I am interest in reversal errors in personal pronoun acquisition. My knowledge comes mostly from studies done with English-speaking children, and I was wondering if there is any languages where this effect would not occur or occurs in a fundamentally different way.

Background: Reversal errors

The referent of pronouns shifts with conversation roles. For pronouns a child must realize that she needs to reverse the pronouns in order to communicate correctly. A father will refer to his daughter as you and himself as me, but the daughter must infer that when she speaks to her father, he becomes you and she becomes me. If a child imitates what she hears, she will refer to herself as you and to everyone else as me - a reversal error.

  • 14
    An amazing factoid is that this reversal occurs even in sign languages where the pronouns involve pointing at the referent. Childred acquiring sign languages learn the pointing as a fixed abstract symbol before they learn that it also involves spatial indication! Sep 16, 2011 at 8:52
  • @hippietrail thanks for finding the typo, in the beta everybody can edit questions, so you can also just fix up such typos yourself and I doubt the author would mind. The factoid about sign-language is really interesting, do you know of a good reference for studies on this? I would really like to look at their numbers. Also, your comment could probably be turned into an answer (if it includes the reference) since it gives good evidence for there not being fundamental differences even in things like sign-language where you would most expect them. Sep 16, 2011 at 9:17
  • Really? I thought there still had to be a minimum so-many-characters change so I didn't even try. Sep 16, 2011 at 9:19

3 Answers 3


Perhaps not.

For as surprising as it sounds, even children learning sign language as their first language go through this same process despite the fact that many if not all sign languages use some kind of pointing for at least first and second person pronouns.

They first learn the abstract symbol for the pronoun ignoring the spatial indication and will point to themselves to indicate "you"! Later they make the same change hearing children make and switch to pointing to the person addressed to mean "you".

Here are some articles I could find on the Internet:

  • 1
    That is totally messed up. In that it is crazy wild. That pointing at something comes later than the abstract symbol. This one little experimental fact would seem to turn over a lot of philosophy concerning signs and symbols and mimicry. Thanks for bringing it up, things are never as simple as we imagine.
    – Mitch
    Sep 16, 2011 at 13:26
  • 3
    This is a perfect example of why languages and linguistics are so endlessly fascinating! (-: Sep 16, 2011 at 15:12
  • 3
    Not that children are like animals, but it's like when you you hide a ball from a dog or cat and you try to point as to where to get it, and all the animal does is look at your hand instead of following the direction of pointing.
    – Mitch
    Sep 16, 2011 at 15:15

I think Japanese (and maybe some other languages) might be special in this case. Japanese uses personal pronouns much less than e.g. English, and second person pronouns are in many cases completely avoided.

When speaking to kids, parents will usually refer to themselves as ママ mama/パパ papa or お母さん okāsan/お父さん otōsan, i.e. Mommy/Daddy or Mother/Father. When referring to the child, they will say child's name+君 kun/ちゃん chan (honorific suffixes), or sometimes a nickname which might or might not have the honorific suffix "built in". Personal pronouns are hardly ever used.

Because of this, the reversal errors do not occur in their usual form. However, different but related problems occur:

  1. The child has to learn to remove the honorific suffix when referring to themselves.
  2. Even without the honorific suffix, referring yourself by name is considered childish. Children have to learn to use a first person pronoun when referring to themselves.
  • 1
    do you know any papers that explore this? I am in particular curious about reversal errors where the child refers to themselves by name (with or without suffix). Feb 14, 2012 at 14:07
  • 1
    @ArtemKaznatcheev Unfortunately not. I only know this from my personal experience.
    – dainichi
    Feb 14, 2012 at 14:12
  • 2
    Clancy, PM (1985) "The acquisition of Japanese" suggests that the use of honorific suffixes is in general very complicated, and often not mastered by junior-high level children. Feb 14, 2012 at 14:28
  • 1
    This answer is of particular interest to me. As a father- when I have previously communicated with my young daughter (now 2 yrs of age) I would always refer to myself as dada and then herself by her proper name- simple sentences: "I'll give you some milk" = "dada will give cailyn some milk" however upon nearly turning 2 - the words "you" and "me" have come up and sorted through. Now when i ask who is "you" she points at herself- who is "me" points at "dada"- but when i say "dada is me" she says no and points to herself again! somewhere @ 1.5 yrs of age- she saw what the words truly meant.
    – user1108
    Jun 17, 2012 at 4:02
  • and since i'm a happy father and blather on and on---- i guess i wonder- when the significance of those words became apparent to her (and by that i mean she was able communicate that she knew the difference between "me" and "you" to myself) was that the point when she finally recognized herself as a conscious entity?
    – user1108
    Jun 17, 2012 at 4:07

It's notable that the inclusive "we" is symmetric, and a that an inclusive sense of "I" and "you" is not completely unthinkable (that is, one can differentiate that the appelative sense of an assertive "I am hungry" is inclusive, and that "we will go now" is nevertheless egoistic; nevermind "feed me", which I'd deem an absurdity for a young child to say :D).

I'm kind of answering to the sign-language related answers. Nevertheless, if reading the question's word "language" as "logic system" (synonymously called "theory" in maths circles), then I might even post this as an answer, deeming an inclusive logic a sub-language. It's not surprising that single valued logic develops before a two valued logic.

Note that me, mi etc means "we" in some languages. Also compare Ger "man", French "on", Eng "one" for an indefinite pronoun, and Proto-Indo-European *sem- "one; together", *se- or *sel- "self, sole".

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.