Tempodivalse:Speakers of other languages practically never fail to mark the object with a non-nominative case. The odds of a rusophone saying *Ja ne smotrel programma* are nil. However, in Esperanto, inadvertently dropping the accusative (among proficient speakers) seems to be rare but not unheard-of - you would obviously identify *Mi ne vidis la programo* as wrong, but might blurt it out anyway, without thinking.

I wonder how to explain this.

One of my learned languages is Japanese, where dropping grammatical endings is basically a feature. It is understood to be a characteristic of informal speech, rising to the level of grammar itself. Spanish drops pronouns. English drops correlatives. "I don't believe (that) the Earth is flat."

I don't have enough formal linguistics to offer an explanation as to why these things get dropped, especially in informal circumstances, but it does seem to be a feature of languages. My conjecture would be that it mainly happens in SVO sentences where you're unlikely to misunderstand the meaning. Without the accusative I can take a good guess what is meant by, "La hundo mangxis la pomo(n)". The apple probably did not eat the dog and Esperanto is SVO enough to offer that additional context. But yeah. Conjecture.

It hardly requires any explanation that a beginner in Esperanto might drop an accusative ending. And even spertuloj are generally not in a comparable position to persons speaking their mother tongue in respect of errors of grammar.

With beginners, I can well imagine that an omission of 'j' can equally well occur as an omission of 'n', particularly in the matter of agreement of adjectives.

As has been mentioned, it is easy when writing a post to get distracted in re-drafting and not make all the changes required by the new sentence structure.

Some of us are very poor typists anyway and we do not have the benefit of a grammar and spelling checker when typing in Esperanto, as English speakers would enjoy when typing on a computer in English.

My intuition is that I would make fewer errors in speaking than in writing. One tends to speak in less complex sentences than one writes.

Of course it may be that the predominance of SVO in Esperanto is a factor in the more frequent omission of 'n' than 'j' (if that actually occurs).

But even 'j' can be redundant. For example in 'La leteroj estas pretaj', the plural has already been expressed in the subject, making the second 'j' redundant. (Redundancy can be a good thing in language)

In any empirical investigation, perhaps one should compare error rates in sentences where a 'j' is redundant with sentences where the 'n' is redundant (like la hundo manĝis la viandon).

kaŝperanto:Is there any collection of more objective evidence along these lines, or is this mostly a subjective conclusion? I would weight the number of omissions against the number of proper uses before drawing any conclusions.

I'm not sure if this question was directed at me, but I'm basing my claim on my graduate education, and specifically the miscue analysis developed by Marie Clay, a literacy pioneer who died a few years back. Her work was instrumental in helping intervention specialists develop data-based hypotheses about why readers miscue (commit a speaking or reading error) the way they do. I can't point to a specific article, but the thought process is bound up in her (very short) book "Running Records."

Tempodivalse:Do speakers of languages with no plural (e.g. Chinese, Indonesian) forget to add -j to plurals in similar circumstances?

I think Mayans speaking Spanish classify as "similar circumstances".

Why? Spanish and Esperanto are structurally almost identical. Q'eqchi' does indeed have the grammatical category of numerus (two of them: singular and plural), however as the Mayan languages are head-marking, the plural is not marked on the plural noun, but on the head of the phrase containing it. For instance you can say "the dog" (li tz'i' ), "with the dog" (rik'in li tz'i' ), "with the dogs" (rik'ineb' li tz'i' ), "I saw the dog" (xwil li tz'i' ) and "I saw the dogs" (xe'wil li tz'i' ), but you cannot say "dogs" without embedding it in a phrase. As you can see the plural of dog is never marked on the word "dog", but on the words "with" and "saw".

Q'eqchi' speakers indeed tend to drop the plural of nouns while speaking Spanish. In my experience they drop them almost 100% after numerals or nouns of measure (dos quetzal de tortilla instead of dos quetzales de tortillas). They also quite often drop the plural marker in subjects, objects and prepositional phrases, but not as often as in the aforementioned cases. In other cases they translate verbatim into Spanish completely avoiding the plural: "mi mamá, mi papá" (linna' linyuwa' ) instead of "mis padres".

With personal pronouns, however, this dropping does not appear (démelas; te los doy). This might be due to the fact, that Q'eqchi' doesn't have personal pronouns[1] and hence there is no built-in rule (in the heads of the Q'eqchi' speakers) which states "don't mark plural there".

So yes: Speakers of languages where the plural is marked differently forget to add a plural marker in similar circumstances.

----
[1] Some grammars classify laa'in, laa'at, etc as personal pronouns. The word laa however is a static verb which is inflected for person. I.e. laa'in doesn't mean "I", but "this is me"; and laa'at doesn't mean "you" but "this is you".

TKNIGHT:

kaŝperanto:Is there any collection of more objective evidence along these lines, or is this mostly a subjective conclusion? I would weight the number of omissions against the number of proper uses before drawing any conclusions.

I'm not sure if this question was directed at me, but I'm basing my claim on my graduate education, and specifically the miscue analysis developed by Marie Clay, a literacy pioneer who died a few years back. Her work was instrumental in helping intervention specialists develop data-based hypotheses about why readers miscue (commit a speaking or reading error) the way they do. I can't point to a specific article, but the thought process is bound up in her (very short) book "Running Records."

Yep, just wondering to what depth you have explored your ideas. As sudanglo mentioned, I would wonder if the errors would be more or less prevalent in spoken Esperanto. This would almost totally eliminate the "typo" or "editing" variables.

Yep, just wondering to what depth you have explored your ideas. As sudanglo mentioned, I would wonder if the errors would be more or less prevalent in spoken Esperanto. This would almost totally eliminate the "typo" or "editing" variables.

Ah, ok. Well, I'd have to classify my opinion as entirely subjective, largely because it's inductive reasoning starting with my circumstance and building out. However, contrary to Sudanglo's opinion, I'd expect to make many more errors speaking because of the nature of my training in the language's visual code, as opposed to the spoken side. But again, that's subjective to my experience as a largely reading/writing student, lacking other speakers in my area, and the time/funds/courage to engage in speaking practice over skype.

Contrary to Sudanglo (and speaking as a komencanto, not as an experienced speaker) I'd contend that I'm more likely to drop the -n, or confuse it with a -j, or make some other miscue because of the synthetic and multi-channel demands of speech compared to writing. While Sudanglo is correct that re-drafting can lead to errors, writing presents the communicator with the chance to revise before showing your work to others--speech doesn't. Speaking requires both the conscious coordination of the speaking apparatus--hyper-conscious, for a komencanto--as well as an active synthesis of grammar and vocabulary, spoken aloud, all at once in a jumble. On the contrary, for the same komencanto, each word can be typed and parsed individually, and the whole thing sits before you for reference as new words are added. Given that, it seems to me more likely that the speaker will remember the -n when they can see the linguistic relationships between the words printed on page or screen, no?

There are speakers of Esperanto in New York. There is even a permanent office at the UN. Neil Blonstein is very active there.

I'm going to suggest a different way of looking at this.

When you learn a language with cases you naturally think of the nominative as the principle form of the word, after all it appears first in declensions and is what you see when look it up in a dictionary.

However native speakers don't necessarily think like that. When Latin lost its cases as it evolved into the romance languages it wasn't the nominative that was preserved. Most romance languages base their nouns on the Latin ablative form apart from French which, just to be different, bases its words on the accusative. Incidentally that explains why the neuter and masculine genders merged.

Maybe speakers of Russian and some other languages subconsciously think of the accusative as the principle form, in which case they don't so much mark the direct object with the accusative as mark the subject with the nominative.

A commentator has said that Russian neuter nouns have never had an accusative, but maybe it would be better to say that they have never had a nominative. I read somewhere the suggestion that the neuter gender of Indo European languages was originally for inanimate objects, in which case, as inanimate objects are not normally the subject of a sentence why would they need a nominative? On the odd occasions they are the subject they could be not marked as such, but context would likely make the meaning clear.

If that's right then speakers don't forget to mark the direct object with the accusative because there is nothing to forget. They'd be more likely to occasionally forget to mark the subject with the nominative. I'd be interested to hear from Tempodivalse if that's something a rusophone might possibly do.

Eo on the other hand is different. The nominative is clearly the main form as a suffix is added to form the accusative and prepositions are usually followed by the nominative.

NickRobinson:Maybe speakers of Russian and some other languages subconsciously think of the accusative as the principle form, in which case they don't so much mark the direct object with the accusative as mark the subject with the nominative.

Very interesting ideas.

When I speak Esperanto (or type it very quickly, which is like speaking in terms of spontaneity), I do occasionally catch myself adding the -n where it doesn't belong. In fact, this seems to occur as often or more often than forgetting the accusative altogether. I never considered that I might intuitively be perceiving the -n as the "default", but it makes sense.

Since I never "learned" Russian, I can't say if my intuitions perceive any particular case as being the "principle" one, though if you just ask me to utter a noun, I will most likely automatically put it in the nominative (which, for certain declensions and genders, coincides with the accusative). Dictionaries, of course, use the nominative form.

Accidentally using the nominative instead of the accusative (or vice versa) is probably one of the rarest mistakes a native would make in Russian (hence my surprise when it happens in rusophone Esperantists or even native Esperantists). I did it once recently, but I immediately felt it was wrong; and in my defense, I use Russian far less often than English now, so I am likely suffering from language attrition.

In my experience, it is more common to mix the non-nominative cases, sometimes with prepositions (*Skuchaju po vas* instead of Skuchaju po vam), and especially with the prepositional/locative (*Na berege* instead of Na beregu).

sudanglo:Some of us are very poor typists anyway and we do not have the benefit of a grammar and spelling checker when typing in Esperanto, as English speakers would enjoy when typing on a computer in English.

LanguageTool is a cross-platform, open-source grammar-checker also available for Eo. It’s pretty useful to me.