Message 2005-12-0004: Re: Minor rewordings of Article 17?

Mon, 26 Sep 2005 21:40:11 +0200

[Previous by date - Minor rewordings of Article 17?]
[Next by date - Fw: Minor rewordings of Article 17?]
[Previous by subject - Re: Minor Rewordings of Article 7?]
[Next by subject - Re: Monotypic genera and the PhyloCode]

Date: Mon, 26 Sep 2005 21:40:11 +0200
From: [unknown]
To: PML <phylocode@ouvaxa.cats.ohiou.edu>
Subject: Re: Minor rewordings of Article 17?

This is a multi-part message in MIME format.

--Boundary_(ID_U4J5arT09WR93qxpiEYhbQ)
Content-type: text/plain; charset=3DISO-8859-1; format=3Dflowed
Content-transfer-encoding: 8BIT

David M's concerns are fully justified in my opinion.

Since "Latin alphabet" is open to interpretation and the very purpose=
 of=20
definitions is to completely exclude ambiguity, would it not be best =
to=20
specify explicitly somewhere in the code "The Phylocode Alphabet for=
=20
construction of names, consists of the letters: a, b, c, d, e, f, g, =
h,=20
i, j, k, l, m, n, o, p, q, r, s, t, u, v, w, y, z." and then refer to=
=20
this alphabet when needed.

As for "pronounceable", I would ask pronounceable for whom? Although=
=20
English is the lingua franca of the scientific community, pronunciati=
on=20
of large numbers of ordinary English words remains problematic for ma=
ny=20
native speakers of other languages. This is a problem that the invent=
ors=20
of artificial languages that pretend to lingua franca status, have=
=20
already tackled: Ido, Novial, Volap=FCk, Interglossa, Esperanto and=
=20
Interlingua. It might be an idea to consult a scholar of such languag=
es=20
to find a formulation that defines "pronounceable" in a less ambiguou=
s,=20
and an unbiased way.

Clearly there is an appeal to common sense in  recommendation 17A as =
it=20
stands. However what is pronounceable by common sense is clearly=20
subjective relative to culture and linguistic aptitude.

I have never got my tongue round Nopalxochia for example, even though=
 I=20
speak several languages. For many people words that start with s=20
followed by a consonant are just as tongue twisting as lx is for me.=
=20
Excluding x was a good choice since its pronunciation is culturally=
=20
ambiguous and phonetically redundant.

Unfortunately any syntactic rules governing pronunciation will=20
inevitably require the Phylocodisation of the natural names of botani=
sts=20
and places when these form the basis of new botanic names, but this i=
s=20
no worse than latinisations like Descartes into Cartesius.

guy@dsv.su.se


David Marjanovic wrote:

>>>
> 17.1. In order to be established, a clade name must be composed of=
=20
> more than one letter and consist exclusively of letters of the Lati=
n=20
> alphabet, which is taken to include j, k, w and y, rare or absent i=
n=20
> classical Latin.
> <<
>
> It is obvious anyway, but maybe z should be included in that list.=
=20
> Like y, it does not occur in any Latin words and was only used to=
=20
> transcribe Greek and other foreign languages (Capital of Dacia:=
=20
> Sarmizegetusa). Or maybe the sentence should end in "consist=20
> exclusively of some or all of the conventional 26 letters of the La=
tin=20
> alphabet" or something.
>
>>>
> Recommendation 17A. Names established under this code should be=
=20
> pronounceable. Thus, every syllable should contain a vowel (or=20
> diphthong), [...]
> <<
>
> The occasional syllabic consonant isn't much of a problem (even tho=
ugh=20
> such things don't exist in Latin or Greek). Many English words end =
in=20
> syllables that have l or n instead of a vowel, and e. g. most kinds=
 of=20
> American English are chock full of syllables with r in the vowel=
=20
> place. What should be limited is consonant clusters. The fossil mam=
mal=20
> *Ausktribosphenos nyktos* (from _Aus_tralia, from the K =3D Cretace=
ous,=20
> and it has tribosphenic teeth) is about the limit of what I find=
=20
> reasonably pronounceable, and I have an Indo-European mother tongue=
,=20
> whereas most other languages don't allow any consonant clusters at =
all!
>

--=20
guy@dsv.su.se
Lundagatan 61, tr III
117 28 Stockholm
Tel:08-658 3775 (home) kl 11.00 -=20


--Boundary_(ID_U4J5arT09WR93qxpiEYhbQ)
Content-type: text/html; charset=3Dus-ascii
Content-transfer-encoding: 7BIT

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
  <meta content=3D"text/html;charset=3DISO-8859-1" http-equiv=3D"Cont=
ent-Type">
  <title></title>
</head>
<body bgcolor=3D"#ffffff" text=3D"#000000">
David M's concerns are fully justified in my opinion.<br>
<br>
Since "Latin alphabet" is open to interpretation and the very purpose
of definitions is to completely exclude ambiguity, would it not be be=
st
to specify explicitly somewhere in the code "The Phylocode Alphabet f=
or
construction of names, consists of the letters: a, b, c, d, e, f, g, =
h,
i, j, k, l, m, n, o, p, q, r, s, t, u, v, w, y, z." and then refer to
this alphabet when needed.<br>
<br>
As for "pronounceable", I would ask pronounceable for whom? Although
English is the lingua franca of the scientific community, pronunciati=
on
of large numbers of ordinary English words remains problematic for ma=
ny
native speakers of other languages. This is a problem that the
inventors of artificial languages that pretend to lingua franca statu=
s,
have already tackled: Ido, Novial, Volap&uuml;k, Interglossa, Esperan=
to and
Interlingua. It might be an idea to consult a scholar of such languag=
es
to find a formulation that defines "pronounceable" in a less ambiguou=
s,
and an <b>unbiased </b>way. <br>
<br>
Clearly there is an appeal to common sense in&nbsp; recommendation 17=
A as it
stands. However what is pronounceable by common sense is clearly
subjective relative to culture and linguistic aptitude. <br>
<br>
I have never got my tongue round Nopalxochia for example, even though=
 I
speak several languages. For many people words that start with s
followed by a consonant are just as tongue twisting as lx is for me.
Excluding x was a good choice since its pronunciation is culturally
ambiguous and phonetically redundant. <br>
<br>
Unfortunately any syntactic rules governing pronunciation will
inevitably require the Phylocodisation of the natural names of
botanists and places when these form the basis of new botanic names,
but this is no worse than latinisations like Descartes into Cartesius=
. <br>
<br>
<a class=3D"moz-txt-link-abbreviated" href=3D"mailto:guy@dsv.su.se">g=
uy@dsv.su.se</a><br>
<br>
<br>
David Marjanovic wrote:<br>
<blockquote cite=3D"mid003001c5c2a7$95f908a0$a83cb23e@IPC" type=3D"ci=
te">
  <blockquote type=3D"cite">
    <blockquote type=3D"cite"><br>
    </blockquote>
  </blockquote>
17.1. In order to be established, a clade name must be composed of mo=
re
than one letter and consist exclusively of letters of the Latin
alphabet, which is taken to include j, k, w and y, rare or absent in
classical Latin.
  <br>
&lt;&lt;
  <br>
  <br>
It is obvious anyway, but maybe z should be included in that list. Li=
ke
y, it does not occur in any Latin words and was only used to transcri=
be
Greek and other foreign languages (Capital of Dacia: Sarmizegetusa). =
Or
maybe the sentence should end in "consist exclusively of some or all =
of
the conventional 26 letters of the Latin alphabet" or something.
  <br>
  <br>
  <blockquote type=3D"cite">
    <blockquote type=3D"cite"><br>
    </blockquote>
  </blockquote>
Recommendation 17A. Names established under this code should be
pronounceable. Thus, every syllable should contain a vowel (or
diphthong), [...]
  <br>
&lt;&lt;
  <br>
  <br>
The occasional syllabic consonant isn't much of a problem (even thoug=
h
such things don't exist in Latin or Greek). Many English words end in
syllables that have l or n instead of a vowel, and e. g. most kinds o=
f
American English are chock full of syllables with r in the vowel plac=
e.
What should be limited is consonant clusters. The fossil mammal
*Ausktribosphenos nyktos* (from _Aus_tralia, from the K =3D Cretaceou=
s,
and it has tribosphenic teeth) is about the limit of what I find
reasonably pronounceable, and I have an Indo-European mother tongue,
whereas most other languages don't allow any consonant clusters at al=
l!
  <br>
  <br>
</blockquote>
<br>
<pre class=3D"moz-signature" cols=3D"72">--=20
<a class=3D"moz-txt-link-abbreviated" href=3D"mailto:guy@dsv.su.se">g=
uy@dsv.su.se</a>
Lundagatan 61, tr III
117 28 Stockholm
Tel:08-658 3775 (home) kl 11.00 -=20
</pre>
</body>
</html>

--Boundary_(ID_U4J5arT09WR93qxpiEYhbQ)--

  

Feedback to <mike@indexdata.com> is welcome!