Brian Stubbs' Semitic/Egyptian > Uto-Aztecan paper

[LingualNoob]

Last year some of you helped me get my feet wet understanding comparative linguistics so that I could try to get a sense about whether or not a preliminary edition of a language proposal showed any obvious technical errors.

I'll say again just like I did last year that I don't pretend that I'm unbiased regarding the implications of the proposal, but that I really just want to understand if it shows promise.

While the book itself is still being prepared for publication, the author published a small introduction to it earlier this year. It can be found here:

http://www.bmaf.org/sites/bmaf.org/files/image/Egyptian-Semitic-in-Uto-Aztecan-by-Brian-Stubbs-Jerry-Grover.pdf

I have no doubt that there will be strong objections to the back-story required for the type of language contact that he's suggesting, but after you're done rolling your eyes and saying "not another one of these", I am hoping that some of you can ignore the back-story for a moment and take a look a the correspondences and cognates that he mentions and let me know if they show any obvious technical errors.

The book is tremendously more detailed than the paper, but I think there's enough in the paper to indicate a few of the basics that he's going to be proposing in the book.


Thanks!

[DKleinecke]

The main problem Stubbs has is that he sounds like a crackpot. The
main immediate objection is that he does not compare Afro-Asiatic with
Uto-Aztecan. The problem of the differances in time and space of the
two families is almost as bad.

I know next to nothing about Uto-Aztecan but what little this article
shows about Stubb's proto-Uto-Aztecan would make me go immediately back
to PUA for a critical examination.

I think what we have here is another example, perhaps the biggest one
yet, of how many pseudo-cognates can be discovered between two unrelated
language groups.

[Peter T. Daniels]

The person asking the question has contacted me privately. Stubbs has a
reason for his investigation. It's up to Mx. Noob to indicate that reason here.

[LingualNoob]

On Sunday, August 9, 2015 at 3:54:17 PM UTC-5, Peter T. Daniels wrote:

> The person asking the question has contacted me privately. Stubbs has a
> reason for his investigation. It's up to Mx. Noob to indicate that reason here.

Take a look at my introduction to this topic and the one from last year (https://groups.google.com/forum/#!searchin/sci.lang/lingualnoob/sci.lang/9uWrRHiLC04/olKfFFDqmP4J).

I have been completely up-front about the fact that both the author and I have biases. In fact, that's the whole reason I sought help from sci.lang in the first place. I've never pretended otherwise in our public or private conversations.

I've never pretended to think that you'd consider the Book of Mormon as a credible source for a back-story, but I have asked you for your technical expertise to help me understand if Stubbs' work is just another piece of pseudo-science from an LDS scholar.

My love of the Book of Mormon and belief in its historocity isn't a product of scientific evaluation, it comes from personal religious experience. I freely admit that my religious biases stand in the way of us seeing eye-to-eye on the back story suggested by Stubbs' research. If that is where this discussion ends with you then I thank you for the technical feedback you've given in our conversations. I honestly do appreciate it.

On the other hand, if you or others are willing to provide feedback about the technical aspects of his proposal then I am all ears.

[LingualNoob]

> The main problem Stubbs has is that he sounds like a crackpot...

Hmm...that's not exactly the type technical feedback I was looking for, but Okay.

> The main immediate objection is that he does not compare Afro-Asiatic with
> Uto-Aztecan.

This is the type of technical feedback I am looking for. How vital do you consider this to be in relation to him being able to convincingly establish the language relationships that he is proposing?

>The problem of the differances in time and space of the
> two families is almost as bad.

Does anybody know what evidence exists regarding Uto-Aztecan that would indicate that the time-depth his proposal implies for Proto-Uto-Aztecan is significantly improbable? Does Stubbs' statement regarding the estimated glottalchronological time-depth make sense?:

"Some may object, citing glottochronology's presumed time-depth of 5,000 years for UA, but holding fast to glottochronological estimates is more a hobby of anthropologists, archaeologists, and non-specialists than of linguists. Most linguists know better and view glottochronological estimates like colds--they usually pass with little permanent damage."

> I know next to nothing about Uto-Aztecan but what little this article
> shows about Stubb's proto-Uto-Aztecan would make me go immediately back
> to PUA for a critical examination.

His book shows that a large number of his *UA cognates come from the work of other Uto-Aztecanists, especially Wick Miller if memory serves.

[Arnaud Fournet]

yes, for example, how can Comanche tobe "lip" derive from supposedly UA *sapal as cited in the pdf??
What's the source for PUA?
A.

[Arnaud Fournet]

Le dimanche 9 août 2015 23:51:15 UTC+2, LingualNoob a écrit :
> > The main problem Stubbs has is that he sounds like a crackpot...
>
> Hmm...that's not exactly the type technical feedback I was looking for, but Okay.

this feedback is nevertheless more technical than you think.
People have given up deriving Arawak from Hebrew about 4 centuries ago.
This is known as "etymological furor".

Before reading 500 pages, a simple preliminary step would be to compare UA with Semitic on the basis of the Swadesh-100 wordlist.
Considering the genetic distance between UA and Semitic, I expect (vague) similarities to fall in the 5-10% range.
A.

[Arnaud Fournet]

Le dimanche 9 août 2015 02:56:02 UTC+2, LingualNoob a écrit :


> The book is tremendously more detailed than the paper, but I think there's enough in the paper to indicate a few of the basics that he's going to be proposing in the book.

For clarification, what exactly is he proposing?

1. that Semitic or Afrasian-looking words can be found in Uto-Aztecan?
2. that Semitic/Afrasian and Uto-Aztecan are close relatives?
3. that Semitic/Afrasian and Uto-Aztecan are distant relatives?

A.

[Yusuf B Gursey]

In <f6736b26-1f4a-4fac...@googlegroups.com>, Arnaud

There seems to be the additional claim that certain specifically NW
Semitic words or forms of words appear in Uto-Aztecan, like Hebrew batt

> A.

[Peter T. Daniels]

On Sunday, August 9, 2015 at 5:51:15 PM UTC-4, LingualNoob wrote:
[no, he didn't]

> > The main problem Stubbs has is that he sounds like a crackpot...
>
> Hmm...that's not exactly the type technical feedback I was looking for, but Okay.
>
> > The main immediate objection is that he does not compare Afro-Asiatic with
> > Uto-Aztecan.
>
> This is the type of technical feedback I am looking for. How vital do you consider this to be in relation to him being able to convincingly establish the language relationships that he is proposing?

I don't know why you chose to reply to my posting as if it were a personal
message to you, but you do yourself and him a disservice by not explaining
the background of the query as you laid it out for me.

[Peter T. Daniels]

Since Mx. Noob appears unwilling to reveal the secret, the answer is (1).

[LingualNoob]

On Monday, August 10, 2015 at 6:15:05 AM UTC-5, Peter T. Daniels wrote:
> On Sunday, August 9, 2015 at 5:51:15 PM UTC-4, LingualNoob wrote:
> [no, he didn't]
>
> > > The main problem Stubbs has is that he sounds like a crackpot...
> >
> > Hmm...that's not exactly the type technical feedback I was looking for, but Okay.
> >
> > > The main immediate objection is that he does not compare Afro-Asiatic with
> > > Uto-Aztecan.
> >
> > This is the type of technical feedback I am looking for. How vital do you consider this to be in relation to him being able to convincingly establish the language relationships that he is proposing?
>
> I don't know why you chose to reply to my posting as if it were a personal

> message to you...

Peter, I'm not clear about what you meant by "a personal message". The quoted text you replied to was from DKleinecke, so I must assume that you are wondering why I sent you a private email yesterday. I sent it because I realized that in our private conversation I might have misrepresented something important from the paper. I had previously told you that Stubbs thought that Uto-Aztecan was a creolization of languages including a lot of Semitic/Egyptian. Yesterday I realized that in the newer paper that we are discussing he might be indicating a genetic relationship between UA and Semitic/Egyptian. I sent you the private message so that you wouldn't make the following mistake:

Mr. Noob is not "unwilling to reveal the secret". He's wondering what the answer is himself. In previous papers Stubbs has proposed that Uto-Aztecan was a creolized language that included heavy Semitic/Egyptian influences. I'm not sure if the new paper is still following that line of thinking when it says:

"In fact, all three of the idioms mentioned (the kw-NWSemitic and p-NWSemitic and Egyptian) appear to have contributed to common UA words found in all branches...It appears that all three were present in what is called Proto-Uto- Aztecan"

So I'm not sure if the answer is (1), with the implication that Proto-Uto-Aztecan is a creolization like he has suggested in the past or if the statement in the new paper is indicating an answer of (2) or (3) depending on what your definition of close/distant is.

In regards to close/distant: In the paper he discusses better preservation of Egyptian in UA than in Coptic and better preservation of Semitic in UA than in Yiddish.

In regards to your statement "you do yourself and him a disservice by not explaining the background of the query as you laid it out for me.", I told you clearly that in the private emails I was "just trying to help paint the picture that he's probably seeing in his mind". I feel it would be a greater disservice to the author to publish my guesses about his thoughts in a public forum.

On the other hand, it's hardly a secret that he's LDS or that his proposal seems to support the back-story of the Book of Mormon and I've already indicated as much here. You seem inclined to want to discuss the details of the Book of Mormon's back-story despite my attempt to keep the discussion focused on the technical details of the proposal. If you want to continue to insinuate that by not engaging in that conversation that I'm not being straightforward then that's up to you. The truth is the same thing that I said in my introduction to this topic:

[bofm...@gmail.com]

Perhaps it is better to ask my underlying question a different way. 11 years ago, someone asked sci.lang what they thought about a paper called "Voiding the Void" that Brian Stubbs had written for LDS audiences (https://groups.google.com/forum/#!searchin/sci.lang/stubbs/sci.lang/eIVVZP_0mQU/waTYHl0rnqgJ). Here is an excerpt from that sci.lang discussion:

On Monday, May 31, 2004 at 5:24:51 PM UTC-5, Peter T. Daniels wrote:
> Rich Wales wrote:
> >
> > "Peter T. Daniels" wrote:
> >
> > > [Brian Stubbs] has not "found" any such roots, . . .
> >
> > Just curious here, but are you saying the above because you have
> > actually read Stubbs's material and concluded it was defective,
> > or because you consider =any= hypothesis of a Semitic/Uto-Aztecan
> > connection to be, by definition, ludicrous and unworthy of any
> > investigation?
>
> The latter.
>
> > In your opinion, what (if any) sort of evidence =would= be needed
> > to establish a pattern of influence upon Uto-Aztecan (or any other
> > indigenous American language family) by speakers of Hebrew (or any
> > other Semitic language or languages)?
>
> Regular correspondences.
>
> Not to mention some explanatory mechanism to account for the purported
> contact.
>
> The Book of Mormon doesn't constitute evidence or a mechanism.
> --
> Peter T. Daniels


I see two fundamental objections in that response:

1. No explanatory mechanism was proposed other than the Book of Mormon.
2. No regular correspondences were proposed.

I know we won't see eye-to-eye on (1), but what about (2)?

Would his paper be very convincing if there was an explanatory mechanism that didn't seem so objectionable?

The answers so far seem to be:

* It would be better if he had included a thorough comparison to Afro-Asiatic.

* His PUA needs to be thoroughly scrutinized.

* He need to account for Camanche "tobe" in order to explain UA *sapal.

* A Swadesh list should be compiled showing > 5%-10% similarities.

Thank you for these helpful answers. What else can I add to this list?

[LingualNoob]

Oops...that reply was done on a different computer where I was logged in differently so it showed a different username. Sorry for the confusion.

[DKleinecke]

I hadn't guessed that there was a LDS connection. That explains why no
historic connection between the locations of the two language groups was
adduced. Since I, like most linguists, cannot accept the Book of Mormon
as historic there is a major problem. It would be best to stay up front
about it.

That Stubbs sounds like a crack-pot is a major issue. Things like his
comments on Yiddish cast a pall over his entire work. If someone can
take such things seriously there is good reason to doubt all of his
work.

The creole explanation for UA is much more plausible (given Stubbs'
assumptions) than a genetic connection.

I am going to get blunt - I suspect Stubbs rejects Afro-Asiatic because
AA includes many black speakers. His use of Egyptian but not Berber is
very strange.

I stand by the last paragraph of my original post - which I copy and
paste here:

[bofm...@gmail.com]

On Monday, August 10, 2015 at 11:48:10 AM UTC-5, DKleinecke wrote:
>
> I hadn't guessed that there was a LDS connection. That explains why no
> historic connection between the locations of the two language groups was
> adduced. Since I, like most linguists, cannot accept the Book of Mormon
> as historic there is a major problem. It would be best to stay up front
> about it.

My apologies then. I thought that mentioning up-front that I had biases and mentioning in the posts that you were a part of from last year that the biases were religious in nature was sufficient, but looking at it now I can see that it was vague enough to miss.

Please take my word for it that it was not my intent to hide that fact. I was just hoping to keep the conversation centered around technical issues. Discussions about the plausibility of the Book of Mormon in general are abundant and hardly limited to linguistic topics. I'm posting on sci.lang hoping to learn from people like yourself whether or not Stubbs' proposal is yet another example of LDS pseudo-scholarship or if it follows the methodologies that historical linguists would expect it to follow.

Despite this misunderstanding and/or disagreement about the relationship between the proposal and the Book of Mormon, several of the responses from yourself and others are shedding light on the topic in ways I hadn't considered yet and I thank you for that.

In regards to some of your other comments:

> That Stubbs sounds like a crack-pot is a major issue. Things like his
> comments on Yiddish cast a pall over his entire work. If someone can
> take such things seriously there is good reason to doubt all of his
> work.
>
> The creole explanation for UA is much more plausible (given Stubbs'
> assumptions) than a genetic connection.

I don't know enough myself to judge the Yiddish comparison. In regards to the creole explanation, he has previously described the connection he is proposing as being a creolization. Since he didn't use that term too explicitly in the most recent paper I don't know if that view has changed.

> I am going to get blunt - I suspect Stubbs rejects Afro-Asiatic because
> AA includes many black speakers. His use of Egyptian but not Berber is
> very strange.

I don't want to jump to the conclusion that you're calling him a racist so I would ask you to clarify this a little. He gives technical arguments relating the proposed cognates to Northwest Semitic and Egyptian. I we are looking for bias then we would be better off recognizing that Semitic/Egyptian tie in to specific references in the Book of Mormon. While AA correlations may provide some support for biases related to the Book of Mormon, the Semitic/Egyptian references provide much more detailed support (The Book of Mormon describes multiple old-world cultures migrating to the Americas. One of them is described as being Hebrew but also being bilingual with Egyptian and it states that the religious records that they brought with them were written in Egyptian...there I go, delving into the back-story). I'm not sure where the color of the speakers came into play here.

Like I said from the beginning, there are biases involved. These biases are relevant to the conversation, but they can also be so distracting that the conversation will likely turn out to be less about the technical details of Stubbs' work and more about the Book of Mormon. I can find conversations related to the historicity/non-historicity of the Book of Mormon anywhere. I'm hoping that sci.lang can provide something more specific.

[Peter T. Daniels]

On Monday, August 10, 2015 at 10:17:29 AM UTC-4, LingualNoob wrote:
> On Monday, August 10, 2015 at 6:15:05 AM UTC-5, Peter T. Daniels wrote:
> > On Sunday, August 9, 2015 at 5:51:15 PM UTC-4, LingualNoob wrote:
> > [no, he didn't]

> > > > The main problem Stubbs has is that he sounds like a crackpot...
> > > Hmm...that's not exactly the type technical feedback I was looking for, but Okay.
> > > > The main immediate objection is that he does not compare Afro-Asiatic with
> > > > Uto-Aztecan.
> > > This is the type of technical feedback I am looking for. How vital do you consider this to be in relation to him being able to convincingly establish the language relationships that he is proposing?
> > I don't know why you chose to reply to my posting as if it were a personal
> > message to you...
>
> Peter, I'm not clear about what you meant by "a personal message". The quoted text you replied to was from DKleinecke, so I must assume that you are wondering why I sent you a private email yesterday. I sent it because I realized that in our private conversation I might have misrepresented something important from the paper. I had previously told you that Stubbs thought that Uto-Aztecan was a creolization of languages including a lot of Semitic/Egyptian. Yesterday I realized that in the newer paper that we are discussing he might be indicating a genetic relationship between UA and Semitic/Egyptian. I sent you the private message so that you wouldn't make the following mistake:
>
> On Monday, August 10, 2015 at 6:17:08 AM UTC-5, Peter T. Daniels wrote:
> > On Monday, August 10, 2015 at 2:15:04 AM UTC-4, Arnaud Fournet wrote:
> > > Le dimanche 9 août 2015 02:56:02 UTC+2, LingualNoob a écrit :
> >
> > > > The book is tremendously more detailed than the paper, but I think there's enough in the paper to indicate a few of the basics that he's going to be proposing in the book.
> > >
> > > For clarification, what exactly is he proposing?
> > >
> > > 1. that Semitic or Afrasian-looking words can be found in Uto-Aztecan?
> > > 2. that Semitic/Afrasian and Uto-Aztecan are close relatives?
> > > 3. that Semitic/Afrasian and Uto-Aztecan are distant relatives?
> >
> > Since Mx. Noob appears unwilling to reveal the secret, the answer is (1).
>
> Mr. Noob is not "unwilling to reveal the secret". He's wondering what the answer is himself. In previous papers Stubbs has proposed that Uto-Aztecan was a creolized language that included heavy Semitic/Egyptian influences. I'm not sure if the new paper is still following that line of thinking when it says:

The "secret" is that Stubbs and Noob are trying to find linguistic justification
for some utterly crazy notions found in the Book of Mormon, notions that were
invented by Joseph Smith on the basis of close to sero knowledge of North
American (pre)history, ethnology, or ethnography. (Or, of course, languages.)

Apparently the myth is that two bands of Hebrew-speakers at different times
crossed the Pacific and interacted with just one ethnic group in what is now
the western US (perhaps in the area Brigham Young coveted as "Deseret") or
Mexico. On the way, one of these bands spent 8 years in Egypt and became
bilingual.

> "In fact, all three of the idioms mentioned (the kw-NWSemitic and p-NWSemitic and Egyptian) appear to have contributed to common UA words found in all branches...It appears that all three were present in what is called Proto-Uto- Aztecan"
>
> So I'm not sure if the answer is (1), with the implication that Proto-Uto-Aztecan is a creolization like he has suggested in the past or if the statement in the new paper is indicating an answer of (2) or (3) depending on what your definition of close/distant is.

As I already told you in email, LOANWOARDS HAVE NOTHING TO DO WITH CREOLIZATION.

> In regards to close/distant: In the paper he discusses better preservation of Egyptian in UA than in Coptic and better preservation of Semitic in UA than in Yiddish.
>
> In regards to your statement "you do yourself and him a disservice by not explaining the background of the query as you laid it out for me.", I told you clearly that in the private emails I was "just trying to help paint the picture that he's probably seeing in his mind". I feel it would be a greater disservice to the author to publish my guesses about his thoughts in a public forum.
>
> On the other hand, it's hardly a secret that he's LDS or that his proposal seems to support the back-story of the Book of Mormon

NO ONE but a Mormon would have the foggiest idea of that.

> and I've already indicated as much here. You seem inclined to want to discuss the details of the Book of Mormon's back-story despite my attempt to keep the discussion focused on the technical details of the proposal. If you want to continue to insinuate that by not engaging in that conversation that I'm not being straightforward then that's up to you. The truth is the same thing that I said in my introduction to this topic:
>
> On Saturday, August 8, 2015 at 7:56:02 PM UTC-5, LingualNoob wrote:
> >
> > I have no doubt that there will be strong objections to the back-story required for the type of language contact that he's suggesting, but after you're done rolling your eyes and saying "not another one of these", I am hoping that some of you can ignore the back-story for a moment and take a look a the correspondences and cognates that he mentions and let me know if they show any obvious technical errors.

You imagine that this paragraph somehow _identifies_ the "back-story"?

[bofm...@gmail.com]

On Monday, August 10, 2015 at 2:27:04 PM UTC-5, Peter T. Daniels wrote:
> The "secret" is that Stubbs and Noob are trying to find linguistic justification
> for some utterly crazy notions found in the Book of Mormon, notions that were
> invented by Joseph Smith on the basis of close to sero knowledge of North
> American (pre)history, ethnology, or ethnography. (Or, of course, languages.)
>
> Apparently the myth is that two bands of Hebrew-speakers at different times
> crossed the Pacific and interacted with just one ethnic group in what is now
> the western US (perhaps in the area Brigham Young coveted as "Deseret") or
> Mexico. On the way, one of these bands spent 8 years in Egypt and became
> bilingual.

^I think this exemplifies my point quite well^

It shows how fast the conversation moves away from the types of technical questions that sci.lang should be adept at answering and focuses it on individuals' perceptions of Joseph Smith, Brigham Young, "Deseret", etc..

It also shows that Peter has already forgotten one thing that I was painstakingly clear about in our private conversation: That although I have my own theories regarding geographic correlations between the southwest and the Book of Mormon, Stubbs does not endorse them nor (to my knowledge) does he endorse other geographic Book of Mormon correlations proposed by other LDS scholars.

I told you that even within the LDS community my ideas were on the outskirts of geographic/ethnographic thinking and I explicitly told you that Stubbs did not endorse them. Here is exactly what I said before discussing Book of Mormon geography:

"I do want to say for the record though that Stubbs does not claim to prefer any Book of Mormon geography (including mine). He stays out of the geographical debate entirely."

In our private conversation you brought up many good questions related to religious aspects of Mormonism that you disagree with. I appreciated your comments and respect your views on those subjects and I enjoyed that exchange. In this public conversation I would prefer that both you and I at least attempt to set aside biases related to our differing views of Mormonism in general and utilize your expertise in linguistics to share with me whatever you are able/willing to share regarding the technical aspects of the proposal that we're discussing.

In regards to the following statement:

> As I already told you in email, LOANWOARDS HAVE NOTHING TO DO WITH CREOLIZATION.

To be fair, your discussion of loanwords was very helpful, but you were not that explicit in saying that it had nothing to do with creolization. My "noobness" is probably what prevented me from formulating your private response into a statement as clear as the one you're giving now. My apologies for the confusion.

[António Marques]

<bofm...@gmail.com> wrote:
>
> Like I said from the beginning, there are biases involved. These biases
> are relevant to the conversation, but they can also be so distracting
> that the conversation will likely turn out to be less about the technical
> details of Stubbs' work and more about the Book of Mormon. I can find
> conversations related to the historicity/non-historicity of the Book of
> Mormon anywhere. I'm hoping that sci.lang can provide something more specific.

The issue here is that it's not at all hard to find chance resemblances
between unrelated languages, all the more once the semantics is allowed to
be loose (which it sort of has to be because we know from doubtlessly
related languages that it is). As such, it's exceedingly important to know
what, precisely, is being proposed, to be able to validate anything: given
that the data is flexible, the net cant be cast wide, otherwise one can
make everything up as one goes along. In that regard, the questions about
AA and PUA are telling: whether going to the proto-languages or not will
entirely depend on what hypothesis one is trying to prove (and so we come
to what you call the backstory, sort of).

As to the hypothesis of there being an American family of languages derived
from a creole that got its lexicon from Hebrew and Egyptian, the
difficulties begin with Hebrew and Egyptian, the vocalism of which is far
from being certain. Then there's the issue of dates. But whether the work
is a sincere scientific effort (no matter its backstory) or just
pseudo-science, is something that only someone who looks into it minutely
and has a reasonable knowledge of UA and AA (and are there many?) can say
authoritatively. We may look cursorily into it and think it looks like
crackpottery based on some first sight metrics, but that's it. There's a
lot of folks here who know a lot about AA, but I don't think we have any
resident UA expert.
--
Sent from one of my newsreaders

[LingualNoob]

On Monday, August 10, 2015 at 5:59:24 PM UTC-5, António Marques wrote:
>
> The issue here is that it's not at all hard to find chance resemblances
> between unrelated languages, all the more once the semantics is allowed to
> be loose (which it sort of has to be because we know from doubtlessly
> related languages that it is).

This is one of the subjects I was hoping to get some opinions on. There is obviously some semantic leeway in the paper, but there are a lot of examples that show very little leeway at all. Example:

Hebrew ya'amiin-o 'he believes him/it' > UA *yawamin-(o) 'believe (him/it)'

There are other examples that seem to me to show a questionable amount of leeway. Example:

Aramaic *yagar 'hill, heap of stones' > UA *yakaC / *yakaR (AMR) 'nose, point, ridge'

You say that semantic leeway has to be loose, but can you give me an idea of how his semantic leeway stacks up compared to other proposals that you've seen?

> As such, it's exceedingly important to know
> what, precisely, is being proposed, to be able to validate anything: given
> that the data is flexible, the net cant be cast wide, otherwise one can
> make everything up as one goes along. In that regard, the questions about
> AA and PUA are telling: whether going to the proto-languages or not will
> entirely depend on what hypothesis one is trying to prove (and so we come
> to what you call the backstory, sort of).

I really don't disagree with your point here, but it's such a slippery slope. People just don't usually seem to have much capacity to talk about anything else once we go down that path.

> As to the hypothesis of there being an American family of languages derived
> from a creole that got its lexicon from Hebrew and Egyptian, the
> difficulties begin with Hebrew and Egyptian, the vocalism of which is far
> from being certain.

I don't know much about these difficulties other than the fact that both you and Stubbs say its difficult. Whatever the difficulties are, does it preclude the possibility of doing a reliable comparison?

> Then there's the issue of dates. But whether the work
> is a sincere scientific effort (no matter its backstory) or just
> pseudo-science, is something that only someone who looks into it minutely
> and has a reasonable knowledge of UA and AA (and are there many?) can say
> authoritatively. We may look cursorily into it and think it looks like
> crackpottery based on some first sight metrics, but that's it. There's a
> lot of folks here who know a lot about AA, but I don't think we have any
> resident UA expert.

Would the technical issues discussed so far rise to the level of "crackpottery" if it weren't for the back-story?

[Peter T. Daniels]

On Monday, August 10, 2015 at 7:55:46 PM UTC-4, LingualNoob wrote:

> Hebrew ya'amiin-o 'he believes him/it' > UA *yawamin-(o) 'believe (him/it)'

What are the individual *UA morphemes?

Hebrew:
ya- 3sg.subj.
'MN 'believe'
-a-ii- imperfect ("present")
-o 3sg.obj.

If the *UA word breaks down similarly, you might have an example.

> There are other examples that seem to me to show a questionable amount of leeway. Example:
>
> Aramaic *yagar 'hill, heap of stones' > UA *yakaC / *yakaR (AMR) 'nose, point, ridge'
>
> You say that semantic leeway has to be loose, but can you give me an idea of how his semantic leeway stacks up compared to other proposals that you've seen?

Carl Darling Buck's *Dictionary of Selected Synonyms ...* (1949)

http://press.uchicago.edu/ucp/books/book/chicago/D/bo3630267.html

will show you "semantic leeway." It is not an etymological dictionary of
Indo-European, and it predates Swadesh lists, so it's sort of a check on
them.

[bofm...@gmail.com]

On Monday, August 10, 2015 at 9:33:21 PM UTC-5, Peter T. Daniels wrote:
> On Monday, August 10, 2015 at 7:55:46 PM UTC-4, LingualNoob wrote:
>
> > Hebrew ya'amiin-o 'he believes him/it' > UA *yawamin-(o) 'believe (him/it)'
>
> What are the individual *UA morphemes?
>
> Hebrew:
> ya- 3sg.subj.
> 'MN 'believe'
> -a-ii- imperfect ("present")
> -o 3sg.obj.
>
> If the *UA word breaks down similarly, you might have an example.

I appreciate you providing the Hebrew breakdown. I spent a couple of hours last night trying to find a source that shows the breakdown in *UA. It turns out that the internet doesn't know everything yet.

I'll head over to the university library this afternoon to see what Wick Miller's database has to say about it unless someone else has it available and can save me the trip.

>
> > There are other examples that seem to me to show a questionable amount of leeway. Example:
> >
> > Aramaic *yagar 'hill, heap of stones' > UA *yakaC / *yakaR (AMR) 'nose, point, ridge'
> >
> > You say that semantic leeway has to be loose, but can you give me an idea of how his semantic leeway stacks up compared to other proposals that you've seen?
>
> Carl Darling Buck's *Dictionary of Selected Synonyms ...* (1949)
>
> http://press.uchicago.edu/ucp/books/book/chicago/D/bo3630267.html
>
> will show you "semantic leeway." It is not an etymological dictionary of
> Indo-European, and it predates Swadesh lists, so it's sort of a check on
> them.

Thank you for the reference.

[Peter T. Daniels]

On Tuesday, August 11, 2015 at 8:58:27 AM UTC-4, bofm...@gmail.com wrote:
> On Monday, August 10, 2015 at 9:33:21 PM UTC-5, Peter T. Daniels wrote:
> > On Monday, August 10, 2015 at 7:55:46 PM UTC-4, LingualNoob wrote:
> >
> > > Hebrew ya'amiin-o 'he believes him/it' > UA *yawamin-(o) 'believe (him/it)'
> >
> > What are the individual *UA morphemes?
> >
> > Hebrew:
> > ya- 3sg.subj.

oops--3masc.sg.subj.

> > 'MN 'believe'
> > -a-ii- imperfect ("present")
> > -o 3sg.obj.

3masc.sg.obj.

> > If the *UA word breaks down similarly, you might have an example.
>
> I appreciate you providing the Hebrew breakdown. I spent a couple of hours last night trying to find a source that shows the breakdown in *UA. It turns out that the internet doesn't know everything yet.
>
> I'll head over to the university library this afternoon to see what Wick Miller's database has to say about it unless someone else has it available and can save me the trip.

A convenient, pretty up-to-date reference on Native American languages is
the linguistics volume of the Smithsonian Handbook:

https://en.wikipedia.org/wiki/Handbook_of_North_American_Indians#Volume_17:_Languages

(who'da thunk it would have a wikiparticle! I didn't know bush43 had cut off
the funding so close to completion. Each of the areal volumes has at least
one chapter on the local languages as well, and over the years I collected
copies of all of them. The missing vols. 1 and 16 were to be "Introduction" [contents not specified] and "Technology and Visual Arts" [presumably lots of the color plates not used in the other volumes]; 18-19 were to be Biographical Dictionary, and 20 Index.)

When I bought my v.17 at the old Smithsonian Museum of the American Indian
in the old Customs House in NYC (before most of the collection went to its
new home in DC), I think it was $60 -- in those days government publications
were sold at cost -- but that may reflect the member's discount of whomever
I had happened to visit it with.

[bofm...@gmail.com]

On Tuesday, August 11, 2015 at 2:05:06 PM UTC-5, Peter T. Daniels wrote:
> > > ...Hebrew:

> > > ya- 3sg.subj.
>
> oops--3masc.sg.subj.
>
> > > 'MN 'believe'
> > > -a-ii- imperfect ("present")
> > > -o 3sg.obj.
>
> 3masc.sg.obj.

Thank you for the clarifications. In regards to "ya-", in the paper I believe Stubbs is telling us that he sees it as fossilized. Here's the quote:

"Some characteristics of UA are different or not at all like Egyptian or Semitic, but reflect influences rather typical of Amerindian language families, which we would expect of a transplant from the outside into the Americas. One example is suppletion in singular vs. plural verb forms. That is, one verb is used for singular subjects and an entirely different word is used when the subject is plural, yet suppletion is nearly non-existent in Semitic or Egyptian. A score of such pairs in UA show such influences on UA. Semitic conjugation morphology (patterns of how verbs are conjugated) is not productive in UA, but hundreds of fossilized forms of both the suffixed / perfective conjugation (singular yašiba; plural yašib-uu) and the prefixed / imperfective conjugation (yi-/ya-, ti-/ta-, etc) are found in UA." p2

I'm not seeing a problem fitting 'MN 'believe'. It seems pretty straightforward to me, but I could be missing something. Please let me know if I am.

I did manage to make it to the university during lunch today. Wick Miller's database #ya-27 lists two attested versions:

Gabrielino: yamáyno "believe it!"
Seranno: yawamin "to believe"

I also grabbed a few other Uto-Aztecan books from the same shelf. I glossed through one of them and found the word described as follows:

Yawii-n-ok=rey
believe-n?-nfu=1pII
'Te creo.'

The book uses the word as an example of how the 'ii' is a troublesome enclitic in Gabrielino. The "=rey" seen above is a reference to a particular version of the enclitic. I'll see if I can find some time tonight to read up on it some more.

Thank you for the handbook reference.

[LingualNoob]

On Tuesday, August 11, 2015 at 4:12:05 PM UTC-5, bofm...@gmail.com wrote:

> The book uses the word as an example of how the 'ii' is a troublesome enclitic in Gabrielino. The "=rey" seen above is a reference to a particular version of the enclitic. I'll see if I can find some time tonight to read up on it some more.

Strike that last part. I misread it earlier.

I can't find anything that describes the morphemes involved in UA *yawamin-(o) 'believe (him/it)'

[Peter T. Daniels]

On Tuesday, August 11, 2015 at 5:12:05 PM UTC-4, bofm...@gmail.com wrote:
> On Tuesday, August 11, 2015 at 2:05:06 PM UTC-5, Peter T. Daniels wrote:
> > > > ...Hebrew:
> > > > ya- 3sg.subj.
> >
> > oops--3masc.sg.subj.
> >
> > > > 'MN 'believe'
> > > > -a-ii- imperfect ("present")
> > > > -o 3sg.obj.
> >
> > 3masc.sg.obj.
>
> Thank you for the clarifications. In regards to "ya-", in the paper I believe Stubbs is telling us that he sees it as fossilized. Here's the quote:
>
> "Some characteristics of UA are different or not at all like Egyptian or Semitic, but reflect influences rather typical of Amerindian language families, which we would expect of a transplant from the outside into the Americas. One example is suppletion in singular vs. plural verb forms. That is, one verb is used for singular subjects and an entirely different word is used when the subject is plural, yet suppletion is nearly non-existent in Semitic or Egyptian. A score of such pairs in UA show such influences on UA. Semitic conjugation morphology (patterns of how verbs are conjugated) is not productive in UA, but hundreds of fossilized forms of both the suffixed / perfective conjugation (singular yašiba; plural yašib-uu) and the prefixed / imperfective conjugation (yi-/ya-, ti-/ta-, etc) are found in UA." p2
>
> I'm not seeing a problem fitting 'MN 'believe'. It seems pretty straightforward to me, but I could be missing something. Please let me know if I am.
>
> I did manage to make it to the university during lunch today. Wick Miller's database #ya-27 lists two attested versions:
>
> Gabrielino: yamáyno "believe it!"
> Seranno: yawamin "to believe"
>
> I also grabbed a few other Uto-Aztecan books from the same shelf. I glossed through one of them and found the word described as follows:
>
> Yawii-n-ok=rey
> believe-n?-nfu=1pII
> 'Te creo.'

I.e. there's no connection whatsoever between the two forms.

[Yusuf B Gursey]

In <2e28def4-0d44-4060...@googlegroups.com>,

bofm...@gmail.com wrote on 8/12/2015:
> On Tuesday, August 11, 2015 at 2:05:06 PM UTC-5, Peter T. Daniels wrote:
>>>> ...Hebrew:
>>>> ya- 3sg.subj.
>>
>> oops--3masc.sg.subj.
>>
>>>> 'MN 'believe'
>>>> -a-ii- imperfect ("present")
>>>> -o 3sg.obj.
>>
>> 3masc.sg.obj.
>
> Thank you for the clarifications. In regards to "ya-", in the paper I
> believe Stubbs is telling us that he sees it as fossilized. Here's the
> quote:
>
> "Some characteristics of UA are different or not at all like Egyptian or

Which is stating what is obvious from what one would expect.

> Semitic, but reflect influences rather typical of Amerindian language

Which is trying to save face.

[LingualNoob]

On Tuesday, August 11, 2015 at 10:12:24 PM UTC-5, Peter T. Daniels wrote:
> ...I.e. there's no connection whatsoever between the two forms.

Does this mean:

1. You see linguistic evidence against the idea that Semitic conjugation could be fossilized in UA *yawamin-(o)?

OR

2. You see no linguistic evidence supporting the idea that Semitic conjugation is fossilized in UA *yawamin-(o)?

If your answer to the above is (2), would the possibility of fossilized Semitic conjugation be more persuasive:

2a. If a larger quantity examples were provided?

2b. If examples of better quality were provided? Can you describe what you would expect to see in a better quality example?

[Yusuf B Gursey]

In <3e3277ba-d0d1-41a3...@googlegroups.com>, LingualNoob

wrote on 8/13/2015:
> On Tuesday, August 11, 2015 at 10:12:24 PM UTC-5, Peter T. Daniels wrote:
>> ...I.e. there's no connection whatsoever between the two forms.
>
> Does this mean:
>
> 1. You see linguistic evidence against the idea that Semitic conjugation
> could be fossilized in UA *yawamin-(o)?

This one. UA conjugation is morphologically distinct from Semitic
conjugation, which is no surprise.

[LingualNoob]

On Wednesday, August 12, 2015 at 11:21:50 PM UTC-5, Yusuf B Gursey wrote:
> > 1. You see linguistic evidence against the idea that Semitic conjugation
> > could be fossilized in UA *yawamin-(o)?
>
> This one. UA conjugation is morphologically distinct from Semitic
> conjugation, which is no surprise.
>

The thing that I'm trying to understand is not whether or not they are morphologically distinct. Stubbs said as much himself and you answered that he is "trying to save face". It's the next part of his statement that I'm referring to. He says that "the prefixed / imperfective conjugation (yi-/ya-, ti-/ta-, etc) are found fossilized in many of his proposed cognate sets". Is there linguistic evidence against the idea that the ya- in UA *yawamin-(o)? could be the result of fossilization of Hebrew "ya- 3masc.sg.subj."?

[Yusuf B Gursey]

In <4951eb79-849a-45fe...@googlegroups.com>, LingualNoob

wrote on 8/13/2015:
> On Wednesday, August 12, 2015 at 11:21:50 PM UTC-5, Yusuf B Gursey wrote:
>>> 1. You see linguistic evidence against the idea that Semitic conjugation
>>> could be fossilized in UA *yawamin-(o)?
>>
>> This one. UA conjugation is morphologically distinct from Semitic
>> conjugation, which is no surprise.
>>
>
> The thing that I'm trying to understand is not whether or not they are
> morphologically distinct. Stubbs said as much himself and you answered that
> he is "trying to save face". It's the next part of his statement that I'm

I was refering to the part of the statement that tried to diminish the
importance of morphological distinctiveness.

[Peter T. Daniels]

On Wednesday, August 12, 2015 at 11:36:41 PM UTC-4, LingualNoob wrote:
> On Tuesday, August 11, 2015 at 10:12:24 PM UTC-5, Peter T. Daniels wrote:

> > ...I.e. there's no connection whatsoever between the two forms.
>
> Does this mean:
>
> 1. You see linguistic evidence against the idea that Semitic conjugation could be fossilized in UA *yawamin-(o)?

You just said -- in the evidence you deleted above my conclusion there --
that the morphology of *UA is nothing like the morphology of Hebrew. What
is your Just-So Story to explain that of the tens of thousands of verb
forms in Hebrew, the _one_ that would have been borrowed intact eeans
'he believes him'? And the further coincidence that the first two syllables,
communicating one morpheme and parts of two others, just happen to resemble
a single intact morpheme?

> OR
>
> 2. You see no linguistic evidence supporting the idea that Semitic conjugation is fossilized in UA *yawamin-(o)?

That too.

> If your answer to the above is (2), would the possibility of fossilized Semitic conjugation be more persuasive:
>
> 2a. If a larger quantity examples were provided?
>
> 2b. If examples of better quality were provided? Can you describe what you would expect to see in a better quality example?

Do you have any examples from the borrowing of words anywhere in the world
that is anything like this unique example?

[LingualNoob]

On Thursday, August 13, 2015 at 6:28:15 AM UTC-5, Peter T. Daniels wrote:
> On Wednesday, August 12, 2015 at 11:36:41 PM UTC-4, LingualNoob wrote:
> > On Tuesday, August 11, 2015 at 10:12:24 PM UTC-5, Peter T. Daniels wrote:
>
> > > ...I.e. there's no connection whatsoever between the two forms.
> >
> > Does this mean:
> >
> > 1. You see linguistic evidence against the idea that Semitic conjugation could be fossilized in UA *yawamin-(o)?
>
> You just said -- in the evidence you deleted above my conclusion there --

My apologies. I was just deleting to keep the post to a smaller size. I wasn't thinking that you or others would suddenly forget the previous post. I'll avoid doing that in the future.

> that the morphology of *UA is nothing like the morphology of Hebrew. What
> is your Just-So Story to explain that of the tens of thousands of verb
> forms in Hebrew, the _one_ that would have been borrowed intact eeans

> 'he believes him'?...

I never said that there was only _one_ intact Hebrew verb form. In the paper, Stubbs tells us that "hundreds of fossilized forms of both the suffixed /perfective conjugation (singular yašiba; plural yašib-uu)and the prefixed / imperfective conjugation (yi-/ya-, ti-/ta-,etc) are found in UA".

He gives several examples on pages 11-12. Here's one:

Uto-Aztecan has four separate forms from the verb bky /bakaa ‘to cry, weep’:
(559) p-Semitic bky/ bakaa ‘he cried, wept’; Syriac bakaa / baka’ > UA *paka’ ‘cry’
(24) kw-Semitic bky/ bakaa ‘he cried, wept’; Hebrew baakaa > UA *kwïkï / *o’kï 'cry’
Because bilabials as first segment in a cluster disappear (-bk- > -k-) in Egyptian/Semitic > UA, the imperfective 3rd person masculine singular *ya-bkV ‘he/it weeps’ with imperfective prefix originally *ya- (later yi-) also matches UA *yakka
(560) Semitic *ya-bkay ‘he/it weeps, cries, masc sg.’ > UA *yaCkaC > *yakka / *yaka ‘cry’
(561) Semitic *ta-bkay ‘she/it weeps, cries, fem sg.’ > UA *takka > NP taka ‘cry’.
So Northern Paiute has both the masc 3rd sg of *ya-bka > yakka and the fem 3rd singular *ta-bka > UA *takka ‘cry’ (the middle consonant geminates/doubles in both as well). UA also has the perfective stem in Aramaic bakay / baka’ ‘cry’ > UA *paka’ of the p-NWSemitic and also *kwïkï/*o’kï of the kw-NWSemitic.

> ...And the further coincidence that the first two syllables,

> communicating one morpheme and parts of two others, just happen to resemble
> a single intact morpheme?

See comments about agglutinated morphemes in creole languages below.

>
> > OR
> >
> > 2. You see no linguistic evidence supporting the idea that Semitic conjugation is fossilized in UA *yawamin-(o)?
>
> That too.
>
> > If your answer to the above is (2), would the possibility of fossilized Semitic conjugation be more persuasive:
> >
> > 2a. If a larger quantity examples were provided?
> >
> > 2b. If examples of better quality were provided? Can you describe what you would expect to see in a better quality example?
>
> Do you have any examples from the borrowing of words anywhere in the world
> that is anything like this unique example?

While Googling for an example like you are asking for I came across the book "An Introduction to Pidgins and Creoles" by John Holm. Since we are discussing a possible creolization of Semitic/Egyptian into UA, the examples and discussion in sections 4.5 and 6.4.2 of the book seem relevant.

It section 4.5 (p127) it says:

"...the extent to which [external lexical influences] are evidenced in creoles suggests that they are accelerated by restructuring...Pidgins and creoles are sometimes claimed to be languages without any inflectional morphology whatsoever...this seems to be true of most fully restructured varieties that are not decreolizing...European morpheme boundaries also disappeared in the creoles: in Creole English one can speak of one aunts or one tools, in which the English word and its plural inflection have become a single creole morpheme with either singular or plural meaning"

To substantiate this particular point the author refers to section (6.4.2) of the book to clarify his point about the disappearance of morpheme boundaries in creole languages. This other section says (p215):

"Unlike nouns in their European lexical source languages, creole nouns are not inflected to indicate number, e.g. CE 'aal di animal__' 'all the animals'. Although some creole words contain fossilized remnants of plural inflections from their lexical source languages (e.g. CE 'tuulz' 'tool' or CF 'zanj' 'angel' from F les anges), these no longer function as inflections..."

It seems appropriate to ask how common it is to find agglutinated morphemes in creole languages. Back to section 4.5 (p129) it says:

"...Baker (1984) surveyed CF lexicons for count nouns having an initial syllable wholly derived from a French article; he found 112 in Haitian, 337 in Rodrigues, 444 in Seychellois and 471 in Mauritian - but only 12 in Réunionnais, providing further evidence that the last was not as extensively restructured. Such morpheme boundary reanalyses are much less frequent in creoles of other lexical bases, but they do occur"

Baker's survey was only looking for count nouns, so it doesn't tell us specifically how likely it would be for ya- or other Semitic conjugations to show up in UA *yawamin-(o)?. On the other hand, it does show that it is common to find agglutinated morphemes in creole languages.

Perhaps the following example from the book fits your request for an example?:

"English creole verbs sometimes agglutinated the -ing ending (e.g. Bahamian to courtin, to loadin, to fishin) and could then take the mesolectal progressive marker -in, e.g. go fishinin." (p129)

The book provides a good discussion of the subject and gives several other examples as well. Here's the Google book preview: https://books.google.com/books?id=B7Nko5XBOegC&lpg=PA215&ots=0BChlrsIU5&dq=creole%20fossilized%20morpheme&pg=PA127#v=onepage&q&f=false

[Arnaud Fournet]

Le jeudi 13 août 2015 19:07:15 UTC+2, LingualNoob a écrit :
> On Thursday, August 13, 2015 at 6:28:15 AM UTC-5, Peter T. Daniels wrote:
> > On Wednesday, August 12, 2015 at 11:36:41 PM UTC-4, LingualNoob wrote:
> > > On Tuesday, August 11, 2015 at 10:12:24 PM UTC-5, Peter T. Daniels wrote:
> >
> > > > ...I.e. there's no connection whatsoever between the two forms.
> > >
> > > Does this mean:
> > >
> > > 1. You see linguistic evidence against the idea that Semitic conjugation could be fossilized in UA *yawamin-(o)?
> >
> > You just said -- in the evidence you deleted above my conclusion there --
>
> My apologies. I was just deleting to keep the post to a smaller size. I wasn't thinking that you or others would suddenly forget the previous post. I'll avoid doing that in the future.
>
> > that the morphology of *UA is nothing like the morphology of Hebrew. What
> > is your Just-So Story to explain that of the tens of thousands of verb
> > forms in Hebrew, the _one_ that would have been borrowed intact eeans
> > 'he believes him'?...
>
> I never said that there was only _one_ intact Hebrew verb form. In the paper, Stubbs tells us that "hundreds of fossilized forms of both the suffixed /perfective conjugation (singular yašiba; plural yašib-uu)and the prefixed / imperfective conjugation (yi-/ya-, ti-/ta-,etc) are found in UA".
>

As a general rule, most loanwords from a language to another are nouns.
Verbs are generally not borrowed, and verbal morphology even less. When verbs are borrowed, they are often no longer a verb in the target language.
so such a claim runs against known examples.
A.

> He gives several examples on pages 11-12. Here's one:
>
> Uto-Aztecan has four separate forms from the verb bky /bakaa ‘to cry, weep’:
> (559) p-Semitic bky/ bakaa ‘he cried, wept’; Syriac bakaa / baka’ > UA *paka’ ‘cry’
> (24) kw-Semitic bky/ bakaa ‘he cried, wept’; Hebrew baakaa > UA *kwïkï / *o’kï 'cry’
> Because bilabials as first segment in a cluster disappear (-bk- > -k-) in Egyptian/Semitic > UA, the imperfective 3rd person masculine singular *ya-bkV ‘he/it weeps’ with imperfective prefix originally *ya- (later yi-) also matches UA *yakka
> (560) Semitic *ya-bkay ‘he/it weeps, cries, masc sg.’ > UA *yaCkaC > *yakka / *yaka ‘cry’
> (561) Semitic *ta-bkay ‘she/it weeps, cries, fem sg.’ > UA *takka > NP taka ‘cry’.
> So Northern Paiute has both the masc 3rd sg of *ya-bka > yakka and the fem 3rd singular *ta-bka > UA *takka ‘cry’ (the middle consonant geminates/doubles in both as well). UA also has the perfective stem in Aramaic bakay / baka’ ‘cry’ > UA *paka’ of the p-NWSemitic and also *kwïkï/*o’kï of the kw-NWSemitic.

I'm afraid the rich prefixal morphology of Semitic verbs offers a large playground for completely fanciful comparisons.
A.

[LingualNoob]

On Thursday, August 13, 2015 at 1:28:55 PM UTC-5, Arnaud Fournet wrote:
> Le jeudi 13 août 2015 19:07:15 UTC+2, LingualNoob a écrit :
> > On Thursday, August 13, 2015 at 6:28:15 AM UTC-5, Peter T. Daniels wrote:
> > > On Wednesday, August 12, 2015 at 11:36:41 PM UTC-4, LingualNoob wrote:
> > > > On Tuesday, August 11, 2015 at 10:12:24 PM UTC-5, Peter T. Daniels wrote:
> > >
> > > > > ...I.e. there's no connection whatsoever between the two forms.
> > > >
> > > > Does this mean:
> > > >
> > > > 1. You see linguistic evidence against the idea that Semitic conjugation could be fossilized in UA *yawamin-(o)?
> > >
> > > You just said -- in the evidence you deleted above my conclusion there --
> >
> > My apologies. I was just deleting to keep the post to a smaller size. I wasn't thinking that you or others would suddenly forget the previous post. I'll avoid doing that in the future.
> >
> > > that the morphology of *UA is nothing like the morphology of Hebrew. What
> > > is your Just-So Story to explain that of the tens of thousands of verb
> > > forms in Hebrew, the _one_ that would have been borrowed intact eeans
> > > 'he believes him'?...
> >
> > I never said that there was only _one_ intact Hebrew verb form. In the paper, Stubbs tells us that "hundreds of fossilized forms of both the suffixed /perfective conjugation (singular yašiba; plural yašib-uu)and the prefixed / imperfective conjugation (yi-/ya-, ti-/ta-,etc) are found in UA".
> >
>
> As a general rule, most loanwords from a language to another are nouns.
> Verbs are generally not borrowed, and verbal morphology even less. When verbs are borrowed, they are often no longer a verb in the target language.
> so such a claim runs against known examples.
> A.
>

In response to one of my earlier posts Peter made it clear that "loanwords have nothing to do with creolization".

>
> As I already told you in email, LOANWOARDS HAVE NOTHING TO DO WITH CREOLIZATION.
>



>

[Arnaud Fournet]

Le jeudi 13 août 2015 21:08:20 UTC+2, LingualNoob a écrit :
> On Thursday, August 13, 2015 at 1:28:55 PM UTC-5, Arnaud Fournet wrote:
> > Le jeudi 13 août 2015 19:07:15 UTC+2, LingualNoob a écrit :
> > > On Thursday, August 13, 2015 at 6:28:15 AM UTC-5, Peter T. Daniels wrote:
> > > > On Wednesday, August 12, 2015 at 11:36:41 PM UTC-4, LingualNoob wrote:
> > > > > On Tuesday, August 11, 2015 at 10:12:24 PM UTC-5, Peter T. Daniels wrote:
> > > >
> > > > > > ...I.e. there's no connection whatsoever between the two forms.
> > > > >
> > > > > Does this mean:
> > > > >
> > > > > 1. You see linguistic evidence against the idea that Semitic conjugation could be fossilized in UA *yawamin-(o)?
> > > >
> > > > You just said -- in the evidence you deleted above my conclusion there --
> > >
> > > My apologies. I was just deleting to keep the post to a smaller size. I wasn't thinking that you or others would suddenly forget the previous post. I'll avoid doing that in the future.
> > >
> > > > that the morphology of *UA is nothing like the morphology of Hebrew. What
> > > > is your Just-So Story to explain that of the tens of thousands of verb
> > > > forms in Hebrew, the _one_ that would have been borrowed intact eeans
> > > > 'he believes him'?...
> > >
> > > I never said that there was only _one_ intact Hebrew verb form. In the paper, Stubbs tells us that "hundreds of fossilized forms of both the suffixed /perfective conjugation (singular yašiba; plural yašib-uu)and the prefixed / imperfective conjugation (yi-/ya-, ti-/ta-,etc) are found in UA".
> > >
> >
> > As a general rule, most loanwords from a language to another are nouns.
> > Verbs are generally not borrowed, and verbal morphology even less. When verbs are borrowed, they are often no longer a verb in the target language.
> > so such a claim runs against known examples.
> > A.
> >
>
> In response to one of my earlier posts Peter made it clear that "loanwords have nothing to do with creolization".

yes, but your "theory" has nothing to do with creolization.
I'm afraid you use the word "creolization" as a kind of wild joker card, in order to "explain" everything you feel like.
Your claim is that semitic-speakers entered in contact with UA speakers at some point in the past. So the issue is about borrowings resulting from these hypothetical contacts.
A.

[Peter T. Daniels]

On Thursday, August 13, 2015 at 3:35:04 PM UTC-4, Arnaud Fournet wrote:
> Le jeudi 13 août 2015 21:08:20 UTC+2, LingualNoob a écrit :
> > On Thursday, August 13, 2015 at 1:28:55 PM UTC-5, Arnaud Fournet wrote:
> > > Le jeudi 13 août 2015 19:07:15 UTC+2, LingualNoob a écrit :
> > > > On Thursday, August 13, 2015 at 6:28:15 AM UTC-5, Peter T. Daniels wrote:
> > > > > On Wednesday, August 12, 2015 at 11:36:41 PM UTC-4, LingualNoob wrote:
> > > > > > On Tuesday, August 11, 2015 at 10:12:24 PM UTC-5, Peter T. Daniels wrote:

> > > > > > > ...I.e. there's no connection whatsoever between the two forms.
> > > > > > Does this mean:
> > > > > > 1. You see linguistic evidence against the idea that Semitic conjugation could be fossilized in UA *yawamin-(o)?
> > > > > You just said -- in the evidence you deleted above my conclusion there --
> > > > My apologies. I was just deleting to keep the post to a smaller size. I wasn't thinking that you or others would suddenly forget the previous post. I'll avoid doing that in the future.
> > > > > that the morphology of *UA is nothing like the morphology of Hebrew. What
> > > > > is your Just-So Story to explain that of the tens of thousands of verb
> > > > > forms in Hebrew, the _one_ that would have been borrowed intact eeans
> > > > > 'he believes him'?...
> > > > I never said that there was only _one_ intact Hebrew verb form. In the paper, Stubbs tells us that "hundreds of fossilized forms of both the suffixed /perfective conjugation (singular yašiba; plural yašib-uu)and the prefixed / imperfective conjugation (yi-/ya-, ti-/ta-,etc) are found in UA".
> > > As a general rule, most loanwords from a language to another are nouns.
> > > Verbs are generally not borrowed, and verbal morphology even less. When verbs are borrowed, they are often no longer a verb in the target language.
> > > so such a claim runs against known examples.

> > In response to one of my earlier posts Peter made it clear that "loanwords have nothing to do with creolization".
>
> yes, but your "theory" has nothing to do with creolization.
> I'm afraid you use the word "creolization" as a kind of wild joker card, in order to "explain" everything you feel like.
> Your claim is that semitic-speakers entered in contact with UA speakers at some point in the past. So the issue is about borrowings resulting from these hypothetical contacts.

Mx. Noob scavenged examples from Holm's book without looking at the definition
of "creole"! A creole is what results when people who have no language in
common and communicate with a pidgin make babies: the infants have only the
"incomplete" input to their language-acquisition system and from this imperfect
example their brains create a full language -- which is what we call a creole.

See Derek Bickerton's book (published under a new title every couple of years)
for the basics and for the implications for the essential nature of human language.

[LingualNoob]

The creolization theory isn't something I invented as an excuse for the lack of underlying morphemes in UA *yawamin-(o)? compared to Hebrew.

I said in one of the earlier posts that although Stubbs doesn't propose a theory about the type of language relationship in his "excerpts" paper, he has said on other occasions that he thinks that the language evidence that he is proposing shows UA to be a creolization involving Hebrew/Egyptian and another (presumably Native American) language.

I was also clear about this fact with Peter in our email conversation last week. Here's a copy/paste from my emeil:

"What Stubbs doesn't say specifically in that "excerpts" paper is whether or not he is proposing a genetic link or just heavy borrowing. In previous papers he had suggested that he thought the proposed language similarities came from creolization."

My own inexperience did confuse creolization with "heavy borrowing". Peter was right to correct me on the fact that "loanwords have nothing to do with creolization", but he is incorrect when he says that creolization is my idea or that it came as a result of finding Holm's book.

[Peter T. Daniels]

On Thursday, August 13, 2015 at 6:49:59 PM UTC-4, LingualNoob wrote:
> On Thursday, August 13, 2015 at 2:35:04 PM UTC-5, Arnaud Fournet wrote:
> > Le jeudi 13 août 2015 21:08:20 UTC+2, LingualNoob a écrit :
> > > On Thursday, August 13, 2015 at 1:28:55 PM UTC-5, Arnaud Fournet wrote:
> > > > Le jeudi 13 août 2015 19:07:15 UTC+2, LingualNoob a écrit :
> > > > > On Thursday, August 13, 2015 at 6:28:15 AM UTC-5, Peter T. Daniels wrote:
> > > > > > On Wednesday, August 12, 2015 at 11:36:41 PM UTC-4, LingualNoob wrote:
> > > > > > > On Tuesday, August 11, 2015 at 10:12:24 PM UTC-5, Peter T. Daniels wrote:

> > > > > > > > ...I.e. there's no connection whatsoever between the two forms.
> > > > > > > Does this mean:
> > > > > > > 1. You see linguistic evidence against the idea that Semitic conjugation could be fossilized in UA *yawamin-(o)?
> > > > > > You just said -- in the evidence you deleted above my conclusion there --
> > > > > My apologies. I was just deleting to keep the post to a smaller size. I wasn't thinking that you or others would suddenly forget the previous post. I'll avoid doing that in the future.
> > > > > > that the morphology of *UA is nothing like the morphology of Hebrew. What
> > > > > > is your Just-So Story to explain that of the tens of thousands of verb
> > > > > > forms in Hebrew, the _one_ that would have been borrowed intact eeans
> > > > > > 'he believes him'?...
> > > > > I never said that there was only _one_ intact Hebrew verb form. In the paper, Stubbs tells us that "hundreds of fossilized forms of both the suffixed /perfective conjugation (singular yašiba; plural yašib-uu)and the prefixed / imperfective conjugation (yi-/ya-, ti-/ta-,etc) are found in UA".
> > > > As a general rule, most loanwords from a language to another are nouns.
> > > > Verbs are generally not borrowed, and verbal morphology even less. When verbs are borrowed, they are often no longer a verb in the target language.
> > > > so such a claim runs against known examples.

> > > In response to one of my earlier posts Peter made it clear that "loanwords have nothing to do with creolization".
> > yes, but your "theory" has nothing to do with creolization.
> > I'm afraid you use the word "creolization" as a kind of wild joker card, in order to "explain" everything you feel like.
> > Your claim is that semitic-speakers entered in contact with UA speakers at some point in the past. So the issue is about borrowings resulting from these hypothetical contacts.
>

> The creolization theory isn't something I invented as an excuse for the lack of underlying morphemes in UA *yawamin-(o)? compared to Hebrew.
>
> I said in one of the earlier posts that although Stubbs doesn't propose a theory about the type of language relationship in his "excerpts" paper, he has said on other occasions that he thinks that the language evidence that he is proposing shows UA to be a creolization involving Hebrew/Egyptian and another (presumably Native American) language.

There is nothing in the structure of UA to suggest a creolization origin.

As A.F. notes, it's not a miscellaneous grab-bag term for vague random resemblance.

> I was also clear about this fact with Peter in our email conversation last week. Here's a copy/paste from my emeil:
>
> "What Stubbs doesn't say specifically in that "excerpts" paper is whether or not he is proposing a genetic link or just heavy borrowing. In previous papers he had suggested that he thought the proposed language similarities came from creolization."
>
> My own inexperience did confuse creolization with "heavy borrowing". Peter was right to correct me on the fact that "loanwords have nothing to do with creolization", but he is incorrect when he says that creolization is my idea or that it came as a result of finding Holm's book.

You've appointed yourself Stubbs's spokesperson, so you're responsible for your statements.

[Dr. HotSalt]

On Saturday, August 8, 2015 at 5:56:02 PM UTC-7, LingualNoob wrote:

(snip)

> http://www.bmaf.org/sites/bmaf.org/files/image/Egyptian-Semitic-in-Uto-Aztecan-by-Brian-Stubbs-Jerry-Grover.pdf

(snip)

No direct linguistic content, but cultural correspondences are indicated here:

https://authorbobfreeman.wordpress.com/2015/04/09/mushrooms-and-religion-by-robert-graves/

The relevant bits are about halfway through in the paragraph beginning "I have eaten the Mexican hallucinogenic mushroom psilocybe Heimsii in Gordon Wasson's company, with the intention of visiting the Mexican paradise called Tlal6can to which it gives access. The god Tlal6c, who was toadheaded, corresponded exactly with Agni and Dionysus." though I recommend reading the whole thing to see why the toad-head remark is significant.

The focus is on the use of amanita muscaris in mystery cults in the Old World and in the New World- the author makes no claims about how such a tradition might have been transferred pre-Columbus (or pre-Ericsson) but merely notes the similarities. You will have to fill in the temporal coincidence, if any.


Dr. HotSalt

[LingualNoob]

Well, If we're at a point where you're rejecting the proposal based on your confidence that PUA didn't undergo creolization thousands of years ago then we're probably done here for now.

I do thank you for taking time to debate a few of the specifics from his paper.

[Peter T. Daniels]

On Friday, August 14, 2015 at 2:47:14 AM UTC-4, LingualNoob wrote:
> On Thursday, August 13, 2015 at 9:51:41 PM UTC-5, Peter T. Daniels wrote:
> > On Thursday, August 13, 2015 at 6:49:59 PM UTC-4, LingualNoob wrote:

> > > I said in one of the earlier posts that although Stubbs doesn't propose a theory about the type of language relationship in his "excerpts" paper, he has said on other occasions that he thinks that the language evidence that he is proposing shows UA to be a creolization involving Hebrew/Egyptian and another (presumably Native American) language.
> > There is nothing in the structure of UA to suggest a creolization origin.
> > As A.F. notes, it's not a miscellaneous grab-bag term for vague random resemblance.
>
> Well, If we're at a point where you're rejecting the proposal based on your confidence that PUA didn't undergo creolization thousands of years ago then we're probably done here for now.
>
> I do thank you for taking time to debate a few of the specifics from his paper.
>

> > You've appointed yourself Stubbs's spokesperson, so you're responsible for your statements.

Learn the characteristics of creole languages. Compare the characteristics of
reconstructed Proto-Uto-Aztecan.

The fact that *UA is not a creole has no bearing on whether magical trans-
Pacific crossings and word-borrowings can have taken place.

The type of posited borrowings, however, is contrary to any known sort of borrowings.

[LingualNoob]

On Friday, August 14, 2015 at 7:40:50 AM UTC-5, Peter T. Daniels wrote:
> On Friday, August 14, 2015 at 2:47:14 AM UTC-4, LingualNoob wrote:
> > On Thursday, August 13, 2015 at 9:51:41 PM UTC-5, Peter T. Daniels wrote:
> > > On Thursday, August 13, 2015 at 6:49:59 PM UTC-4, LingualNoob wrote:
>
> > > > I said in one of the earlier posts that although Stubbs doesn't propose a theory about the type of language relationship in his "excerpts" paper, he has said on other occasions that he thinks that the language evidence that he is proposing shows UA to be a creolization involving Hebrew/Egyptian and another (presumably Native American) language.
> > > There is nothing in the structure of UA to suggest a creolization origin.
> > > As A.F. notes, it's not a miscellaneous grab-bag term for vague random resemblance.
> >
> > Well, If we're at a point where you're rejecting the proposal based on your confidence that PUA didn't undergo creolization thousands of years ago then we're probably done here for now.
> >
> > I do thank you for taking time to debate a few of the specifics from his paper.
> >
> > > You've appointed yourself Stubbs's spokesperson, so you're responsible for your statements.
>
> Learn the characteristics of creole languages. Compare the characteristics of
> reconstructed Proto-Uto-Aztecan.
>

That sounds like a good idea and I don't mind homework assignments :)

Is there a particular book that you'd recommend?

As I've Googled to try to understand the characteristics of creolized languages I've tended to find results related to creolizations that appear to have taken place within the last ~400 +/- years. Those examples provide important data, but in addition to a book dealing with creolizations in general, is there a book or other resource that you can point me to that shows examples of languages that are known to have been creolized at a time-depth measured in thousands of years?

>
> The fact that *UA is not a creole has no bearing on whether magical trans-
> Pacific crossings and word-borrowings can have taken place.
>

Yeah, I know. Your responses are still motivated by your feelings about Mormonism. Mine are too, but I'm honestly not trying to be bull-headed about Stubbs' research. I'm just trying to understand the strengths and weaknesses of his proposal and the debate that we've been having has been tremendously helpful in that regard.

Although I knew that Stubbs saw his proposed evidence as a probable creolization, it wasn't until you provided the morpheme breakdown of Hebrew ya'amiin-o and I researched it trying to find matching UA sub-morphemes that I began to understand why he was suggesting that his proposed evidence pointed towards a creolization as opposed to direct genetic descent.

Your morpheme breakdown taught me that the evidence behind that proposed cognate is limited to semantics and phonetics and that it lacks specific Hebrew sub-structure. You may think that I'm not listening, but the truth is that this exercise had the potential to produce evidence that would cause serious doubts in my mind about the whole proposal. You shouldn't give up on the analysis of his cognates so quickly.

You think that I'm using the creolization idea as a "grab-bag" (as A.F. puts it) to simply excuse anything that doesn't support my biases, but that is not true. If our little investigation into the sub-morphemes of UA *yawamin-(o) had revealed that there is productive UA-specific morpheme sub-structure behind *yawamin-(o) then I would not only accept it as strong evidence against that one proposed cognate, I would also see it as evidence against the proposed creolization.

I was rather disappointed that you simply assumed that I would (or even could) just use the creolization theory as a "grab-bag" and that you seemed to quit engaging with me on the proposal's linguistic evidence as a result. What's the big deal if you can't prove that one example isn't a creolization? There are hundreds of other proposed cognates in his paper.

Why would you want to point the discussion to the "magical" properties of my religion instead of continuing to address the evidence he's proposing? If the proposal is bogus, shouldn't the Comparative Method be able to take that much data and show a lot of examples with strong evidence of contradictory (not just missing) sub-structure?

This doesn't mean that I won't continue to challenge your responses, but if his proposal is as ridiculous as you say it is, wouldn't it be more effective to empty my creolization "grab-bag" based on the linguistic evidence he's proposing rather than trying to get me to agree with your opinion that it's a "fact" that PUA didn't experience creolization thousands of years ago?

The fact is that sci.lang is the only outlet that I know of where people like me can go to receive knowledgeable, critical feedback from people who are proficient in historical linguistics. I hope you can understand that I didn't engage in this conversation supposing that I would walk away with your stamp of approval on Stubbs' proposal. I'm engaging in the conversation to see what happens to the proposal when it's challenged. I may be a little stubborn in my attempted defense of his paper, but that doesn't mean that I'm not listening to what you're saying.

If you are still interested in addressing the linguistic evidence, let's simply look at another proposed cognate. This one catches my attention:

(853) Aramaic ђippušit-aa ‘beetle-the’; Arabic *xunpus > UA *wippusi ‘beetle’ (-np- > -pp- in both Aramaic & UA)
("ђ > w/o" is one of the regular sound correspondences described in the paper)

[Peter T. Daniels]

On Friday, August 14, 2015 at 12:41:48 PM UTC-4, LingualNoob wrote:
> On Friday, August 14, 2015 at 7:40:50 AM UTC-5, Peter T. Daniels wrote:
> > On Friday, August 14, 2015 at 2:47:14 AM UTC-4, LingualNoob wrote:
> > > On Thursday, August 13, 2015 at 9:51:41 PM UTC-5, Peter T. Daniels wrote:
> > > > On Thursday, August 13, 2015 at 6:49:59 PM UTC-4, LingualNoob wrote:
> >
> > > > > I said in one of the earlier posts that although Stubbs doesn't propose a theory about the type of language relationship in his "excerpts" paper, he has said on other occasions that he thinks that the language evidence that he is proposing shows UA to be a creolization involving Hebrew/Egyptian and another (presumably Native American) language.
> > > > There is nothing in the structure of UA to suggest a creolization origin.
> > > > As A.F. notes, it's not a miscellaneous grab-bag term for vague random resemblance.
> > >
> > > Well, If we're at a point where you're rejecting the proposal based on your confidence that PUA didn't undergo creolization thousands of years ago then we're probably done here for now.
> > >
> > > I do thank you for taking time to debate a few of the specifics from his paper.
> > >
> > > > You've appointed yourself Stubbs's spokesperson, so you're responsible for your statements.
> >
> > Learn the characteristics of creole languages. Compare the characteristics of
> > reconstructed Proto-Uto-Aztecan.
> >
>
> That sounds like a good idea and I don't mind homework assignments :)
>
> Is there a particular book that you'd recommend?

You already have Holm!

> As I've Googled to try to understand the characteristics of creolized languages I've tended to find results related to creolizations that appear to have taken place within the last ~400 +/- years. Those examples provide important data, but in addition to a book dealing with creolizations in general, is there a book or other resource that you can point me to that shows examples of languages that are known to have been creolized at a time-depth measured in thousands of years?

No. How would we "know" that?

> > The fact that *UA is not a creole has no bearing on whether magical trans-
> > Pacific crossings and word-borrowings can have taken place.
>
> Yeah, I know. Your responses are still motivated by your feelings about Mormonism.

I had the "feelings" long before I knew you and he were pushing a Mormon
agenda. It's just not good linguistics.

> Mine are too, but I'm honestly not trying to be bull-headed about Stubbs' research. I'm just trying to understand the strengths and weaknesses of his proposal and the debate that we've been having has been tremendously helpful in that regard.
>
> Although I knew that Stubbs saw his proposed evidence as a probable creolization, it wasn't until you provided the morpheme breakdown of Hebrew ya'amiin-o and I researched it trying to find matching UA sub-morphemes that I began to understand why he was suggesting that his proposed evidence pointed towards a creolization as opposed to direct genetic descent.
>
> Your morpheme breakdown taught me that the evidence behind that proposed cognate is limited to semantics and phonetics and that it lacks specific Hebrew sub-structure. You may think that I'm not listening, but the truth is that this exercise had the potential to produce evidence that would cause serious doubts in my mind about the whole proposal. You shouldn't give up on the analysis of his cognates so quickly.
>
> You think that I'm using the creolization idea as a "grab-bag" (as A.F. puts it) to simply excuse anything that doesn't support my biases, but that is not true. If our little investigation into the sub-morphemes of UA *yawamin-(o) had revealed that there is productive UA-specific morpheme sub-structure behind *yawamin-(o) then I would not only accept it as strong evidence against that one proposed cognate, I would also see it as evidence against the proposed creolization.
>
> I was rather disappointed that you simply assumed that I would (or even could) just use the creolization theory as a "grab-bag" and that you seemed to quit engaging with me on the proposal's linguistic evidence as a result. What's the big deal if you can't prove that one example isn't a creolization? There are hundreds of other proposed cognates in his paper.

Sorry, but "one example isn't a creolization" is incoherent.

> Why would you want to point the discussion to the "magical" properties of my religion instead of continuing to address the evidence he's proposing? If the proposal is bogus, shouldn't the Comparative Method be able to take that much data and show a lot of examples with strong evidence of contradictory (not just missing) sub-structure?
>
> This doesn't mean that I won't continue to challenge your responses, but if his proposal is as ridiculous as you say it is, wouldn't it be more effective to empty my creolization "grab-bag" based on the linguistic evidence he's proposing rather than trying to get me to agree with your opinion that it's a "fact" that PUA didn't experience creolization thousands of years ago?

Once you understand the notion of "TMA," as the creolists call it, you'll
revise your attitude.

> The fact is that sci.lang is the only outlet that I know of where people like me can go to receive knowledgeable, critical feedback from people who are proficient in historical linguistics. I hope you can understand that I didn't engage in this conversation supposing that I would walk away with your stamp of approval on Stubbs' proposal. I'm engaging in the conversation to see what happens to the proposal when it's challenged. I may be a little stubborn in my attempted defense of his paper, but that doesn't mean that I'm not listening to what you're saying.
>
> If you are still interested in addressing the linguistic evidence, let's simply look at another proposed cognate. This one catches my attention:
>
> (853) Aramaic ђippušit-aa ‘beetle-the’; Arabic *xunpus > UA *wippusi ‘beetle’ (-np- > -pp- in both Aramaic & UA)
> ("ђ > w/o" is one of the regular sound correspondences described in the paper)

Not sure what that first letter is supposed to be -- I'm seeing an h with
both a crossbar and an ng-hook. Presumably it's supposed to be crossed-h (equivalent to h-underdot).

The given form is not listed in Brockelmann's Syriac lexicon, but Sokoloff
has it in his Babylonian Aramaic (i.e. mostly the Talmud) dictionary, so at
least 1000 years later than you need it to be. He finds that Brockelmann's
form is (using x to represent that letter) xabshushta; the root is XBSh
'to imprison'. No p, no n; the Arabic form may be a folk-etymologized dissimilation: if they knew that -CC- in Aramaic words often corresponded to
-nC- in Arabic, they might analogize too far. However, the * on the Arabic
form is a red flag. What is it reconstructed from?

[Yusuf B Gursey]

In <54b906bf-1b02-4932...@googlegroups.com>, Peter T.

It's xunfus in Arabic, which would give *xunpus

[Arnaud Fournet]

Le vendredi 14 août 2015 18:41:48 UTC+2, LingualNoob a écrit :

> If you are still interested in addressing the linguistic evidence, let's simply look at another proposed cognate.

It can't be a cognate.
I would suggest comparand-um/-a instead.
A.

This one catches my attention:
>
> (853) Aramaic ђippušit-aa ‘beetle-the’; Arabic *xunpus > UA *wippusi ‘beetle’ (-np- > -pp- in both Aramaic & UA)
> ("ђ > w/o" is one of the regular sound correspondences described in the paper)

According to Wikipedia PUA had a fricative *h
How do you explain that ђ or x was not just changed into *h, which is about the same sound?
A.

[LingualNoob]

On Friday, August 14, 2015 at 2:42:08 PM UTC-5, Arnaud Fournet wrote:
> Le vendredi 14 août 2015 18:41:48 UTC+2, LingualNoob a écrit :
>
> > If you are still interested in addressing the linguistic evidence, let's simply look at another proposed cognate.
>
> It can't be a cognate.
> I would suggest comparand-um/-a instead.
> A.
>

That's fine. We'll go with "comparandum".

>
> This one catches my attention:
> >
> > (853) Aramaic ђippušit-aa ‘beetle-the’; Arabic *xunpus > UA *wippusi ‘beetle’ (-np- > -pp- in both Aramaic & UA)
> > ("ђ > w/o" is one of the regular sound correspondences described in the paper)
>
> According to Wikipedia PUA had a fricative *h
>

ђ = IPA #175 (https://en.wikipedia.org/wiki/Sj-sound)

>
> How do you explain that ђ or x was not just changed into *h, which is about the same sound?
> A.

You are correct. I should have copied/pasted the full description of the correspondence from the paper: "Semitic initial voiceless pharyngeal ђ > UA *hu, or w/o/u, and non-initially ђ > w/o/u"

[Peter T. Daniels]

On Friday, August 14, 2015 at 5:28:46 PM UTC-4, LingualNoob wrote:
> On Friday, August 14, 2015 at 2:42:08 PM UTC-5, Arnaud Fournet wrote:
> > Le vendredi 14 août 2015 18:41:48 UTC+2, LingualNoob a écrit :

> > > If you are still interested in addressing the linguistic evidence, let's simply look at another proposed cognate.
> > It can't be a cognate.
> > I would suggest comparand-um/-a instead.
>

> That's fine. We'll go with "comparandum".
>
> > This one catches my attention:
>
> > > (853) Aramaic ђippušit-aa ‘beetle-the’; Arabic *xunpus > UA *wippusi ‘beetle’ (-np- > -pp- in both Aramaic & UA)
> > > ("ђ > w/o" is one of the regular sound correspondences described in the paper)
> > According to Wikipedia PUA had a fricative *h
>
> ђ = IPA #175 (https://en.wikipedia.org/wiki/Sj-sound)

that sound is unknown outside Swedish. It is a mistake in Aramaic or Semitic
generally.

> > How do you explain that ђ or x was not just changed into *h, which is about the same sound?
>

> You are correct. I should have copied/pasted the full description of the correspondence from the paper: "Semitic initial voiceless pharyngeal ђ > UA *hu, or w/o/u, and non-initially ђ > w/o/u"

That he uses the wrong symbol -- consistently! -- shows that you basically
can't trust anything he says about Semitic. It's a symbol you learn on the
first day of class on any Semitic language except Akkadian.

[Peter T. Daniels]

Oh. Is that all. I wonder why he wants to pretend it's /p/.

[Arnaud Fournet]

Le vendredi 14 août 2015 23:28:46 UTC+2, LingualNoob a écrit :
> On Friday, August 14, 2015 at 2:42:08 PM UTC-5, Arnaud Fournet wrote:
> > Le vendredi 14 août 2015 18:41:48 UTC+2, LingualNoob a écrit :
> >
> > > If you are still interested in addressing the linguistic evidence, let's simply look at another proposed cognate.
> >
> > It can't be a cognate.
> > I would suggest comparand-um/-a instead.
> > A.
> >
>
> That's fine. We'll go with "comparandum".

yes, avoid the regular vocabulary about genetic linguistics like cognate, sound correspondence, etc. if the target of your theory is about loanwords.
A.

>
> >
> > This one catches my attention:
> > >
> > > (853) Aramaic ђippušit-aa ‘beetle-the’; Arabic *xunpus > UA *wippusi ‘beetle’ (-np- > -pp- in both Aramaic & UA)
> > > ("ђ > w/o" is one of the regular sound correspondences described in the paper)
> >

Besides, may we have the original data supporting the reconstruction UA *wippusi ‘beetle'? Which UA languages have that word?
Incidentally, could UA have geminates? Is it proved?
A.

> > According to Wikipedia PUA had a fricative *h
> >
>
> ђ = IPA #175 (https://en.wikipedia.org/wiki/Sj-sound)

From a phonological PoV, I would expect any voiceless fricative like x X or h. to be adapted as h, if h exists in the target language.
Besides, we have no exact idea how the letters were actually phoned in ancient languages. Their values are mostly a matter of convention. So it's rather pointless to make a fuss about that fricative being velar, uvular or pharyngeal.
For the same reason, we don't know if UA *h was glottal or not.
A.

>
> >
> > How do you explain that ђ or x was not just changed into *h, which is about the same sound?
> > A.
>
> You are correct. I should have copied/pasted the full description of the correspondence from the paper: "Semitic initial voiceless pharyngeal ђ > UA *hu, or w/o/u, and non-initially ђ > w/o/u"

it should be only ђ or whatever h symbol > UA *h,
the other changes are unacceptable.
A.

[Peter T. Daniels]

On Saturday, August 15, 2015 at 3:22:10 AM UTC-4, Arnaud Fournet wrote:

> From a phonological PoV, I would expect any voiceless fricative like x X or h. to be adapted as h, if h exists in the target language.
> Besides, we have no exact idea how the letters were actually phoned in ancient languages. Their values are mostly a matter of convention. So it's rather pointless to make a fuss about that fricative being velar, uvular or pharyngeal.

Given that Semitic has all three -- voiceless velar, pharyngeal, and larnyngeal
fricatives, as well as voiced velar and laryngeal fricatives, it is hardly
"pointless" to make those distinctions in the five reconstructed phonemes.

But it's become clear from this thread alone that your knowledge of Semitic
languages and linguistics is minimal.

[LingualNoob]

Once again I've provided incorrect information. I said 'ђ' was IPA #175 but I was wrong. At the time I wrote that I didn't have the book in front of me so I assumed it was #175 based off of looking at an IPA chart and thinking that it looked right (although looking at it again it doesn't even look right). I know this was pretty poor of me.

Today I searched for his description of the symbol in the preliminary copy of his book and found that he spends a paragraph describing why he uses that particular symbol. I don't want to quote a whole paragraph verbatim from an unpublished book so I will simply say that it can be considered the same as h-underdot for our purposes.

I'm honestly trying to understand everything that's been said about "Aramaic ђippušit-aa ‘beetle-the’; Arabic *xunpus", but I'm not sure I understand how everything that has been discussed all fits together. Here is my understanding at this point:

1. The Proto-Arabic form *xunpus seems correct.

2. There are doubts about Stubbs' assertion that Arabic -np- > -pp- in Aramaic.

3. None of us have substantiated or refuted Stubbs' assertion that -np- > -pp- in UA.

4. The Aramaic form ђippušit-aa is properly attested to mean 'beetle-the', but the source where you find it (Sokoloff) cannot be used to assert that it meant 'beetle-the' earlier than 400 AD. Further, Sokoloff says that ђippušit-aa is derived from Brockelmann's form xabshushta whose root 'XBSh' means 'to imprison'.

5. 'ђ' > 'w' in Aramaic ђippušit-aa > UA *wippusi is not objectionable as long as Stubbs shows that it is a regular correspondence.

Is all ^this^ correct/complete based on what has been discussed so far?

[Peter T. Daniels]

On Saturday, August 15, 2015 at 4:06:04 PM UTC-4, LingualNoob wrote:
> On Saturday, August 15, 2015 at 7:18:10 AM UTC-5, Peter T. Daniels wrote:
> > On Saturday, August 15, 2015 at 3:22:10 AM UTC-4, Arnaud Fournet wrote:
> >
> > > From a phonological PoV, I would expect any voiceless fricative like x X or h. to be adapted as h, if h exists in the target language.
> > > Besides, we have no exact idea how the letters were actually phoned in ancient languages. Their values are mostly a matter of convention. So it's rather pointless to make a fuss about that fricative being velar, uvular or pharyngeal.
> >
> > Given that Semitic has all three -- voiceless velar, pharyngeal, and larnyngeal
> > fricatives, as well as voiced velar and laryngeal fricatives, it is hardly
> > "pointless" to make those distinctions in the five reconstructed phonemes.
> >
> > But it's become clear from this thread alone that your knowledge of Semitic
> > languages and linguistics is minimal.
>
> Once again I've provided incorrect information. I said 'ђ' was IPA #175 but I was wrong. At the time I wrote that I didn't have the book in front of me so I assumed it was #175 based off of looking at an IPA chart and thinking that it looked right (although looking at it again it doesn't even look right). I know this was pretty poor of me.
>
> Today I searched for his description of the symbol in the preliminary copy of his book and found that he spends a paragraph describing why he uses that particular symbol. I don't want to quote a whole paragraph verbatim from an unpublished book so I will simply say that it can be considered the same as h-underdot for our purposes.

The IPA equivalent of Semitic h-underdot is the simple crossed-h [ħ], Unicode U0127. (It's in the Latin Extended-A group because it's used in the
standard orthography of Maltese, a European language.)(I don't know what
the numbering of IPA letters you use represents.)

> I'm honestly trying to understand everything that's been said about "Aramaic ђippušit-aa ‘beetle-the’; Arabic *xunpus", but I'm not sure I understand how everything that has been discussed all fits together. Here is my understanding at this point:
>
> 1. The Proto-Arabic form *xunpus seems correct.

What do you mean by "Proto-Arabic"? Yusuf conjectures that he put a star on
that form because he had a reason for not giving the true Arabic form xunfus.

> 2. There are doubts about Stubbs' assertion that Arabic -np- > -pp- in Aramaic.

There are no doubts. The n is not found in that word in the other Semitic
languages.

> 3. None of us have substantiated or refuted Stubbs' assertion that -np- > -pp- in UA.

That step would only be necessary if there were some reason to suppose the
word was borrowed from Arabic.

> 4. The Aramaic form ђippušit-aa is properly attested to mean 'beetle-the', but the source where you find it (Sokoloff) cannot be used to assert that it meant 'beetle-the' earlier than 400 AD. Further, Sokoloff says that ђippušit-aa is derived from Brockelmann's form xabshushta whose root 'XBSh' means 'to imprison'.

Note that the hypothesis does not involve Aramaeans crossing the Pacific,
only "Hebrews."

The -aa suffix no longer serves as a definite article in either Syriac or
Talmudic Aramaic. The -aa form is simply the basic form of the noun, which
only loses the -aa when the word is the first member of the "construct"
construction.

The TA form isn't _derived_ from the Syriac form; they're parallel developments
from the root for 'encase' (the more general sense than 'imprision', which
fits better with beetles). The -u- is fairly unusual in a nominal form; it
may help indicate an animal or an insect name.

> 5. 'ђ' > 'w' in Aramaic ђippušit-aa > UA *wippusi is not objectionable as long as Stubbs shows that it is a regular correspondence.

Not the Semitist's problem.

[LingualNoob]

On Saturday, August 15, 2015 at 5:02:04 PM UTC-5, Peter T. Daniels wrote:
> On Saturday, August 15, 2015 at 4:06:04 PM UTC-4, LingualNoob wrote:
> > On Saturday, August 15, 2015 at 7:18:10 AM UTC-5, Peter T. Daniels wrote:
> > > On Saturday, August 15, 2015 at 3:22:10 AM UTC-4, Arnaud Fournet wrote:
> > >
> > > > From a phonological PoV, I would expect any voiceless fricative like x X or h. to be adapted as h, if h exists in the target language.
> > > > Besides, we have no exact idea how the letters were actually phoned in ancient languages. Their values are mostly a matter of convention. So it's rather pointless to make a fuss about that fricative being velar, uvular or pharyngeal.
> > >
> > > Given that Semitic has all three -- voiceless velar, pharyngeal, and larnyngeal
> > > fricatives, as well as voiced velar and laryngeal fricatives, it is hardly
> > > "pointless" to make those distinctions in the five reconstructed phonemes.
> > >
> > > But it's become clear from this thread alone that your knowledge of Semitic
> > > languages and linguistics is minimal.
> >
> > Once again I've provided incorrect information. I said 'ђ' was IPA #175 but I was wrong. At the time I wrote that I didn't have the book in front of me so I assumed it was #175 based off of looking at an IPA chart and thinking that it looked right (although looking at it again it doesn't even look right). I know this was pretty poor of me.
> >
> > Today I searched for his description of the symbol in the preliminary copy of his book and found that he spends a paragraph describing why he uses that particular symbol. I don't want to quote a whole paragraph verbatim from an unpublished book so I will simply say that it can be considered the same as h-underdot for our purposes.
>
> The IPA equivalent of Semitic h-underdot is the simple crossed-h [ħ], Unicode U0127. (It's in the Latin Extended-A group because it's used in the
> standard orthography of Maltese, a European language.)(I don't know what
> the numbering of IPA letters you use represents.)
>
> > I'm honestly trying to understand everything that's been said about "Aramaic ђippušit-aa ‘beetle-the’; Arabic *xunpus", but I'm not sure I understand how everything that has been discussed all fits together. Here is my understanding at this point:
> >
> > 1. The Proto-Arabic form *xunpus seems correct.
>
> What do you mean by "Proto-Arabic"? Yusuf conjectures that he put a star on
> that form because he had a reason for not giving the true Arabic form xunfus.
>

From Stubbs' paper: "* marks a proto-form or original sound or word as reconstructed by linguists."

...which led me to call it "Proto-Arabic".

>
> > 2. There are doubts about Stubbs' assertion that Arabic -np- > -pp- in Aramaic.
>
> There are no doubts. The n is not found in that word in the other Semitic
> languages.
>
> > 3. None of us have substantiated or refuted Stubbs' assertion that -np- > -pp- in UA.
>
> That step would only be necessary if there were some reason to suppose the
> word was borrowed from Arabic.
>
> > 4. The Aramaic form ђippušit-aa is properly attested to mean 'beetle-the', but the source where you find it (Sokoloff) cannot be used to assert that it meant 'beetle-the' earlier than 400 AD. Further, Sokoloff says that ђippušit-aa is derived from Brockelmann's form xabshushta whose root 'XBSh' means 'to imprison'.
>
> Note that the hypothesis does not involve Aramaeans crossing the Pacific,
> only "Hebrews."
>
> The -aa suffix no longer serves as a definite article in either Syriac or
> Talmudic Aramaic. The -aa form is simply the basic form of the noun, which
> only loses the -aa when the word is the first member of the "construct"
> construction.
>
> The TA form isn't _derived_ from the Syriac form; they're parallel developments
> from the root for 'encase' (the more general sense than 'imprision', which
> fits better with beetles). The -u- is fairly unusual in a nominal form; it
> may help indicate an animal or an insect name.
>
> > 5. 'ђ' > 'w' in Aramaic ђippušit-aa > UA *wippusi is not objectionable as long as Stubbs shows that it is a regular correspondence.
>
> Not the Semitist's problem.
>
> > Is all ^this^ correct/complete based on what has been discussed so far?

Thank you for the clarifications. Is it safe to say then that that if we assume his ђ > w correspondence and ignore the problem with the time depth that TA "ђippušit-aa ‘beetle-the’ > UA *wippusi ‘beetle’" is a strong semantic and phonetic match?

In Marcus Jastrow's "A Dictionary of the Targumim, The Talmud Babli, and Yerushami, and the Midrashic Literature" on page 459 it lists the form and says "ch. same" (ch.=Chaldeac). Does the fact that Chaldeac has the same form do anything to help with the time-depth problem?

[Peter T. Daniels]

On Saturday, August 15, 2015 at 8:40:12 PM UTC-4, LingualNoob wrote:

> Thank you for the clarifications. Is it safe to say then that that if we assume his ђ > w correspondence and ignore the problem with the time depth

And the fact that the hypothesis concerns Hebrew, not Aramaic

> that TA "ђippušit-aa ‘beetle-the’ > UA *wippusi ‘beetle’" is a strong semantic and phonetic match?

As A.F. said, the basis for reconstructing that form and meaning for the *UA
word should be provided.

> In Marcus Jastrow's "A Dictionary of the Targumim, The Talmud Babli, and Yerushami, and the Midrashic Literature" on page 459 it lists the form and says "ch. same" (ch.=Chaldeac). Does the fact that Chaldeac has the same form do anything to help with the time-depth problem?

Socratic questions: Same as what? Do you know what "Chaldaean" meant to
Jastrow?

Sokoloff started his series of Aramaic dictionaries with the Babylonian one
because the existing ones -- namely, Jastrow's from 1901 -- were so unreliable. Jastrow, a good rabbinic scholar (from Philadelphia, BTW), was
constitutionally unable to recognize that many words in the Talmud etc. were
borrowed from Greek and came up with bizarre Semitic explanations. The XBSh
for this one, though, is from Brockelmann, the most erudite Semitist of his
generation.

[LingualNoob]

On Saturday, August 15, 2015 at 9:53:05 PM UTC-5, Peter T. Daniels wrote:
> On Saturday, August 15, 2015 at 8:40:12 PM UTC-4, LingualNoob wrote:
>
> > Thank you for the clarifications. Is it safe to say then that that if we assume his ђ > w correspondence and ignore the problem with the time depth
>
> And the fact that the hypothesis concerns Hebrew, not Aramaic
>

Stubbs hypothesizes that the language spoken Semitic group involved here was NorthWest Semitic influenced significantly by Aramaic. He discusses the Aramaic tendencies in a couple of places in the paper but particularly on page 11.

>
> > that TA "ђippušit-aa ‘beetle-the’ > UA *wippusi ‘beetle’" is a strong semantic and phonetic match?
>
> As A.F. said, the basis for reconstructing that form and meaning for the *UA
> word should be provided.
>

The basis for the reconstruction is described in the unpublished book.

>
> > In Marcus Jastrow's "A Dictionary of the Targumim, The Talmud Babli, and Yerushami, and the Midrashic Literature" on page 459 it lists the form and says "ch. same" (ch.=Chaldeac). Does the fact that Chaldeac has the same form do anything to help with the time-depth problem?
>
> Socratic questions: Same as what? Do you know what "Chaldaean" meant to
> Jastrow?
>
> Sokoloff started his series of Aramaic dictionaries with the Babylonian one
> because the existing ones -- namely, Jastrow's from 1901 -- were so unreliable. Jastrow, a good rabbinic scholar (from Philadelphia, BTW), was
> constitutionally unable to recognize that many words in the Talmud etc. were
> borrowed from Greek and came up with bizarre Semitic explanations. The XBSh
> for this one, though, is from Brockelmann, the most erudite Semitist of his
> generation.

Fair enough.

[Arnaud Fournet]

Pontificating as usual about your own inability to understand what people say.
I was not talking about Proto-Semitic. We have no clear idea what Akkadian h stands for. It stems from PS voiceless velar, it stems from, this does not mean it's a voiceless velar.
A.

[Arnaud Fournet]

Le samedi 15 août 2015 22:06:04 UTC+2, LingualNoob a écrit :
> On Saturday, August 15, 2015 at 7:18:10 AM UTC-5, Peter T. Daniels wrote:
> > On Saturday, August 15, 2015 at 3:22:10 AM UTC-4, Arnaud Fournet wrote:
> >

>
> Once again I've provided incorrect information. I said 'ђ' was IPA #175 but I was wrong. At the time I wrote that I didn't have the book in front of me so I assumed it was #175 based off of looking at an IPA chart and thinking that it looked right (although looking at it again it doesn't even look right). I know this was pretty poor of me.
>
> Today I searched for his description of the symbol in the preliminary copy of his book and found that he spends a paragraph describing why he uses that particular symbol. I don't want to quote a whole paragraph verbatim from an unpublished book so I will simply say that it can be considered the same as h-underdot for our purposes.

Anyway, it's only a matter of symbols and graphic conventions.
A.

>
> I'm honestly trying to understand everything that's been said about "Aramaic ђippušit-aa ‘beetle-the’; Arabic *xunpus", but I'm not sure I understand how everything that has been discussed all fits together. Here is my understanding at this point:
>
> 1. The Proto-Arabic form *xunpus seems correct.
>
> 2. There are doubts about Stubbs' assertion that Arabic -np- > -pp- in Aramaic.

This makes little sense in the first place to posit that Arabic would give loanwords to Aramaic. The direction of borrowing is usually the opposite.
A.

>
> 3. None of us have substantiated or refuted Stubbs' assertion that -np- > -pp- in UA.

you have not even proved that pp in UA is a valid hypothesis.

A.

>
> 4. The Aramaic form ђippušit-aa is properly attested to mean 'beetle-the', but the source where you find it (Sokoloff) cannot be used to assert that it meant 'beetle-the' earlier than 400 AD. Further, Sokoloff says that ђippušit-aa is derived from Brockelmann's form xabshushta whose root 'XBSh' means 'to imprison'.

Anyway, there's no problem with a Semitic root *x_p "beetle", with different derivatives: xunpus, xippuSit, etc.
A.

>
> 5. 'ђ' > 'w' in Aramaic ђippušit-aa > UA *wippusi is not objectionable as long as Stubbs shows that it is a regular correspondence.

No, it's objectionable, as I have already stated.
we should have all Semitic H > UA *h.
The rest is unacceptable.
A.

>
> Is all ^this^ correct/complete based on what has been discussed so far?

No.
A.

[Arnaud Fournet]

it's very weak phonetically, as stated before.
Besides, we are still waiting for the UA data supporting the reconstruction *wippusi.
A.

[Yusuf B Gursey]

In <144c9ee0-6686-4f30...@googlegroups.com>, Arnaud

Nevertheless, the asssimilation of /n/ to a following consonant in NW
Semitic is an established phenomenon, though I don't know if it works
with labials. OTOH Arabic is said to have examples of dissimilation of
doubled consonants CC > nC. If xunfur is a loanword from Aramaic it is
a very old loanword when Aramaic differentiated between /ħ/ and /x/.
Aramaic Heth regularly becomes Arabic /ħ/ in loanwords from Aramaicc
and it is the unmarked form of Heth that became Arabic script ħa:'. I
presume that Old Arabic /f/ was bilabial or somehow resembling [p], and
of course comes from PS */p/. Islam was introduced into Ceylon and the
Indian Ocean lands by Arab sailors from the SE Arabian coast and not by
Persians (and later Turks) as in the rest of Asia. In Arabic script
Tamil and Malay one has modified Arabic script <f> to represent /p/ and
not modified <b> as in Persian. There is also Arabic fi:l "elephant"
(mentioned in the Qur'an) for Persian pi:l "elephant" and a few other
early loanwords, as well as Aramaic /p/ regularly appearing as /f/ in
loanwords. In later loanwords Arabic usually renders foreign /p/ as
/b/.

[Peter T. Daniels]

Do you really not understand how philology works? There are transcriptions of
Akkadian and other cuneiform-written names into alphabetic scripts that
preserve the distinctions; and there is converging comparative evidence.

Moreover, the voiceless pharyngeal fricative under discussion here is _not_
represented by Akkadian voiceless-velar-fricative, but by glottal-that-
influences-the-adjacent-vowel-from-a-to-e.

You _really_ should not comment on things you know nothing about.

Which may well be what the readers for IJAL will say about the article you
submitted to them.

[Peter T. Daniels]

On Sunday, August 16, 2015 at 3:47:46 AM UTC-4, Yusuf B Gursey wrote:
> In <144c9ee0-6686-4f30...@googlegroups.com>, Arnaud
> Fournet wrote on 8/16/2015:
> > Le samedi 15 août 2015 22:06:04 UTC+2, LingualNoob a écrit :

> >> 1. The Proto-Arabic form *xunpus seems correct.
> >>
> >> 2. There are doubts about Stubbs' assertion that Arabic -np- > -pp- in
> >> Aramaic.
> >
> > This makes little sense in the first place to posit that Arabic would give
> > loanwords to Aramaic. The direction of borrowing is usually the opposite.
>

> Nevertheless, the asssimilation of /n/ to a following consonant in NW
> Semitic is an established phenomenon,

There IS NO N in the Aramaic/Proto-West Semitic form.

[Yusuf B Gursey]

In <a8b0b97e-0600-4c6a...@googlegroups.com>, Peter T.

OK. No problem.

[Arnaud Fournet]

You're an idiot, who knows only vague generalities, overbaked and overcooked for the dummies, like you.
You understand even less than you know, of course.
A.

[LingualNoob]

Here's an example that I'm curious to see your reactions to because it involves verbal nouns and semantic leeway:

"(724) Semitic parʕoš ‘flea (jumper)’ (from the Semitic verb prʕš ‘jump’) > UA *par’osi / *paro’osi ‘jackrabbit’; the jackrabbit, like the flea, is also a jumper, and in UA *paro’osi ‘jackrabbit’ we see all 4 consonants and 2 identical vowels in two of the most extraordinary jumpers of the animal kingdom."

I think the argument Stubbs is making here isn't that 'flea' became 'jackrabbit', but rather that since Semitic has the noun 'flea' based on the verb prʕš that it is reasonable to see the noun 'jackrabbit' based on the same verb.

I honestly don't know if you will find this kind of comparison to be acceptable and would appreciate your thoughts on it.

[Yusuf B Gursey]

In <54acab8d-5c8d-4f45...@googlegroups.com>, LingualNoob

wrote on 8/17/2015:
> Here's an example that I'm curious to see your reactions to because it
> involves verbal nouns and semantic leeway:
>
> "(724) Semitic parʕoš ‘flea (jumper)’ (from the Semitic verb prʕš ‘jump’) >

Arabic bur*gh*u:*th* also bar*gh*u:*th* (more common in colloquials)
Hebrew parʕoš; Assyrian (acc. to BDB) par paršu-u , puršu-u

So it's Hebrew PS would be *par*gh*u:*th*, *pur*gh*u:*th*

[Peter T. Daniels]

Then all you have to do is show that they borrowed the 'jump' verb.

I'd guess that if you were going to call something "flea" in a new enviroment
it would be its itchiness rather than its jumpiness that was most salient.
There are perfectly good Hebrew words for rabbit-sorts of things, since
they're among the ones listed as unclean.

[Peter T. Daniels]

On Sunday, August 16, 2015 at 7:28:31 PM UTC-4, Yusuf B Gursey wrote:
> In <54acab8d-5c8d-4f45...@googlegroups.com>, LingualNoob
> wrote on 8/17/2015:

> > Here's an example that I'm curious to see your reactions to because it
> > involves verbal nouns and semantic leeway:
> >
> > "(724) Semitic parʕoš ‘flea (jumper)’ (from the Semitic verb prʕš ‘jump’) >
>
> Arabic bur*gh*u:*th* also bar*gh*u:*th* (more common in colloquials)
> Hebrew parʕoš; Assyrian (acc. to BDB) par paršu-u , puršu-u
>
> So it's Hebrew PS would be *par*gh*u:*th*, *pur*gh*u:*th*

And we know from Greek transcriptions in the Septuagint (quite a while after
the posited magic-carpet ride) that Hebrew had kept the distinction between
`Ayn and Ghain, even though they were spelled with the same letter (because
in Phoenician they had merged).

But Th > Sh prehistorically (Th > T in Aramaic).

[Yusuf B Gursey]

In <f380b361-1e92-4760...@googlegroups.com>, Peter T.

Good point.

[Arnaud Fournet]

I have found Uralic-looking material in Mayan languages as well.
A.

[Peter T. Daniels]

That puts you right in Lyle Campbell's wheelhouse, since those are his two
areas of specialization -- and one of the pairings for which he drew up a list
of deceptive-looking chance similarities.

BTW Ed Vaida has retired from the editorial board of WORD (the journal of the
ILA), effective at the end of the 2015 volume.

[LingualNoob]

On Sunday, August 16, 2015 at 9:09:05 PM UTC-5, Peter T. Daniels wrote:
> On Sunday, August 16, 2015 at 7:28:31 PM UTC-4, Yusuf B Gursey wrote:

> > In LingualNoob

> > wrote on 8/17/2015:
>
> > > Here's an example that I'm curious to see your reactions to because it
> > > involves verbal nouns and semantic leeway:
> > >
> > > "(724) Semitic parʕoš ‘flea (jumper)’ (from the Semitic verb prʕš ‘jump’) >
> >
> > Arabic bur*gh*u:*th* also bar*gh*u:*th* (more common in colloquials)
> > Hebrew parʕoš; Assyrian (acc. to BDB) par paršu-u , puršu-u
> >
> > So it's Hebrew PS would be *par*gh*u:*th*, *pur*gh*u:*th*
>
> And we know from Greek transcriptions in the Septuagint (quite a while after
> the posited magic-carpet ride) that Hebrew had kept the distinction between
> `Ayn and Ghain, even though they were spelled with the same letter (because
> in Phoenician they had merged).
>

I'm trying to understand...back in the days of the magic-carpet, would ʕ = 'Ayn or would ʕ = Ghain in prʕš?

[LingualNoob]

Where can I find Campbell's list of deceptive-looking chance similarities so I can see how a list that is known to be coincidental compares to Stubbs' proposal?

[Yusuf B Gursey]

In <7bc948be-137f-4c11...@googlegroups.com>, LingualNoob

Ghain. It was around in the time of the LXX

[Peter T. Daniels]

Proto-Semitic had two sounds, *ʕ and *ɣ. They were preserved to this day in
Arabic ع and غ (and in a few other languages), and there are distinct letters
for them in Ugaritic, too. By the time the Phoenician "alphabet" was devised,
probably late in the 2nd millennium BCE, the distinction had been lost in
Phoenician and there's only the one letter ע.

But even though Hebrew adopted a variety of the Phoenician alphabet, with
only the one letter, we know from the spelling of Gaza (as in the Strip) in
the Greek Bible (the Septuagint) that the name spelled `aza in Hebrew was
still pronounced Ghaza. However, the distinction had been lost by the time
the dots were added to the Hebrew text in the second half of the first
millennium CE; if the distinction had been preserved they would have marked
the letter in some way, as with the distinction between שׂ Sin and שׁ Shin.

> > But Th > Sh prehistorically (Th > T in Aramaic).

So depending on your timeframe, it would be either (PSem) PRGhTh or (Heb) PRGhSh.

[Peter T. Daniels]

On Monday, August 17, 2015 at 10:31:57 AM UTC-4, LingualNoob wrote:
> On Monday, August 17, 2015 at 5:54:25 AM UTC-5, Peter T. Daniels wrote:
> > On Monday, August 17, 2015 at 4:45:35 AM UTC-4, Arnaud Fournet wrote:

> > > I have found Uralic-looking material in Mayan languages as well.

> > That puts you right in Lyle Campbell's wheelhouse, since those are his two
> > areas of specialization -- and one of the pairings for which he drew up a list
> > of deceptive-looking chance similarities.
>

> Where can I find Campbell's list of deceptive-looking chance similarities so I can see how a list that is known to be coincidental compares to Stubbs' proposal?

I don't remember -- probably in his historical linguistics textbook, but he
also wrote dozens of polemics against "long rangers" making the same point.

[Arnaud Fournet]

Le lundi 17 août 2015 12:54:25 UTC+2, Peter T. Daniels a écrit :
> On Monday, August 17, 2015 at 4:45:35 AM UTC-4, Arnaud Fournet wrote:
> > Le dimanche 16 août 2015 14:00:17 UTC+2, Peter T. Daniels a écrit :
> > > On Sunday, August 16, 2015 at 1:39:51 AM UTC-4, Arnaud Fournet wrote:
> > > > Le samedi 15 août 2015 14:18:10 UTC+2, Peter T. Daniels a écrit :
> > > > > On Saturday, August 15, 2015 at 3:22:10 AM UTC-4, Arnaud Fournet wrote:
>
> > > Which may well be what the readers for IJAL will say about the article you
> > > submitted to them.
> >
> > I have found Uralic-looking material in Mayan languages as well.
> > A.
>
> That puts you right in Lyle Campbell's wheelhouse, since those are his two
> areas of specialization -- and one of the pairings for which he drew up a list
> of deceptive-looking chance similarities.

This morning, I found that quite a lot of Quechua words look Uralic.
This is interesting as Quechua has the same system as Na-Dene: voiceless - glottalized - aspirated.
So if my cognates have the same consonants in Na-Dene and Quechua, then it's bingo.
A.

>
> BTW Ed Vaida has retired from the editorial board of WORD (the journal of the
> ILA), effective at the end of the 2015 volume.

Anyway, don't you need to be a member of some association to have the right to submit to Word?
A.

[Peter T. Daniels]

On Monday, August 17, 2015 at 1:04:25 PM UTC-4, Arnaud Fournet wrote:

> > BTW Ed Vaida has retired from the editorial board of WORD (the journal of the
> > ILA), effective at the end of the 2015 volume.
>
> Anyway, don't you need to be a member of some association to have the right to submit to Word?

No.

After the hiatus caused by the incapacity of the former Managing Editor, the
journal was relaunched in 2015 with Taylor & Francis as the publisher, and
it's very much in need of article submissions.

[Arnaud Fournet]

hm!? I was not aware of this interruption. What happened?
A.

[Peter T. Daniels]

I wouldn't expect you to regularly consult WORD. It's not a matter for public discussion.

[LingualNoob]

Here's one that compares attested forms (one of which is Arabic rather than Hebrew/Aramaic), but which sounds like it fits very well phonetically and semantically:

(87) Arabic ʕgz / ʕagaza ‘to age, grow old (of women)’ > Tr wegaca- 'grow old (of women)’

Two of Stubbs' postulated regular correspondences that seem relevant (pasted in full this time):

(Proto-Semitic *z > c (ts) in UA)

(The Semitic voiced pharyngeal ʕ > UA w/o/u, i.e., some form of rounding, as the Phoenician ʕ symbol > Greek o)

[Peter T. Daniels]

On Tuesday, August 18, 2015 at 8:29:20 AM UTC-4, LingualNoob wrote:

> Here's one that compares attested forms (one of which is Arabic rather than Hebrew/Aramaic), but which sounds like it fits very well phonetically and semantically:

So now you have Arabs joining in this vast pilgrimage? With such large
populations making the trek you'd think there would be _some_ archeological
evidence somewhere in Asia or America. Not to mention similar "linguistic
evidence" in the languages of all the peoples they happened to run into on
the way.

> (87) Arabic ʕgz / ʕagaza ‘to age, grow old (of women)’ > Tr wegaca- 'grow old (of women)’
>
> Two of Stubbs' postulated regular correspondences that seem relevant (pasted in full this time):
>
> (Proto-Semitic *z > c (ts) in UA)
>
> (The Semitic voiced pharyngeal ʕ > UA w/o/u, i.e., some form of rounding, as the Phoenician ʕ symbol > Greek o)

Drop the "as" clause. Rounding isn't involved in the coloring of a to o by
`ayn, but backing.

[LingualNoob]

On Tuesday, August 18, 2015 at 7:40:38 AM UTC-5, Peter T. Daniels wrote:
> On Tuesday, August 18, 2015 at 8:29:20 AM UTC-4, LingualNoob wrote:
>
> > Here's one that compares attested forms (one of which is Arabic rather than Hebrew/Aramaic), but which sounds like it fits very well phonetically and semantically:
>
> So now you have Arabs joining in this vast pilgrimage? With such large
> populations making the trek you'd think there would be _some_ archeological
> evidence somewhere in Asia or America. Not to mention similar "linguistic
> evidence" in the languages of all the peoples they happened to run into on
> the way.
>

The difficulty I see with this example is different than the difficulty you are suggesting. Stubbs addresses your point in the paper we're discussing (p13) saying:

"Some may question citing cognate forms from various Semitic languages instead of only one, but it is quite acceptable. For example, what we have of Classical Hebrew vocabulary in existing texts is but a fraction of what existed in the spoken dialect(s); so when a match with the expected Hebrew reflex of an existing Arabic form is found, for example, there is little reason to doubt its existence in the ancient spoken cognate language Hebrew. In fact, that is what the philologists who compiled the Hebrew lexicons have always done: validate the Hebrew terms based on cognate terms. There is no word for squirrel in the Hebrew Old Testament, yet two Arabic words for squirrel are in UA, whose sound correspondences match unattested Hebrew cognates: e.g. (57) Arabic singaab = Hebrew *siggoob > UA sikkuC ‘squirrel’. Another example is Semitic *km’ ‘truffle’ (575) found in both Arabic to the south and Ugaritic (of Northwest Semitic) to the north, so the term’s existence in Hebrew, located between the two, would be likely, even though Old Testament authors had no occasion to talk about truffles either."

I expect that your response to Stubbs' statement will be something like "he is just doing a bunch of hand-waving in order to make an excuse to broaden his source lexicon pool". That's something I've considered but when I look into it I find that Stubbs is hardly the only linguist to use Arabic terms to postulate that a similar term might be found in other Semitic languages. For example, after you questioned this one today I did some googling and quickly found (from searching the semitic database for the meaning 'old' on the website http://starling.rinet.ru/babel.php?lan=en):

Number: 24
Proto-Semitic: *ʕagūz-
Meaning: 'old woman'
Arabic: ʕagūz-
Notes: Cf. Arab ʕgz [-u-] 'become old (of a woman)'.

...I'm not citing this unreferenced database entry from a website in hopes of convincing you that *ʕagūz- 'old woman' is a Proto-Semitic form. I'm simply saying that stuff like this is pretty easy to find and shows me that Stubbs' statement about the utilization of languages like Arabic seems valid on the surface because many other linguists seem to do the same thing. Am I sure that his use of Arabic words is appropriate? No, but I'm not having heartburn about it based on what I'm seeing so far.

On the other hand, I was expecting you to downplay this one based on Trask's term "Reaching Down". The following statement from Campbell & Poser (2008) is what I see as problematic for this example:

"Related to the appeal to non-cognate forms from a language family is the tendency for distant genetic relationship enthusiasts to compare a word from but one language (or a very few languages) of one family with some word thought to be similar in one (or a few) languages in some other family – Larry Trask (1999) called this “reaching down.” A form which has a clearly established etymology in its own family by virtue of having cognates in a number of sister languages stands a better chance of perhaps having cognate associations with words of languages that may be even more remotely related than some isolated form does in some language which has no known cognates elsewhere within its family and hence no prima facie evidence of potential greater age. Inspectionally resemblant lexical sets of this sort are not convincing." (p208)

What I was hoping for was that you might be familiar with related forms because this seems like it would be a strong example if it showed a broader distribution. If not, then I tend to add this one to the pool of "cool, but not convincing by itself".

[Peter T. Daniels]

On Tuesday, August 18, 2015 at 11:38:47 AM UTC-4, LingualNoob wrote:
> On Tuesday, August 18, 2015 at 7:40:38 AM UTC-5, Peter T. Daniels wrote:
> > On Tuesday, August 18, 2015 at 8:29:20 AM UTC-4, LingualNoob wrote:

> > > Here's one that compares attested forms (one of which is Arabic rather than Hebrew/Aramaic), but which sounds like it fits very well phonetically and semantically:
> > So now you have Arabs joining in this vast pilgrimage? With such large
> > populations making the trek you'd think there would be _some_ archeological
> > evidence somewhere in Asia or America. Not to mention similar "linguistic
> > evidence" in the languages of all the peoples they happened to run into on
> > the way.
>
> The difficulty I see with this example is different than the difficulty you are suggesting. Stubbs addresses your point in the paper we're discussing (p13) saying:
>
> "Some may question citing cognate forms from various Semitic languages instead of only one, but it is quite acceptable. For example, what we have of Classical Hebrew vocabulary in existing texts is but a fraction of what existed in the spoken dialect(s); so when a match with the expected Hebrew reflex of an existing Arabic form is found, for example, there is little reason to doubt its existence in the ancient spoken cognate language Hebrew. In fact, that is what the philologists who compiled the Hebrew lexicons have always done: validate the Hebrew terms based on cognate terms. There is no word for squirrel in the Hebrew Old Testament, yet two Arabic words for squirrel are in UA, whose sound correspondences match unattested Hebrew cognates: e.g. (57) Arabic singaab = Hebrew *siggoob > UA sikkuC ‘squirrel’. Another example is Semitic *km’ ‘truffle’ (575) found in both Arabic to the south and Ugaritic (of Northwest Semitic) to the north, so the term’s existence in Hebrew, located between the two, would be likely, even though Old Testament authors had no occasion to talk about truffles either."

A problem with both these examples is that they refer to critters of
restricted geographic range. I know what a truffle is in English -- a
fungus that grows underground and is hunted using pigs -- but it seems
unlikely that such a thing was known in the Semitic-speaking area; nor
is it likely that an alien animal-name would be applied to a local small
mammal, unless the Arabs brought their pet singaabs with them.

> I expect that your response to Stubbs' statement will be something like "he is just doing a bunch of hand-waving in order to make an excuse to broaden his source lexicon pool". That's something I've considered but when I look into it I find that Stubbs is hardly the only linguist to use Arabic terms to postulate that a similar term might be found in other Semitic languages. For example, after you questioned this one today I did some googling and quickly found (from searching the semitic database for the meaning 'old' on the website http://starling.rinet.ru/babel.php?lan=en):
>
> Number: 24
> Proto-Semitic: *ʕagūz-
> Meaning: 'old woman'
> Arabic: ʕagūz-
> Notes: Cf. Arab ʕgz [-u-] 'become old (of a woman)'.

If they don't have it in any other language, then they can't trace it to PSem.

> ...I'm not citing this unreferenced database entry from a website in hopes of convincing you that *ʕagūz- 'old woman' is a Proto-Semitic form. I'm simply saying that stuff like this is pretty easy to find and shows me that Stubbs' statement about the utilization of languages like Arabic seems valid on the surface because many other linguists seem to do the same thing. Am I sure that his use of Arabic words is appropriate? No, but I'm not having heartburn about it based on what I'm seeing so far.
>
> On the other hand, I was expecting you to downplay this one based on Trask's term "Reaching Down". The following statement from Campbell & Poser (2008) is what I see as problematic for this example:
>
> "Related to the appeal to non-cognate forms from a language family is the tendency for distant genetic relationship enthusiasts to compare a word from but one language (or a very few languages) of one family with some word thought to be similar in one (or a few) languages in some other family – Larry Trask (1999) called this “reaching down.” A form which has a clearly established etymology in its own family by virtue of having cognates in a number of sister languages stands a better chance of perhaps having cognate associations with words of languages that may be even more remotely related than some isolated form does in some language which has no known cognates elsewhere within its family and hence no prima facie evidence of potential greater age. Inspectionally resemblant lexical sets of this sort are not convincing." (p208)
>
> What I was hoping for was that you might be familiar with related forms because this seems like it would be a strong example if it showed a broader distribution. If not, then I tend to add this one to the pool of "cool, but not convincing by itself".

I myself am not a Semitic lexicon, and the David Cohen Dictionnaire des
racines semitiques hasn't gotten near `ayn yet; the closest we have to
a Semitic etymological dictionary is Sokoloff's reedition of Brockelmann's
Lexicon syriacum (in English!). The word isn't found in Syriac.

[DKleinecke]

The bottom line is that long range comparisons like Stubb's and Arnaud's
are not viewed sympatheticially by mainline historical linguists. There is
a minority group (like Arnaud) that indulges in them. Stubb's article is
eccentric enough that it will be dismissed out of hand by "experts". A
position that "the experts are wrong and I am right" is the hallmark of
crackpots. Basically Stubbs needs to read the mainline literature and
conform more to its style of reasoning - if his thesis does indeed survive
such a test.

In the case of Mormon history I don't know how to handle the temporal
problem. But briefly - since 2600 BP is far too short a time to squeeze
in all the development of UA. According to Wikipedia proto-UA dates at
about 7000 BP. I forbear mentioning the geographic problems.

[Yusuf B Gursey]

In <0e314adb-3121-4d6b...@googlegroups.com>, DKleinecke

There are (or at least were) many in Russia. Karl H. Menges, the late
Turkologist and a serious linguist, wrote many such articles and I had
heard complaints that he allowed too much of other people's work on the
subject be published in Central Asiatic Journal towards the end of his
life. He took Dumezil's Turkic - Quechua comparisions seriously as
indicated in a line in "Turkic Languages and Peoples." Denis Sinor also
did such work, though many of his comparisions involve such things as
animal names and they are treated as loanwords.

I understand the difficulties involved and I appreciate the need for
caution, but at the same time I don't think one should entirley give up
on such attempts and I enjoy reading serious work along those lines.

[LingualNoob]

On Tuesday, August 18, 2015 at 2:15:51 PM UTC-5, DKleinecke wrote:
> The bottom line is that long range comparisons like Stubb's and Arnaud's
> are not viewed sympatheticially by mainline historical linguists. There is
> a minority group (like Arnaud) that indulges in them. Stubb's article is
> eccentric enough that it will be dismissed out of hand by "experts". A
> position that "the experts are wrong and I am right" is the hallmark of
> crackpots. Basically Stubbs needs to read the mainline literature and
> conform more to its style of reasoning - if his thesis does indeed survive

> such a test...

As Stubbs said in his paper, the whole proposal he is about to present includes 1500+ examples like we've been discussing with many more supporting details for each example. His exact Uto-Aztecan reconstructions are already published in his book "Uto-Aztecan: A Comparative Vocabulary" (Stubbs, 2011) (I don't have a copy of the Comparative Vocabulary. His proposal includes enough details about each reconstruction that I've been satisfied with it alone).

Campbell and other prominent Uto-Aztecanists use content from Stubbs comparative vocabulary book in their recent publications and commonly cite content from his previous publications as well.

Your following statement is definitely true: "long range comparisons like Stubb's...are not viewed sympatheticially by mainline historical linguists"

But I take exception to this statement: "Stubbs needs to read the mainline literature and conform more to its style of reasoning"

I've spent a lot of time not only pouring over the examples in the book, but also reading up on comparative linguistics and trying to see where the proposal could be fudging things. I've especially focused on Campbell's books since he is both an Uto-Aztecanist and an expert in Historical Linguistics and has devoted considerable effort into defining the requirements that should be met by anyone proposing long-distant language relationships.

I can see some things here and there that don't fall perfectly in-line with Campbell's advice (like the "reaching down" I mentioned in the last example), but even after I set aside the examples that I think might be questionable, the case he makes still seems overwhelmingly convincing.

...I know, I'm not a linguist and I'm definitely still an amateur, but I have to say that if the linguistic evidence in this proposal is obviously wrong or presented outside of acceptable standards, then historical linguists need to do a much better job of explaining how to present a long-range proposal.

It's not like he's doing what Greenberg did. Stubbs proposes a large system of regular sound correspondences and his comparisons aren't just a bunch of CV or CVC vagueries. While there are examples of questionable semantic leeway, they are few compared to the number of examples of precise semantics.

I wish I could quote from his book already but I'm stuck having to simply say that in addition to the 1500+ lexical comparisons (many of which show impressive submerged features), the proposal lives up to the other claims made in the paper, including:

* "The UA basic vocabulary from Egyptian and Semitic are numerous: body parts, plant and animal terms, nouns of nature (sun, moon, star, sky, rock, water, etc.; see 7.1 in the book.)"

* "A considerable amount of Semitic morphology or fossilized items of Semitic verb conjugations"

* "many unusual semantic combinations in Semitic and Egyptian are preserved in the corresponding UA meanings"

* "The author's book, Uto-Actecan: A Comparative Vocabulary (2011), includes some 2700 Uto-Aztecan cognate sets. Those with substantial similarity to Semitic or Egyptian, and according to the proposed sound correspondences, are about 30%."

* "among the 2700 Uto-Aztecan cognate sets, the vast majority of those sets have cognates or reflexes, that is, descendant words in less than half of the 30 UA languages. Only 45 cognate sets have reflexes in nearly all 30 UA languages or appear in all 8 of the 8 UA branches. Yet 26 of those 45 sets appear in the Near-East sets (see 7.7 in the book). That amounts to about 60% of the widespread UA words."

* "the Semitic or Egyptian forms proposed to underlie the UA forms often answer questions and explain puzzles in UA that Uto-Aztecanists have not yet been able to explain"

* Related to the above quote^ he then mentions some items in the paper that the book presents very convincingly. For example, the paper simply says: "Uto-Aztecanists agree on each UA language's reflex that corresponds to PUA *p. However, five UA languages--Tarahumara, Mayo, Yaqui, Arizona Yaqui, and Eudeve--show both initial b and p corresponding to PUA *p. This split is usually ignored as an inconvenient inconsistency in these languages. However, the initial b forms in these languages correspond to Egyptian b or Semitic b of p-NWSemitic, and the initial p forms to Semitic/Egyptian p..."

* The paper also mentions several other examples of UA problems that are solved by recognizing the Semitic/Egyptian ties he presents. In the full proposal the examples he is referring to seem very impressive.

I know you haven't seen the whole proposal yet, but when you do I believe you'll see that, whether you agree with his conclusions or not, he does do an impressive job of presenting his proposal within the Comparative Method and within the professional standards defined by folks like Campbell.

[LingualNoob]

On Tuesday, August 18, 2015 at 2:15:51 PM UTC-5, DKleinecke wrote:

> In the case of Mormon history I don't know how to handle the temporal
> problem. But briefly - since 2600 BP is far too short a time to squeeze
> in all the development of UA. According to Wikipedia proto-UA dates at
> about 7000 BP. I forbear mentioning the geographic problems.

Yes, the glottochronological estimates that I've seen don't fit the timeframe for Book of Mormon cultures. I can speculate about the disparity, but I doubt that such speculation would be convincing. There's also the possibility that the proposal is substantially correct in some aspects and incorrect in others. Maybe the answer is that Stubbs has --only-- discovered that UA is related to Afro-Asiatic in some way. I think it's best to just let the language evidence that he's proposing lead the way. Maybe it will confirm my biases, maybe not. In any case, I do hope that such a carefully prepared proposal won't simply be ignored by his peers.

[Yusuf B Gursey]

In <5586aca1-3cb0-4ef1...@googlegroups.com>, LingualNoob

wrote on 8/18/2015:
> Here's one that compares attested forms (one of which is Arabic rather than
> Hebrew/Aramaic), but which sounds like it fits very well phonetically and
> semantically:
>
> (87) Arabic ʕgz / ʕagaza ‘to age, grow old (of women)’ > Tr wegaca- 'grow old
> (of women)’
>

ʕagaza aorist yaʕgizu is "to be powerless" (said to be from an earlier
meaning of being left behind)

ʕagaza aorist yaʕguzu is "to grow old" (of women) which is derivative
of the earlier meaning.

> Two of Stubbs' postulated regular correspondences that seem relevant (pasted
> in full this time):
>
> (Proto-Semitic *z > c (ts) in UA)
> (The Semitic voiced pharyngeal ʕ > UA w/o/u, i.e., some form of rounding, as
> the Phoenician ʕ symbol > Greek o)

/ʕ/ is a consonant, which Greek lacks, so it was left available to
represent Greek /o/. Rounding does not follow /ʕ/, which is similar to
the sound you make during mild retching.

[Yusuf B Gursey]

In <31c68ee0-1780-410f...@googlegroups.com>, Peter T.

Likely you can't, since it seems to be a derivative meaning from
"powerless, helpless" as acknowledged by Arab philologists.

[Yusuf B Gursey]

In <f4e0806c-e22f-430a...@googlegroups.com>, LingualNoob

Yes, Starotsin's legacy. Some of the reconstructions of this group have
been criticized, as I know from those made of its alledged "Altaic"
reconstructions.

I don't want ot knock it down entirely, as I myself consult it on
occasion.

[LingualNoob]

On Tuesday, August 18, 2015 at 5:07:08 PM UTC-5, Yusuf B Gursey wrote:
> In <5586aca1-3cb0-4ef1...@googlegroups.com>, LingualNoob
> wrote on 8/18/2015:
> > Here's one that compares attested forms (one of which is Arabic rather than
> > Hebrew/Aramaic), but which sounds like it fits very well phonetically and
> > semantically:
> >
> > (87) Arabic ʕgz / ʕagaza ‘to age, grow old (of women)’ > Tr wegaca- 'grow old
> > (of women)’
> >
>
> ʕagaza aorist yaʕgizu is "to be powerless" (said to be from an earlier
> meaning of being left behind)
>
> ʕagaza aorist yaʕguzu is "to grow old" (of women) which is derivative
> of the earlier meaning.
>

Do you have an idea of when it derived?

[Yusuf B Gursey]

In <e62bbac0-efeb-4ee2...@googlegroups.com>, LingualNoob

wrote on 8/19/2015:
> On Tuesday, August 18, 2015 at 5:07:08 PM UTC-5, Yusuf B Gursey wrote:
>> In <5586aca1-3cb0-4ef1...@googlegroups.com>, LingualNoob
>> wrote on 8/18/2015:
>>> Here's one that compares attested forms (one of which is Arabic rather than
>>> Hebrew/Aramaic), but which sounds like it fits very well phonetically and
>>> semantically:
>>>
>>> (87) Arabic ʕgz / ʕagaza ‘to age, grow old (of women)’ > Tr wegaca- 'grow
>>> old (of women)’
>>>
>>
>> ʕagaza aorist yaʕgizu is "to be powerless" (said to be from an earlier
>> meaning of being left behind)
>>
>> ʕagaza aorist yaʕguzu is "to grow old" (of women) which is derivative
>> of the earlier meaning.
>>
>
> Do you have an idea of when it derived?

No. But you have comparisions with NW Semitic languages and since it
does not appear in them it is after Proto-Arabic split from them.

[Arnaud Fournet]

Le mardi 18 août 2015 21:52:36 UTC+2, LingualNoob a écrit :
> On Tuesday, August 18, 2015 at 2:15:51 PM UTC-5, DKleinecke wrote:
> > The bottom line is that long range comparisons like Stubb's and Arnaud's
> > are not viewed sympatheticially by mainline historical linguists. There is
> > a minority group (like Arnaud) that indulges in them. Stubb's article is
> > eccentric enough that it will be dismissed out of hand by "experts". A
> > position that "the experts are wrong and I am right" is the hallmark of
> > crackpots. Basically Stubbs needs to read the mainline literature and
> > conform more to its style of reasoning - if his thesis does indeed survive
> > such a test...
>
> As Stubbs said in his paper, the whole proposal he is about to present includes 1500+ examples like we've been discussing with many more supporting details for each example. His exact Uto-Aztecan reconstructions are already published in his book "Uto-Aztecan: A Comparative Vocabulary" (Stubbs, 2011) (I don't have a copy of the Comparative Vocabulary. His proposal includes enough details about each reconstruction that I've been satisfied with it alone).

Has this book been peer-reviewed, before publishing?
How much reliability does it have?
A.

>
> Campbell and other prominent Uto-Aztecanists use content from Stubbs comparative vocabulary book in their recent publications and commonly cite content from his previous publications as well.
>
> Your following statement is definitely true: "long range comparisons like Stubb's...are not viewed sympatheticially by mainline historical linguists"
>
> But I take exception to this statement: "Stubbs needs to read the mainline literature and conform more to its style of reasoning"
>
> I've spent a lot of time not only pouring over the examples in the book, but also reading up on comparative linguistics and trying to see where the proposal could be fudging things. I've especially focused on Campbell's books since he is both an Uto-Aztecanist and an expert in Historical Linguistics and has devoted considerable effort into defining the requirements that should be met by anyone proposing long-distant language relationships.
>
> I can see some things here and there that don't fall perfectly in-line with Campbell's advice (like the "reaching down" I mentioned in the last example), but even after I set aside the examples that I think might be questionable, the case he makes still seems overwhelmingly convincing.
>
> ...I know, I'm not a linguist and I'm definitely still an amateur, but I have to say that if the linguistic evidence in this proposal is obviously wrong or presented outside of acceptable standards, then historical linguists need to do a much better job of explaining how to present a long-range proposal.

Basically, there's no difference between a long-range proposal and a short-range proposal. You need to collect sets of cognate words, that can be predictably derived from proto-forms according to regular and recurrent sound-laws. And the proto-language and the sound-laws should be acceptable, i.e they should not be odd to the point of looking ad-hoc.

the word "long-range" itself is in fact an insult. It means that the proposal is unacceptable in the first place.
Same thing with the word "macro-comparative", these words are anathemas.
A.

[Arnaud Fournet]

I reject the word "long range", which is just a word of abuse.
What I try to do is just comparative work, nothing else.
A.

> There is
> a minority group (like Arnaud) that indulges in them.

yes, because I think the endeavor is possible.
A.

> Stubb's article is
> eccentric enough that it will be dismissed out of hand by "experts".

it's clear that a direct comparison between UA and NW Semitic will be rejected as absurd, by about everybody. This idea is suffering from overstretch.
It would make more sense to explain which Amerindian families are likely relatives of UA and which are not.
A.

> A
> position that "the experts are wrong and I am right" is the hallmark of
> crackpots. Basically Stubbs needs to read the mainline literature and
> conform more to its style of reasoning - if his thesis does indeed survive
> such a test.

The "experts" are wrong, when they claim that macro-comparative work is impossible. I agree that in the recent past, people involved in that field have done a great job ruining its reputation among academic circles.
But I'm quite confident this reputation can be restored, if good work is published. That's what I'm trying to do.
A.

[Arnaud Fournet]

Le lundi 17 août 2015 19:04:25 UTC+2, Arnaud Fournet a écrit :
> Le lundi 17 août 2015 12:54:25 UTC+2, Peter T. Daniels a écrit :
> > On Monday, August 17, 2015 at 4:45:35 AM UTC-4, Arnaud Fournet wrote:
> > > Le dimanche 16 août 2015 14:00:17 UTC+2, Peter T. Daniels a écrit :
> > > > On Sunday, August 16, 2015 at 1:39:51 AM UTC-4, Arnaud Fournet wrote:
> > > > > Le samedi 15 août 2015 14:18:10 UTC+2, Peter T. Daniels a écrit :
> > > > > > On Saturday, August 15, 2015 at 3:22:10 AM UTC-4, Arnaud Fournet wrote:
> >
> > > > Which may well be what the readers for IJAL will say about the article you
> > > > submitted to them.
> > >
> > > I have found Uralic-looking material in Mayan languages as well.
> > > A.
> >
> > That puts you right in Lyle Campbell's wheelhouse, since those are his two
> > areas of specialization -- and one of the pairings for which he drew up a list
> > of deceptive-looking chance similarities.
>
> This morning, I found that quite a lot of Quechua words look Uralic.
> This is interesting as Quechua has the same system as Na-Dene: voiceless - glottalized - aspirated.
> So if my cognates have the same consonants in Na-Dene and Quechua, then it's bingo.
> A.

It works !

The extremely interesting point about Quechua is that it has the same k / q and voiceless / glottalized / aspirated distinctions as Na-Dene and especially Eyak.
So here we go:

1. Aspirated uvular *qh
Eyak qha "to bite"
Quechua qhatki "bitter"
Finnish katkera "bitter"
Besides, Quechua also has qhamsay ‘to bite’.
Mongolian *qaža- ‘to bite’

2. Plain voiceless uvular *q
Eyak qhahtL "(tree) bark"
Quechua qara "bark, skin"
Finnish kärnä "bark" (funny ä is long a: from *ah)
Now you may object that Eyak has qh instead of q, but it seems that Eyak spreads aspiration onto the initial. So that the root is *qah- with *q and not *qh.
Yukaghir qār ‘(tree) bark’ (Nikolayeva 2006:379) has the right length: long a:
Besides, Mongolian qajir(a)-su(n) ‘scales, hard bark’ with *qaj < *qah

Another item is *qo!- "to cover"
Eyak qa!- "covered with"
Quechua qata "roof ; also cover, bedding, coverage"
Uralic *kota "house"
Besides Mongolian *qota "town, group of huts"

Another item is *qi: "moon"
Eyak qi: "new moon"
Uralic *kiiNi (with *i: and not with the bad *u)
Quechua killa "moon" (double ll as in Spanish =ly)
Apparently Quechua changes *qi into *ki as in Yukaghir.
Not in Mongolian.

3. Glottalized uvular *q!
Eyak q!a!c "hand"
Uralic *käte "hand", *käme "palm of hand"
Funny ä stands for *a: < *a+glottal stop
Quechua *q!apa "palm of hand"
Mongolian *gar (written mongolian qa(:)r) "hand, arm"

I've not looked at the velar series k kh and k! yet.
I suppose it should work as well.

Comments are welcome.
A.

[Peter T. Daniels]

On Wednesday, August 19, 2015 at 4:31:11 AM UTC-4, Arnaud Fournet wrote:
> Le lundi 17 août 2015 19:04:25 UTC+2, Arnaud Fournet a écrit :

> > This morning, I found that quite a lot of Quechua words look Uralic.
> > This is interesting as Quechua has the same system as Na-Dene: voiceless - glottalized - aspirated.

That's hardly unusual among the languages of the world.

> > So if my cognates have the same consonants in Na-Dene and Quechua, then it's bingo.

No. It's a common typological feature. Have you used Hockett, *Manual of
Phonology* (1955)?

> It works !
>
> The extremely interesting point about Quechua is that it has the same k / q and voiceless / glottalized / aspirated distinctions as Na-Dene and especially Eyak.
> So here we go:

All of this means nothing unless you have regular patterns of correspondence
across numerous complete lexical items.

Show a dozen more items for each correspondence.

AIUI Uralic vowels can't be reconstructed precisely, because vowel harmony
makes it impossible to discover a specific base form.

[LingualNoob]

Perhaps this belongs in a separate thread?

[LingualNoob]

On Tuesday, August 18, 2015 at 11:09:37 PM UTC-5, Yusuf B Gursey wrote:
> In <e62bbac0-efeb-4ee2...@googlegroups.com>, LingualNoob
> wrote on 8/19/2015:
> > On Tuesday, August 18, 2015 at 5:07:08 PM UTC-5, Yusuf B Gursey wrote:
> >> In <5586aca1-3cb0-4ef1...@googlegroups.com>, LingualNoob
> >> wrote on 8/18/2015:
> >>> Here's one that compares attested forms (one of which is Arabic rather than
> >>> Hebrew/Aramaic), but which sounds like it fits very well phonetically and
> >>> semantically:
> >>>
> >>> (87) Arabic ʕgz / ʕagaza ‘to age, grow old (of women)’ > Tr wegaca- 'grow
> >>> old (of women)’
> >>>
> >>
> >> ʕagaza aorist yaʕgizu is "to be powerless" (said to be from an earlier
> >> meaning of being left behind)
> >>
> >> ʕagaza aorist yaʕguzu is "to grow old" (of women) which is derivative
> >> of the earlier meaning.
> >>
> >
> > Do you have an idea of when it derived?
>
> No. But you have comparisions with NW Semitic languages and since it
> does not appear in them it is after Proto-Arabic split from them.
>

Thanks. I appreciate your help. I'm writing this example off.

[LingualNoob]

On Wednesday, August 19, 2015 at 12:32:50 AM UTC-5, Arnaud Fournet wrote:
> Le mardi 18 août 2015 21:52:36 UTC+2, LingualNoob a écrit :
> > On Tuesday, August 18, 2015 at 2:15:51 PM UTC-5, DKleinecke wrote:
> > > The bottom line is that long range comparisons like Stubb's and Arnaud's
> > > are not viewed sympatheticially by mainline historical linguists. There is
> > > a minority group (like Arnaud) that indulges in them. Stubb's article is
> > > eccentric enough that it will be dismissed out of hand by "experts". A
> > > position that "the experts are wrong and I am right" is the hallmark of
> > > crackpots. Basically Stubbs needs to read the mainline literature and
> > > conform more to its style of reasoning - if his thesis does indeed survive
> > > such a test...
> >
> > As Stubbs said in his paper, the whole proposal he is about to present includes 1500+ examples like we've been discussing with many more supporting details for each example. His exact Uto-Aztecan reconstructions are already published in his book "Uto-Aztecan: A Comparative Vocabulary" (Stubbs, 2011) (I don't have a copy of the Comparative Vocabulary. His proposal includes enough details about each reconstruction that I've been satisfied with it alone).
>
> Has this book been peer-reviewed, before publishing?
>

Kenneth Hill apparently reviewed it in the International Journal of American Linguistics. When I've tried to find and read the review I find a JSTOR link to it but JSTOR's website has an error that keeps redirecting your browser if you try to visit the link so I've never been able to actually read the review.

Stubbs says that his Uto-Aztecan Comparative Vocabulary was "favorably received among Uto-Aztecanists, though no two Uto-Aztecan specialists will agree on all aspects and reconstructions, as Kenneth Hill notes in a favorable review in the International Journal of American Linguistics (Hill 2012), and after any linguistic comparative work, adjustments inevitably follow."

>
> How much reliability does it have?
>

While I can't comment on it's reliability myself, I've seen it cited by several other Uto-Aztecanists and Lyle Campbell uses it as the basis for exercises in his new textbook "Historical Linguistics" (2013).

>
> A.
>
> >
> > Campbell and other prominent Uto-Aztecanists use content from Stubbs comparative vocabulary book in their recent publications and commonly cite content from his previous publications as well.
> >
> > Your following statement is definitely true: "long range comparisons like Stubb's...are not viewed sympatheticially by mainline historical linguists"
> >
> > But I take exception to this statement: "Stubbs needs to read the mainline literature and conform more to its style of reasoning"
> >
> > I've spent a lot of time not only pouring over the examples in the book, but also reading up on comparative linguistics and trying to see where the proposal could be fudging things. I've especially focused on Campbell's books since he is both an Uto-Aztecanist and an expert in Historical Linguistics and has devoted considerable effort into defining the requirements that should be met by anyone proposing long-distant language relationships.
> >
> > I can see some things here and there that don't fall perfectly in-line with Campbell's advice (like the "reaching down" I mentioned in the last example), but even after I set aside the examples that I think might be questionable, the case he makes still seems overwhelmingly convincing.
> >
> > ...I know, I'm not a linguist and I'm definitely still an amateur, but I have to say that if the linguistic evidence in this proposal is obviously wrong or presented outside of acceptable standards, then historical linguists need to do a much better job of explaining how to present a long-range proposal.
>
> Basically, there's no difference between a long-range proposal and a short-range proposal. You need to collect sets of cognate words, that can be
>

This is wishful thinking. As much as I hope that Stubbs' long-range proposal is accurate and gets taken seriously by professionals, any long-range proposal needs to meet a higher bar than a short-range proposal if it wants to get anyone's attention.

>
predictably derived from proto-forms according to regular and recurrent sound-laws. And the proto-language and the sound-laws should be acceptable, i.e they should not be odd to the point of looking ad-hoc.
>

We don't get to decide what changes happened to languages in the past. Changes have taken place in languages that seem very strange. I'm sure it's easier to convince linguists on seemingly-natural sound-law changes, but you can't simply -decide- that a language could only have changed in ways you predetermine to be 'acceptable'. A better approach would be to recognize that if you are proposing a regular correspondence that does not seem natural that you need to both try to explain it and show a large amount of evidence in support of it.

...at least that's my take-away from what I've read.

>
> the word "long-range" itself is in fact an insult. It means that the proposal is unacceptable in the first place.
>

That's just not true. It's not insulting. It's a recognition of the fact that linguists have had to deal with a lot of bogus proposals that share one similar trait: They propose long-distance language relationships. "Long-range" is a generalization, but it is not an insult.

What should we expect? That linguists would ignore every previous attempt to tie together old-world and new-world languages? If that's what you think then you're saying that their previous work successfully debunking such proposals should be ignored. We need to recognize that any long-range proposal has a VERY high bar to reach. It's a reality.

Take the Clovis-first debate as an example. For decades, Clovis-first was the revered standard for colonization of the Americas. During that time a huge number of pseudo-scientific proposals were put forward attempting to prove Clovis-first wrong. Interspersed in the dizzying array of fringe theories was some very good science (i.e. Monte Verde) showing that Clovis-first was wrong. The authors of the good science had to put up with a lot of insults from their peers, but were eventually vindicated. The thing to recognize is that the vindication of the good pre-Clovis science didn't vindicate all the pseudo-pre-Clovis science.

People blame the scientists that held onto the Clovis-first theory for not being open-minded, but that's just not true. They were in a situation where they were being bombarded by pseudo-science. It was a signal-to-noise ration problem. There was so much pseudo-science 'noise', that actual scientific signals couldn't be heard until they were so strong that they could overcome the noise. Once the good science was able to make a strong enough case, Clovis-first was finally accepted as wrong.

Although individual linguists will choose to insult those who throw their hat into the long-range ring, the term itself is not an insult. It's a recognition that we're surrounded by a bunch of noise.

People will use this situation to insult you. What you need to learn to do is not to bite when they bait you. An example of this was Peter's jabs about your paper not being acceptable for publication. I have no idea if your papers are good or not, but that's beside the point. His purpose for writing that was not to help you, it was to make you angry. It's the same thing when he talks about what he calls the "magic carpet" back-story to Stubbs' proposal.

He will have his fun at your expense. There's nothing you can do about that. If you think your ideas have merit then ignore the jabs that people make.