Re: swallow vs. nightingale

--- In cybalist@yahoogroups.com, "Grzegorz Jagodzinski"
<grzegorj2000@...> wrote:

>
> ---- Original Message ----
> From: tgpedersen
> To: cybalist@yahoogroups.com
> Sent: Wednesday, November 14, 2007 3:11 AM
> Subject: [tied] Re: swallow vs. nighingale
>
> >>>> A short "leap aside", if you please. I perfectly know that
> >>>> this is a list on IE, not Nostratic, but I only want to show
> >>>> several examples of irregular development of initial groups
> >>>> which is the subject here. This irregular development is
> >>>> documented well in Altaic, and especially in Mongolian. Only
> >>>> some examples:
>
> >>>> u ~ 0: Mongolian usu - Turkish su (< sub) "water" (cf. also
> >>>> Buriat dialectal (so: Mongolian) hub ~ sub "river..."
> >>>> (adjective))
>
> [...]
>
> >>>> There exist much more such examples. Naturally such
> >>>> disappearings of initial sounds are not regular in any of
> >>>> cited languages.
>
> >>> Why 'naturally'?
>
> >> Dear Torsten, I wrote "naturally" because I think so. And I think
> >> so because of the knowledge I have. I really hope that you do not
> >> understand why "naturally", and that your question really was not
> >> to be provocative. Sorry, I do not like quarrels. We may be of
> >> different opinions but it does not means that we must provocate
> >> one another.
>
> >> Anyway, I would never ask such a question because the situation
> >> seems to me clear. Obviously not to you.
>
> > That's right. That's why I asked you the question and why you
> > didn't ask yourself the question.
>
> >> "Naturally" means that disappearing of initial sounds is not
> >> regular in Altaic, however they occur in some words.
>
> > No it doesn't.
>
> > The latter statement states as a contingent fact, that
> > disappearing of initial sounds is not regular in Altaic, however
> > they occur in some words. But if you add 'naturally' in front of
> > it, you are saying that this fact is derived from something else,
> > some higher principle which you don't mention.
>
> >> It was a notice for those who have little knowledge on Altaic. If
> >> disappearing had been regular, I would not have put it as the
> >> argument. And if anyone else had put such examples as an
> >> argument, I would have guessed that the examples must illustrate
> >> some irregular changes. There would not have been another
> >> possibility unless my opponent had been schizophrenic or he had
> >> not thought logically.
>
> > That's your reason for putting statement 2). I still don't know
> > what your reason was for preceding it with 1) 'naturally'.
>
> The main subject of this discussion is not the meaning of the word
> "naturally" but the problem if there exist irregular phonetic
> changes or not. And really, boring of side threads instead of the
> main one might even make the reader confused.

OK, so you don't want to answer the question.

> Anyway, here is my explanation for you if you still have so much
> troubles with understanding my words.

I have no problem understanding your words. Occasionally I fail to
understand why you put them together the way you do.

> Webster's New Worlds Dictionary defines "naturally" as, among
> others, "as one might expect" (and not "naturally", as you writes,
> Torsten).

OK, so when you use 'naturally' it means "as one might expect" and not
"naturally".

> The discussion was on irregular phonetic changes. I gave a list of
> some phonetic changes in Altaic.

So you did.

> And I wrote that the cited chages are "naturally" irregular.

So they are, according to you, 'irregular, as one might expect'.

> The reader might expect that I had quoted them on purpose - even if
> that was not stated literally.

Most thing people say what they say on purpose. Why is that relevant here?

> Instead, I wrote "naturally"

Instead of what?

> - to confirm that the examples had really been quoted with the
> purpose to illustrate the thesis that irregular changes do exist.

In order to confirm that you had quoted those words on purpose, you
added the word 'naturally' in front of the next sentence?

> If I had omitted that word, there would not have been any relation
> between the list of quoted examples and the previous thread.

????

> I had really thought that that would be read by people who know the
> meaning of the word "naturally".

I know the meaning of the word 'naturally'. I asked you why used it in
that sentence.

> But obviously I was wrong.

No.

> You, Torsten, did not realize that the English word "naturally"
> means "as one might expect".

How does my asking you why you are placing the word 'naturally' in
front of a sentence lead you to the conclusion that I don't know what
'naturally' means?

> But now you know it, I hope.

Like before.

> Nevertheless, I think that all is clear now, and I think the topic
> is closed. EOT, like people write on popular mailing lists.

You realize you can't account for your own convoluted thought
processes, so, realizing offense is the best defense, you make a claim
that I don't understand a common English word and declare the debate
closed?

> Really, let's talk about more serious problems instead.

Yes, let's change the subject.

> >> All I wanted was to show that irregular disappearing of initial
> >> sounds is present in other languages than IE as well. If no
> >> examples were known or discussed in the literature, we should
> >> consider seriously the discussed hypothesis that some birds names
> >> in some IE languages are borrowings from some substrate.

That has no bearing on the question. We should discuss it anyway.

> > Obviously at some level you must have sensed that there was a
> > possibility you chose to ignore, namely that those words in Altaic
> > with 'irregularly' disappearing initial sounds themselves may have
> > come from a substrate in which they were regular.
>
> Not 'irregularly' but irregularly, without quotation marks. There
> are several circumstances which make the degree of probability of
> such possibility very low.
>
> 1) Various sounds disappear, both (various) vowels and consonants.
> If the discussed hesitation

I assume you mean alternation

> had been due to substrate influence, we might have expected that it
> would have been a trace of a morphological element there, with the
> function of article for instead. But such an alternating element
> should be invariable. Instead of an invariable element we observe a
> set of very miscellaneous alternations.

No, it shouldn't. The 'extra' element which occurs and doesn't occur
in the word is not necessarily an independent morpheme.
Assume some donor language has forms *aCVC and *CVC- of some class of
word, and that in the course of borrowing the rules #VCV- -> #CV- and
#CVC- -> #VC- apply. Words of that class would appear in loans in the
receiving language as #CVC- and #VC- where the first C would be
unpredictable.

> 2) Influence of substratum should be the strongest on peripheric
> area of a language group, or on territory occupied by a given
> language not very long ago. Anyway, we should observe differences in
> strength of substrate influence on the area of the group. We do not
> observe such an effect in the case of Altaic.
>
> 3) We cannot observe that longer forms would exist in one group of
> ethnolects (dialects, languages etc.) while shorter forms in another
> group of ethnolects. Instead, in a given dialects we have longer
> forms of some words and shorter forms of another words. In other
> words, the changes are not areal, and we cannot explain them by
> areal phonetic processes (either in Altaic languages or in substrate
> languages, if they were the source of these words). The only
> explanation that would make sense would be that the variable element
> was a morpheme in substrate - in some word forms it was present, and
> in some words it was absent.
>
> 4) Some of anlaut variations are specific for single dialects:
> while one dialect of a given language has the full form, another
> dialect has the shorter form (without the initial sound).

I only have your word on the subject to go by. I find it confusing
that 4) claims that some variations are related to particular
languages, in direct contradiction to 2) and 3) which claim none are.
Which one do you want me to believe?

> We cannot point what languages were the substrate - and naturally we
> cannot reject their existence in the past only because of this.

That's right.

> However, if the substrate does not exist any more now, the process
> of borrowing must have happened in the past, and rather in the
> distant past,

And there's your thumb on the scale again: past -> distant past, with
nothing proffered to support it.

> probably in the period when modern dialects and languages had not
> existed yet. A substrate language could not influence dialects of
> languages which had not formed.

That is equivalent to a claim that the word was borrowed into the
proto-language. But since the proto-language is by definition not
documented, there is no way we can prove that.

> Naturally, there is a chance that two forms, the full one and the
> shorter one, existed side by side for centuries, and finally only
> one of them has survived in one dialect and only one of them (but
> the other one) in another dialect. But the question is how much the
> chance is. Note also that we do not speak about only two dialects,
> so the chance is much time less.

Irrelevant.

> And the science differs from other forms of cognition in this,
> among others, that the science rejects little probable solutions if
> one can propose others, more probable (and simpler).

If you don't have definition of probability of a solution, that
statement is vacuous.

> The result is that, as far as I know, most Altaicists reject the
> possibility that the anlaut hesitations were caused by substrate.
> Simply, it would be too complicated explanation, based on little
> probable events. On the other hand, the possibility that initial
> sounds vanish irregularly, in various time, and in various languages
> and dialects, is more probable.

The problem with your analysis here is that you don't realize what
'irregular' means. First, it is not an absolute, but a relative term.
Items are not irregular in vacuo, but in relation to a given set of
rules. In other words, if you assume that the set of 'irregular' words
obey a different set of rules, namely those of the donor language and
of the transmission from the donor to the recipient language, then
they are no longer irregular, but regular with respect to the other
set of rules. Second, 'irregular' is not a predicate, thus not a
designation of a property, but a negation, ie the designation of the
absence of a property, namely regularity, which is designated by the
predicate 'regular'. Third, regularity is the very point of any
scientific description of an object by a hypothesis. Calling a
phenomenon 'irregular' amounts to declaration of bankruptcy: sorry,
guys, we can't explain this. I've said it before, but you don't seem
to get it.

>
> All the rest is the question of estimation of credibility of such
> or another explanation. In my opinion (like in the opinion of many
> Altaicists), the view that the hesitations of initial sounds is a
> result of irregular phonetic changes, is much better motivated than
> any other view. I understand that there are people who never admit
> the idea that irregular phonetic changes exist, and I do not hope
> that I will ever convince such a fellow believer. Discussion with
> such a person is simply a waste of time. The main point in which
> beliefs differ from science is that beliefs are not open to
> any arguments.

Of course irregularity exists, namely in those theories which fail to
explain the phenomena in question.

> Nevertheless, the discussion would have sense if it appeared that
> the above discussed circumstances are meaningless, ineffective,
> badly estimated etc.

That does not make sense. Please rephrase.

> Another reason for revision would be a piece of evidence that
> irregular changes are impossible. However, while proving of
> existence of something is enough simple, it is very hard to prove
> that something does not exist.

The problem is that you think 'irregularity' is an objectively
existing property. Then the above would make sense. It isn't, so it
doesn't.

> Of course, we may have doubts if something really exists. But it is
> more important if we have also evidence.
>
> >> But as I showed, such examples are known also outside the IE
> >> family, and the presence of "fleeting a-" cannot be taken as a
> >> serious argument for such a substrate.
>
> > If you had only cited 2), the statement about the supposed
> > irregular disappearing of initial sounds in Altaic, you would have
> > had nothing but an analogy, and that is not enough for the
> > conclusion you wanted, namely that the 'fleeting-a' is a 'regular
> > irregularity' as that of Altaic.
>
> No, "nothing but analogy" is enough. Namely, we talk about general
> laws of language development in fact. Two antagonistic views must be
> considered:
> a) irregular phonetic changes are possible
> b) irregular phonetic changes are impossible.

Once again: 'irregular' is not an absolute property, so neither
statement makes sense.

> I am convinced that the Altaic argument is enough for accepting the
> variant a, and, as I presented, there are reasons for such a
> convinction. But if yes, the presence of the a-mobile in bird names
> itself is not enough for creating hypotheses of existence a
> "bird-substrate language" in the past.

It is supporting evidence for such a hypothesis.

> It is true that the situation is not clear.

Then stop pretending it is.

> 1) Personally, I am convinced that "somebody had been here before
> us" in Europe, i.e. Indo-Europeans are not the first humans to live
> here. So, existence of a substrate language for IE languages is very
> probable in western Europe, practically certain. Which is more, I
> suspect that indeed, there existed Semitic, or at least Semitoid
> substrate or adstrate (or even superstrate) there. However, the
> anlaut alternations in two or three bird names (and in one or two
> other words) cannot be proofs for anything.

No one claimed they were, they are supporting evidence.

> 2) Both forms of a given word, the one with a- and the one without
> it, may have developed from an inherited lexeme. The cited Altaic
> examples show that irregular phonetic changes in this position
> happen to occur in languages in the world,

No. There is no such thing as an 'irregular change' in vacuo, changes
are irregular relative to a theory, ie in historical linguistics a set
of rules.

> so why not in IE. Which is more, some IE languages has preserved
> traces of initial laryngeals, in more or less regular way (not all
> Greek-specific initial vowels are from laryngeals, on the other hand
> in Armenian not all laryngeals have left traces). Another
> IE-specific feature of this type is s-mobile: its presence and
> absence seems to be quite irregular.

We don't know why it's there. Our ignorance is a property of us, not
of the data.

> There is also base (even if less certain) for supposing k-mobile
> and d/l-mobile (in IE words like "tear" or "tongue"). As
> explanations for s-mobile, and possibly also for k-mobile and
> d(l)-mobile does not need substrate influence, also a-mobile does
> not [need?] anything external.

We have accepted that we haven't found (yet?) an explanation for the
s-, k- and d- prefixes. That doesn't mean we should do the same for
the a- prefix.

> 3) One form of a given word may be inherited (or: borrowed already
> in the common IE stage) while the other form may be borrowed to a
> specific branch. The source may unnecessarily be Semitic, like in
> the instance of "ore" (where the original source was probable
> Sumerian, and attested Semitic forms were borrowed parallelly).

I note with satisfaction that you read my proposals in cybalist.
Personally I think the origin should be sought further east; there's
something fishy about the whole multiple loci of discovery' idea.

> 4) The word in question is a Wanderwort, and both its forms have
> been taken from different sources "on its way". An example of this
> model may be IE "apple" (with the a-less forms like in Latin
> "ma:lum" < *maHl- and with a-having forms like English "apple" <
> *h2a-bl-, to mention the most known variations, from earlier
> *mVl-/*-bVl-). Another example may be the Balto-Slavic-Albanian term
> for "nut". In Slavic and Albanian only o-having forms exist (Slavic
> ore^xU < *a-rais^a-, Albanian arrë) while in Baltic - only a-less
> forms exist (Lith. ríes^as), both supposed to be from North
> Caucasian source.

See my earlier post on 'apple'.

> 5) The observed similarity is only seeming, and actually both forms
> come from two different IE protoforms.

Then we'd have to accept that PIE speakers had two words for water:
*akW- and *ap- (and *am-?). I don't think so.

> Anyway, the only aim of my posts in this thread was not to negate
> existence of Semitoid influence in western branches of IE. I have
> only called the attention to the fact (yes, the fact) that the
> presence of a-/zero alternation in several IE words does not give a
> base for reconstruction a bird substrate or for any Semitoid
> influence (which, on the other hand, cannot be rejected, but on a
> different base).

That's is categorical mistake: a rule of proof is a rule, not a fact.

> Particularly:
> a) there is no ground for the statement that the a-mobile is a trace
> of the Semitic article,

Some have mentioned the possibility, but you're probably right.

> b) the observed anlaut hesitation may be explained in many different
> ways, not necessarily with the help of substrate.

Irrelevant. The fact that several competing hypothesis on a set of
facts exist, does not in itself disprove any of them. And if it were
so, which one would it be? The latest? The first?

> Naturally, the discussed possibility cannot be fully rejected - but
> it is less probable than other possible explanations. According to
> the most basic principles of the science (like Occam's razor),

Occam's Razor is secondary principle, it is used for culling among
theories having passed the first tests.

> such an explanation should only be mentioned.

Why? Occam's razor states that the theory with the least machinery,
set of concepts ('entities'), should be preferred. The theory you
propose, which incorporates unexplained exceptions ('irregularities')
would have been culled before it got to Occam's, on the primary
principle that the theory which explains the most (ie. with the fewest
irregularities) should be preferred.

> Giving it too much credence (or treating it as an argument for
> anything else) is a serious methodological mistake.

No. But proposing a theory which calls some phenomena 'irregular' is.
Imagine a researcher who proposes a theory that some cell changes are
'irregular'. He would get funding (hopefully). Everyone would see him
for what he was: arrogant and intellectually lazy.


Torsten