Re: [xmca] Palindromes and Semordnilap

From: Mike Cole (
Date: Sun Jan 07 2007 - 20:30:19 PST

Thgir. oN?

On 1/7/07, David Kellogg <> wrote:
> I've often wondered if "teach" and "cheat" (which I have used to teach the
> "five word rule" on using sources) are truly palindromic. The problem, of
> course, is the "ch" sound, which in IPA is composed of two phonemes, namely
> /t/ and /sh/ (it's not really written like that in IPA, it's more like the
> f-hole on a violin without the crossbar).
> If we treat /t/ and /sh/ as two different phonemes, which of course
> it/they is/are in most languages, then "teach" and "cheat" are not
> palindromic, because the /t/ and /sh/ are not reversed. The reverse of
> "teach" would be something like "shteat".
> So you see maybe Puppeg Gruppifuppin's hyper-orthographic undergrads
> were onto something. "Say yes", on the other hand, is a true spoken
> palindrome, meaning you can actually digitalize it and play it backwards,
> and it's understandable.
> Mike was wondering, in his pragmatist way, if any of this has more than
> curiosity value and in particular if it has anything to do with phonics and
> reading instruction. I don't really know, and I'm a little wary of the
> various studies that show that Chinese kids have a MUCH lower rate of
> dyslexia than kids in other countries because of their writing system. But
> there IS a well known form of dyslexia which involves reversing phonemes,
> rather as I tend to do with words like "left" and "right" and "yin" and
> "yang". We could call it "semordnilap", involuntary palindromery. Well,
> semordnilap is apparently non-existent in China.
> In any case, I think that BOTH spoken palindromes ("teach-cheat" and
> "say yes") point to the problem of using phonemes, which as Peg points out
> were really derived from written language, to analyze spoken language. With
> "teach" and "cheat", we are unsure on exactly what constitutes a phoneme
> (/t/ and /sh/ can be separate phonemes in English too you know). And "say
> yes" indicates TWO different phonemes when there is really only one.
> That brings me to a rather silly remark I made a few posts ago, which
> David Preiss fortunately caught me on. I was criticizing Tomasello's
> statement that the "only" reason why language develops a "separate" layer of
> lexicogrammar (which unlike phonology and unlike semantics does not
> interface directly with the physical world) is the need to have many ways of
> saying the same thing.
> I said that I thought that the need to have one way of describing many
> things was more pressing, and of course this is true: that's why Piaget
> complains that children's first words are so polysemic. But of course you
> don't need lexicogrammar to use one utterance to refer to many different
> things; the fact that in "Say Yes" one phoneme can refer to more than
> grapheme is just one example.
> The other meaning of "palimpsest" is a "magic slate"--you know, those
> waxy things with covered with a sheet of plastic that you can write on and
> then erase by lifting the sheet with a ripping sound. Freud was fascinated
> by them, and constructed a whole theory of memory around the metaphor. They
> are very useful when someone catches you writing something silly, no?
> Rrrrip!
> David Kellogg
> Seoul National University of Education
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around
> _______________________________________________
> xmca mailing list
xmca mailing list

This archive was generated by hypermail 2b29 : Thu Feb 01 2007 - 10:11:31 PST