Nasal assimilation

41 views
Skip to first unread message

cntrational

unread,
Mar 10, 2012, 5:21:14 AM3/10/12
to loj...@googlegroups.com
In most languages, when nasals come before stops, they generally assimilate to the stop. So, we don't say "inpossble", we say "impossble" -- /np/ is a cluster that's rare in most languages; it's usually changed to /mp/. Similarly for /mt/ → /nt/ and /mk/ → /ŋk/. Which makes it really weird that Lojban *does* allow clusters like /np/. What's up with that? Seems like a pretty glaring omission in the Lojban phonotactic rules, especially considering how common nasal assimilation is in the world.

mi'e cntr

Jorge Llambías

unread,
Mar 10, 2012, 6:19:27 AM3/10/12
to loj...@googlegroups.com

In English such assimilations don't really happen (cf. "input",
"gumtree", "camcorder"). "Impossible" came into English with the
assimilation already there. The designers of Lojban were English
speakers, not linguists, so Lojban has a strong bias towards English
phonotactics, not towards what's common in the world.

mu'o mi'e xorxes

Sid

unread,
Mar 10, 2012, 6:36:18 AM3/10/12
to loj...@googlegroups.com
I'd personally say "imput" in rapid speech, and "gumtree" and
"camcorder" are more lenient about phonotactics because they're
compound words. But, yeah, English isn't too strict about it in slower
speech, though I'd argue that assimilation happens a lot in rapid
speech. Ask most English speakers to say "He got ten points." rapidly,
they'll probably say "tem points".

2012/3/10 Jorge Llambías <jjlla...@gmail.com>:

> --
> You received this message because you are subscribed to the Google Groups "lojban" group.
> To post to this group, send email to loj...@googlegroups.com.
> To unsubscribe from this group, send email to lojban+un...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/lojban?hl=en.
>

gleki

unread,
Mar 10, 2012, 6:46:42 AM3/10/12
to loj...@googlegroups.com
If we replace np /nb with mp / mb will there be any clashes in existing gismu (of course taking into account that two gismu can't differ in the last vowel etc.) ?

> To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.

Sid

unread,
Mar 10, 2012, 7:07:22 AM3/10/12
to loj...@googlegroups.com
Actually changing the forms of the gismu to adjust for this seems like
something that will never happen.

>> > lojban+un...@googlegroups.com.


>> > For more options, visit this group at
>> > http://groups.google.com/group/lojban?hl=en.
>> >
>

> --
> You received this message because you are subscribed to the Google Groups
> "lojban" group.

> To view this discussion on the web visit
> https://groups.google.com/d/msg/lojban/-/fmaLB9VGBzYJ.


>
> To post to this group, send email to loj...@googlegroups.com.
> To unsubscribe from this group, send email to

> lojban+un...@googlegroups.com.

Pierre Abbat

unread,
Mar 10, 2012, 8:25:42 AM3/10/12
to loj...@googlegroups.com
On Saturday, March 10, 2012 06:46:42 gleki wrote:
> If we replace np /nb with mp / mb will there be any clashes in existing
> gismu (of course taking into account that two gismu can't differ in the
> last vowel etc.) ?

I don't know offhand, but there are plenty of clashes in lujvo if nasals are
assimilated.

Pierre
--
ve ka'a ro klaji la .romas. se jmaji

gleki

unread,
Mar 10, 2012, 8:56:34 AM3/10/12
to loj...@googlegroups.com
I just meant whether it's important to pronounce m  and n carefully in speech without creating ambiguities

>> > lojban+unsubscribe@googlegroups.com.


>> > For more options, visit this group at
>> > http://groups.google.com/group/lojban?hl=en.
>> >
>
> --
> You received this message because you are subscribed to the Google Groups
> "lojban" group.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/lojban/-/fmaLB9VGBzYJ.
>
> To post to this group, send email to loj...@googlegroups.com.
> To unsubscribe from this group, send email to

> lojban+unsubscribe@googlegroups.com.

Craig Daniel

unread,
Mar 10, 2012, 9:37:09 AM3/10/12
to loj...@googlegroups.com
Personally, to differentiate, my lojban /n/ is usually velar.

(I also used to pronounce my ' as [T], a nonstandard-but-valid value
for it, because I was having trouble making it sufficiently distinct
from x, but my x has improved and I don't need to do that anymore.)

- mi'e .kreig.daniyl.

And Rosta

unread,
Mar 11, 2012, 8:21:37 AM3/11/12
to loj...@googlegroups.com
Jorge Llambías, On 10/03/2012 11:19:

English place assimilation is very active, but (i) it targets /n, d, t/, (ii) it's leftwards, (iii) it's mostly optional (a lazy way of saying that in certain environments it's optional and in certain environments it's obligatory).

It's true that the designers of Lojban knew little linguistics, but quite a few aspects of Lojban phonotactics are weird for English too.

--And.

Escape Landsome

unread,
Mar 11, 2012, 8:33:49 AM3/11/12
to loj...@googlegroups.com
Perhaps, you touched someting.

Another well-known invented language is sanskrit, and it was designed
so as to render this kind of "phonetic accident" impossible.

Sid

unread,
Mar 11, 2012, 10:28:29 AM3/11/12
to loj...@googlegroups.com
Um. 1. Sanskrit was not invented, at least in the way Lojban is. 2.
The habit of Sanskrit speakers of writing assimilations down doesn't
have much to do with this.

mi'e cntr

> --
> You received this message because you are subscribed to the Google Groups "lojban" group.
> To post to this group, send email to loj...@googlegroups.com.

> To unsubscribe from this group, send email to lojban+un...@googlegroups.com.

John E Clifford

unread,
Mar 12, 2012, 10:47:20 AM3/12/12
to loj...@googlegroups.com
Well, Sanskrit is, if not an invented language, at least an artificial one as it has come down to us.  Panini's grammar is more prescriptive than descriptive, at least as it came to be used. As for assimilation, the rules in Panini prescribe what assimilation takes place (all the obvious ones and some rather surprising ones), so that any deviation from this is just bad Sanskrit (Nehru, I think it was, was looked down on by many Brahmins because his Sanskrit had an Uttar Pradeshi accent).  I'm not sure that accidental assimilations in Lojban would be so terrible, but they might in some cases ruin the isomorphism -- better to insert a y or whatever than that.


From: Sid <cntra...@gmail.com>
To: loj...@googlegroups.com
Sent: Sunday, March 11, 2012 9:28 AM
Subject: Re: [lojban] Nasal assimilation

Um. 1. Sanskrit was not invented, at least in the way Lojban is. 2.
The habit of Sanskrit speakers of writing assimilations down doesn't
have much to do with this.

mi'e cntr

On 11 March 2012 18:03, Escape Landsome <esca...@gmail.com> wrote:
> Perhaps, you touched someting.
>
> Another well-known invented language is sanskrit, and it was designed
> so as to render this kind of "phonetic accident" impossible.
>
> --
> You received this message because you are subscribed to the Google Groups "lojban" group.
> To post to this group, send email to loj...@googlegroups.com.
> To unsubscribe from this group, send email to lojban+unsub...@googlegroups.com.

> For more options, visit this group at http://groups.google.com/group/lojban?hl=en.
>

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsub...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages