[Corpora-List] Phonemes of primitive languages

[John A Goldsmith]

Phoneme frequencies, as far as I can see --- and I've looked -- don't 
display a Zipfian distribution. There are some good reasons not to 
expect them to, in the sense that we've come to expect Zipfian 
distributions to arise from processes that frequently introduce new 
items into the set in question, all the while preferring to repeat items 
that have appeared with higher frequency. Words can be thought of as 
generated by such a process without too much straining of the 
imagination, but phonemes cannot.
John Goldsmith

On 12/03/2010 02:53 PM, Rich Cooper wrote:
> The minimum description length work on compression and learning, especially
> the AIXI paper in some AAAI publication, describes a combination linear and
> discrete model of a learner that should be something akin to Zipf's
> insights.
>
> -Rich
>
> Sincerely,
> Rich Cooper
> EnglishLogicKernel.com
> Rich AT EnglishLogicKernel DOT com
> 9 4 9 \ 5 2 5 - 5 7 1 2
>
> -----Original Message-----
> From: corpora-bounces at uib.no [mailto:corpora-bounces at uib.no] On Behalf Of
> Vlado Keselj
> Sent: Friday, December 03, 2010 12:33 PM
> To: Yuri Tambovtsev
> Cc: corpora at uib.no
> Subject: Re: [Corpora-List] Phonemes of primitive languages
>
>
> On Sat, 4 Dec 2010, Yuri Tambovtsev wrote:
>
>    
>> Dear colleagues, can we think that in primitive that is seminal
>> languages the speech sounds were also simple? I compared the frequency
>> of occurrence of speech sounds of some 300 languages which I have in my
>> phonetics corpora and came to the conclusion that simple phonemes occur
>> more frequently and in more world languages. Is there any law under it
>> or is it just a chance? And what is a chance in a language speech chain?
>> Looking forward to hearing from you to yutamb at mail.ru Be well, Yuri
>> Tambovtsev, Novosibirsk
>>      
> There could be something similar to the Zipf's law.  I believe that in his
> PhD thesis, Zipf postulated that his law is more universal than a
> relation between word ranks and their frequencies in a text, and that it
> was a consequence of minimization of human effort required in
> communication.
>
> --Vlado
>
> _______________________________________________
> Corpora mailing list
> Corpora at uib.no
> http://mailman.uib.no/listinfo/corpora
>
>
> _______________________________________________
> Corpora mailing list
> Corpora at uib.no
> http://mailman.uib.no/listinfo/corpora
>    


_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora