invited to the party, my parents: Jack and Jill.

Esperanto grammar (as well as more regular punctuation):

1. I invite-is I-an parent-ojn, Jack-on, and Jill-on to the-party-o.
2. I invite-is I-an parent-ojn who name-as Jack and Jill to the-party-o.
  • 'I' becomes 'my' by adding -a.
  • Since 'my parents' is a direct object we also add -n, so 'I-an parent-ojn'.
  • I'm not sure if Jack and Jill need -o's in the second sentence.

The "it problem"

The dog furiously chased the cat until it suddenly fell.

What is 'it' referring to?

  1. The dog chased the cat until the cat fell.
  2. The dog chased the cat until the dog fell.

I can't think of a way to disambiguate this while keeping the 'it'.

1. The-dog-o fury-e chase-is the-cat-on until the-cat-o sudden-e fall-is.
2. The-dog-o fury-e chase-is the-cat-on until the-dog-o sudden-e fall-is.

Sample translation

Source. With slight alterations:

English grammar is complex, making it difficult to remember, master and use
logically. Ensuring you use the correct grammar can be tricky, especially
when you are in conversation with someone and they are speaking at an
alarmingly fast pace. Learning grammar is like learning to drive, you can
learn all of the theory, rules and regulations, but you won't be good at it
unless you practice it and it starts to become second nature to you. Grammar
is extremely important, incorrect use of grammar can confuse the person you
are speaking to and even change the meaning of what you are communicating,
what's more is native English speakers are hyper-aware of grammar and will
notice almost immediately if a grammatical error is made, even if this is
the smallest of errors. Don't get me started on accents...

Esperanto grammar (kind of):

English-a grammar-o be-as complex-a, such-that difficult-as remember-i,
master-i and use-i logic-e it-n. Ensure-i that you use-as correct-a
grammar-o can-as be-i trick-a, special-e when you converse-as with
person-o and they speak-as at alarm-e fast-a pace-o. Learn-i grammar-on
be-as like learn-i drive-i, you can-as learn-i all the theory-ojn,
rule-ojn, regulation-ojn, but you not good-os at it unless you
practice-as it-n and it start-as become-i second-a nature-o to you.
Grammar-o be-as extreme-e important-a, not-correct-a use-o of grammar-o
can-as confuse-i the person-on whom you speak-as, and even change-as the
meaning-on of what you communicate-as, more-e, native-a English-a
speaker-oj hyper-e aware-as of grammar-o and notice-os almost immediate-e
if grammar-a error-o make-itas, even if this be-as the most small
error-o. Not start-u I-n on accent-oj...

Most of this is probably inaccurate, but it is interesting still.

Conclusion

Clearly, mixing Esperanto grammar and English vocabulary leads to language that is unpleasant and difficult to read, speak, write or listen to. I would say this is mostly due to the heavy use of '-' (maybe apostrophes would have been better) and the difference in the pronunciation systems -- more irregularity in English.

Perhaps a system similar to this may be useful in informal human-readable speech tagging, to disambiguate speech, or more generally, help separate grammar from vocabulary (I have not looked into that). Importantly, English vocabulary here could be swapped out for some other language's vocabulary.