>"Aren't" as a contraction for "am not" is what I have said all my 
>life, as a native speaker of the American version of English.
>"Amn't" is what I guess I would logically say if "aren't" didn't
>exist, and it is in fact what a Russian friend of mine says all
>the time.  To her, "aren't" for "am not" sounds strange, since it
>is actually incorrect but she is perfectly comfortable with "amn't."  
>She doesn't like "am I not" because she thinks it sounds too formal.

That's interesting.  Many Americans (myself included) learned to use
"aren't I?" in tag questions for the first person; but you're the
first I've ever heard who confessed to using it wherever "*amn't"
might have gone.  As in, for instance, "I aren't going"?  I don't
think I've ever heard that; people will simply use the alternative
contraction of "I'm not" to avoid that.

I don't believe, in fact, I've *ever* heard "aren't" used with the
first person outside of a tag question.  So it's partly a syntactic
matter why this particular contraction gets used -- in a tag question
you can't use the alternative contraction because the order of the
pieces is changed from pronoun + be + not to be + pronoun + not.

And you pretty much have to contract in a tag question, because the
full forms (", is he not?", ", am I not?", etc) are very highly marked
for formality.  As many before you have pointed out, they make us feel

So we attempt whatever strikes us as the best compromise, and "aren't"
seems to work.  But, like using "wake" in the present perfect, the
best solution isn't always good enough for some speakers, and they
simply learn to avoid the construction.  Luckily, it's a fairly rare
occasion when one has to ask a question in the first person.

--- Followup:

>>Please tell me more about using "wake" in the present 
>>perfect. I have always felt so uncomfortable about this 
>>that I use circumlocutions to avoid using it. 
>All of these make sense to me: 
>"I woke/woke up/awoke at 7:00 AM." 
>"I should be wakened/awakened . . ." 
>"I wake up/awaken around ..."  
>"I was wakened/awoken ..." 
>"I woke/awoke ..." 
>"Hush! You'll wake the baby!" 
>"When do you normally wake/awaken?" 
>"The noise does/did/will  
>    wake me/awaken me/wake me up." 
>"I/he waked up at 7:00 AM" gives me the willies.

Me too.  I mentioned "wake in the present perfect" because it's my
best example of an English "defective" verb; i.e, one that lacks a
principal part, in this case the past participle, the one that's used
in the present perfect construction.  That's formed, as you know, by
using a form of "be" followed by the past participle of the next verb
in the phrase.  The problem is that "wake" just doesn't *have* a past

People get around the problem in a number of ways.  You can always
avoid the present perfect with a little care in rephrasing, or
you can use one of the other verbs that are formed with available
derivational scraps:

   wake      woke        *
   waken     wakened     wakened
   awake     awoke       *
   awaken    awakened    awakened

The -en suffix is a relatively non-productive causative affix;
white/whiten, dark/darken, red/redden, stiff/stiffen, live/liven,
(but: blue/*bluen, orange/*orangen, tall/*tallen, big/*biggen), etc.
Since "wake" is inchoative (i.e, it refers to changing from one state
to another), its causative means to cause someone to wake, which can
be oneself, so there's an inchoative sense available, too, and that
means "waken" can be used as a synonym for "wake". And, causative
verbs with -en are regular, so there's a guaranteed past participle.

The a- prefix is an adjective-forming derivational morpheme that goes on
a lot of interesting words. It's not very productive, either, so the
words are often archaic. Exx: agley, abaft, atop, anew, etc. There may
be several sources for it; I haven't looked up the etymology for
"awake(n)". Anyway, "awake" *is* an adjective, and names the state
begun by the inchoative "wake(n)". And there's the other end of the
chain -- you frequently get word sets in English like

  causative: open/close   inchoative: open/close  stative: open/closed

where the same form is used in all three, or sometimes only
two, of the different senses.

The upshot (that's another interesting word, isn't it?  Is it an
old metaphor?) is that people vary quite a lot in how they interpret
the meanings of these four verbs, two regular and two irregular.
One principle that's commonly used is to assume that if there are
two different words there must be two different meanings, and make
up (or discover) a distinction for them to express.  In this case,
that's somewhat balanced by the need to find a perfectible participle.

Other English defective verbs include 'beware' (usable in the
imperative only), 'blowdry' (try forming the past tense and you'll see
what I mean), 'born' (technically, a "deponent" verb, with only
passive forms), and the modal auxiliaries, but they're so irregular
anyway that's hardly surprising.  Latin had lots more.

Anyway, you're in the same boat as the rest of us.  Funny we can't
just declare it a regular verb, but there it is.  Even The Academy
can't make it sound right if it sounds wrong.

 -John Lawler                     More grammar
  Linguistics Program   University of Michigan
 "..and, who knows? Maybe the horse will sing."