Monday, October 20, 2003

The pseudo-pausal

We just switched over to saying mashiv haruach umorid hagashem. Or is that morid hageshem? Yup, its more pausal stuff, a followup to Succat David Hanofelet!

Some folks say hageshem, some say hagashem. The difference between the two is the hagashem assumes that the phrase mashiv haruach umorid hageshem is the termination of the previous paragraph in shmoeh eserei. Hageshem, which is not pausal, assumes it is the beginning of the next paragraph (for the phrase stands between two paragraphs). Only at the end of a paragraph would it be at the end of a sentence and thus warrant pausal form.

While this is so, the matching phrase, morid hatal, or as some say it, mashiv haruach umorid hatal, is by all accounts in pausal form. The pausal form is heh patach tet kametz lamed, that is, with a kametz as opposed to a patach under the tet. This is so even according to those who say hageshem.

Why? This is something in Hebrew known as the pseudo-pausal. It has pausal form but does not come as a result ot being at the end of a sentence or midsentence.

Rather, when you have roots of the form XYY, often the second Y drops off. Thus, RBB as myriad, but we say rav. HRR as mountain, but we say har. GNN as garden, but we say gan. TLL as dew, but we say tal.

However, is certain forms, the gemination of the second root letter manifests. For example, Harerei kedem, telalim, revevot.

When the definite article "the" = ha is added to such a word, rather than manifesting the second geminated root letter, the patach becomes a kametz, in a pausal manner(perhaps as compensatory lengthening. Often, when we would ordinarily double a letter with a dagesh, but we cannot because it is a gutteral, we lengthen the vowel under the preceding letter in compensation). So, if you look in the beginning of Bereishit, you will encounter over and over hagan, with a kametz under the gimel, even where there is no trup of etnachta or silluq. You will not encounter hagan with a patach under the gimel.

So too hatal.


Tally said...

This is incorrect. There is no "pseudo-pausal" rule in Hebrew.
Compensatory lengethening only occurs with a gutteral letter and only on the letter prior to the gutteral, as a replacment for the dagesh. Never on the letter itself. Besides, these aren't even gutterals.

Please provide a source for your "rule".

See the Radak in Michlol who deals with "Hagan" vs. "Gan".
"Hatal" is unrelated and would only receive a kamatz if there were a true pause (although we do find "Tal" with a kamatz with a zakef-katon in Tanach.

See Sefer Emes LYaakov, Bereishis 3:19, who deals with your question on "HaTal" with a kamatz vs. "HaGeshem" with a segol and poses an answer. If there were a "pseudo-pausal" rule, R' Yaakov would have used that obvious answer.

joshwaxman said...

bli neder, i'll try to reply in a few hours.

kol tuv,

joshwaxman said...

There are two levels upon which this disagreement could take place -- whether it is a contest of authorities, or a contest of the facts. I would prefer the latter, because even the authorities are not operating with a masorah. There is indeed a masorah as to what nikkud and dagesh goes where, but the authorities then examined all of this data, this evidence, and in their brilliance formulated rules and explanations to account for all of the data.

So, while I can, to an extent, play in this contest of appealing to authorities, I would rather not.

Now, to address your points, as best I can...

This is incorrect
You believe it to be incorrect.

Compensatory lengethening only occurs with a gutteral letter
Yes, that is the regular case in which we encounter it. This is because the guttural cannot take the dagesh. But that does not mean that the phenomenon would not manifest itself in other instances, such as the one under discussion, where there cannot be a dagesh.

as a replacment for the dagesh
yes, that is why it is called "tashlum dagesh". i am with you so far.

Never on the letter itself
from your comments here, i believe it possible that you are misunderstanding me. in the word hattal, the compensatory lengthening is not because of anything to do with the letter ט. It has to do with the letter ל. The triliteral root would be tll, but there is only one lamed. So we can place a dagesh chazak in the lamed. Only we cannot because it is in the end of the word. This is thus "as a replacment for the dagesh", as you write, and it is "on the letter prior to the" letter, on the tes prior to the lamed.

I think you may have misunderstood my intent as talking about the dagesh on the tes. I was not.

Please provide a source for your "rule".
No need for snide quotation marks. An examination of the data reveals such a rule. But what do you consider the source for a rule? Is this an appeal to authority?

If you want to know an authority figure, if I recall correctly, I heard it from Dr. Richard Steiner (a rather famous scholar in Semitic languages) in one of the several courses I took with him in Semitic languages. I don't recall if he provided the explanation about tashlum dagesh, or if it was just the "rule" of the pseudo-pausal.

I would note that modern Semitic language scholars differ with medieval Semitic language scholars as to whether bi-literal roots exist, which could account for a disparity between what you have read and what Dr. Steiner said (to my recollection).

See the Radak in Michlol who deals with "Hagan" vs. "Gan".
It would make sense that Radak would have an alternate explanation of the phenomenon, since he would be working with a different theory of the language. This is alright.

See Sefer Emes LYaakov
I am certain that Rav Yaakov Kaminetsky, zatz"al, was a Gadol, as well as quite knowledgeable in dikduk. Even so, my impression is that Dr. Steiner is a greater expert in this field. Also, if Rav Kaminetzky was basing himself on classic sources for his dikduk, then it would make sense that he would not have spotted this "obvious answer".

Now that I have clarified in what regard I intended "compensatory lengthening", and have clarified my "source", what say you?

All the best,

Tally said...

I'm happy to debate over facts. I'm still waiting for some.

It's quite funny how you claim to be above playing the "My Rabbi is smarter than your Rabbi" game, but you do it yourself in the last paragraph. At least I know that I am not able to make such a judgement call over who was "greater" in dikduk. How childish.

Yes, I am are aware that the rules were derived based on the data; but there have been a number of books written on the subject that contain said rules. I was looking for a source, which you don't have. I'm all for remembering what you learned in Semitic Languages 101, but the rule has to work. The burden is on you to prove that it does.

How about the word "Gal"? It follows the XYY pattern and we find HaGal with a patach in the Torah under non-pausal trup.

At the end of the day, there is the Radak who says clearly that the word is HaTal (patach) unless it takes a true pausal-form (zakef koton included). If you want to ignore that under the psuedo-intellectual guise that he was working under a different theory of the language, that's your prerogative.

joshwaxman said...

first off, i want to apologize if i came off too strongly. i didn't mean to insult, but you seemed to write quite harshly, and with hostility. and sometimes I adopt a no-nonsense approach.

"It's quite funny how you claim to be above..."
To clarify, my point was that i could contest you on either level, if you wished. To demonstrate that I could, I put in that last paragraph. But I do think it silly.

Are you willing to admit that you entirely misunderstood what I wrote the first time, and that all of your initial objections were based on this misunderstanding? (It could well be my fault. Perhaps my initial words were not clear enough.) If not, please explain how your words make sense, about how the present example is not an instance of compensation for dagesh and is on the letter, rather than preceding the letter. An apology or clarification would be a menshlich thing to do, before proceeding with another line of attack. And you may indeed be a mentch.

"I was looking for a source, which you don't have."
Which I cannot produce on the spot, in the form of a written book. So? Your question was implying that I made this up out of whole cloth. I most certainly did not.

"I'm all for remembering what you learned in Semitic Languages 101"
Who is acting childish now? Why the need to insult? I took more than 'Semitic languages 101', whether or
not I am right about this. Tell me, am I as much an idiot as you initially assumed, in your initial misreading of the theory?

"How about the word "Gal"?"
Great question, and straight to the point. And the answer is that I don't know. *Perhaps* you shlugged me up. Or *perhaps* there is an answer. Off the cuff, how about the following? The plural of טל is טללים. The plural of גל is גלים. In contrast, the word גללים refers to animal droppings, rather than many waves or many piles of stones. So, are you so positive that this is a true XYY pattern?

This theory is certainly falsifiable by counter-evidence. (On the other hand, perhaps I am just not of sufficient caliber to successfully defend the approach.)

"psuedo-intellectual guise"
why so combative? i do seem to recall that there is such a difference. are you saying that I am wrong in this? (it certainly is possible.)

"At the end of the day, there is the Radak who says clearly"
yes, there is a Radak. but Radak is not halacha leMoshe miSinai.

Please, before you respond to my other points, preface it with a statement on whether or not you entirely misunderstood my post initially.

Again, I apologize. I am much nicer in person, and when unprovoked.

All the best,

joshwaxman said...

"At least I know that I am not able to make such a judgement call over who was "greater" in dikduk. How childish."

Yes, I agree with you that you are not able to make such a judgement call. But I don't think it childish to make such a judgement call, and believe I am correct in this. Regardless, an appeal to authority is silly.

kol tuv,

Tally said...

Thank you for the apology, and I offer mine as well. I think this medium makes certain comments come across as rude when they aren't intended that way, and I appreciate that as a blogger you do get attacked anonymously and therefore may tend to open up in defensive mode when no offense was intended.
That said, I adopt the same no nonsense approach as you do. And I'm sure you are a mentch in real-life, as I can assure you I am as well.

I admit, of course, that I misunderstood what you initially wrote vis-a-vis the dagesh. However, as far as I know, compensatory lengethening only occurs over a gutteral letter (including Resh). So you are stretching (lengthening??) that concept to a place where I haven't yet seen it applied, and was asking for a source. If you can find compensatory lengethening in another context with non-gutterals, that would help make your case.

Forget "Gal" (although I still believe that question to be valid); the word "Tal" itself is found in the Torah as "HaTal" - with a patach, in Bamidbar 11,9. Shouldn't that alone disprove your rule?

Of course the Radak isn't Halacha LeMoshe MiSinai; but then again, nothing is. If we can't offer up Rishonim as evidence, then argument is pointless. This doesn't detract from Dr. Steiner; perhaps you can ask him for clarity and we can all benefit from his answer.

For the record, I think you are greatly misinformed about Reb Yaakov's knowledge with resepct to dikduk. Even a cursory reading of his writings reveals a very deep understanding of the language. I'm not bringing Reb Yaakov as a "proof" simply because he was a Gadol, in some kind of "all-Gedolim-know-everything" brainwashed way. Other Gedolim offer answers to this question (I believe there is a Chabad siddur that gives some suggestions as to why there is a Kamtatz), but I wouldn't infer from their lack of using this rule that there was no such rule. When you speak to Dr. Steiner please show him the piece.

joshwaxman said...


As noted at the google leining group, there are other examples. for example, regarding הרב, there is Bemidbar 35:8, with a patach.

I'll try to get confirmation / clarification next time I speak to Dr. Steiner. (I haven't spoken to him in a while.) Assuming that I remembered what I remembered correctly, and he actually does say this, it would not be because he didn't know of other approaches to deal with or solve this problem (such as that of Rav Yaakov or the Radak).

I could readily relax the rule to account for the data -- that we will *sometimes* see this seeming pausal form in non-pausal contexts, specifically in "doubly-closed", XYY forms, where the plural indicates that it is such a form. This would lead to lack of falsifiability. But it might well account for all the data, and might be what Dr. Steiner meant. We might have the ability to partly "confirm" by seeing if this pattern in non-pausal contexts is only true for such examples, if e.g. XY patterns as opposed to XYY patterns exist. (but maybe i am wrong about making such a distinction.)

I didn't mean to knock Rav Yaakov Kaminetsky, BTW. But there is a difference between someone who knows Hebrew dikduk very well, based on classical sources in Rishonim, on the one hand, and a Semitic philologist, on the other. I can't begin to explain the difference. But besides Hebrew linguistics of the middle ages, Dr. Steiner also knows the current research, and competing theories, on Biblical Aramaic, Galilean Aramaic, Biblical Hebrew, etc. He knows phonology in general, and has novel research on e.g. dating sound changes in Hebrew. (see e.g.
) As another example, he was able to argue, e.g., that "the letter known as Hebrew sin was pronounced as a fricative-lateral[2] and in another he argued that the pronunciation of the letter tsade as an affricate, /ts/, is very old and widespread, against others who had doubted this.[3] These books have convinced most specialists." There is more, but I'll leave it at that.

here is an analysis of 'compensatory lengthening' (by a different author) on the basis of the loss of case endings in Proto-Hebrew.
does Rav Yaakov ever deal with this? this is a different *type* of analysis.

This is a difference in breadth, and type, of knowledge. That was not intended to 'knock' the knowledge of Rav Yaakov Kaminetzky, chas veshalom. But at the same time, that Rav Kaminetzky does not offer an otherwise 'obvious' solution, which might make sense to a modern Semitic linguist, is not so persuasive to me.

Can you tell me offhand whether they treat טל and הר as XY with duplication in certain instances or as XYY, primarily?

kol tuv,

Tally said...

Of course, I understand the difference between Reb Yaakov and Dr. Steiner. But we are not talking about a complex rule here; this isn't phonology or Galilean Aramaic, where one would have an "expertise" over another. This feels like pretty basic dikduk, not PhD level stuff. But we agreed to not appeal to authority, so let's again look at the facts.

You are now relaxing this rule so much that it is hardly a rule, if it ever was one. We also find Saf-HaSaf/BaSaf (patach), Sal-HaSal (patach), Taf-HaTaf (patach), Dak-KaDak (patach), and many more, where in each of those cases it ONLY takes a pausal form under a true pause. Happy to give examples of each from Tanach.

Therefore, saying that an XYY-word "can" take this form (which of course was never under dispute) doesn't answer your original question. In cases like this where we mostly find the root XYY with a patach and sometimes with a kamatz, we need to look at the individual word in question, as it is found in Tanach, and use that. (This is standard practice in the liturgy, and there are many examples where we use a word exactly as it is found in Tanach, even though the "rules" would dicate otherwise.) Since we ONLY find "Tal" with a kamatz under a true pausal form, and in fact DO find HaTal with a patach, that is obviously how it needs to be in the liturgy, unless it is a pausal form.

I eagerly await clarification from Dr. Steiner.

joshwaxman said...

first off, i am not promising *any* clarification from Dr. Steiner.

"But we are not talking about a complex rule here"
if it is in which cases it makes sense to talk about compensatory lengthening, then it is a rule about the behavior of phonology and the way it behaves in a language. despite the fact that it feels to you like "pretty basic dikduk".

"this isn't phonology"
this is precisely phonology!!!! it is sound changes in different phonological contexts.

"We also find Saf-HaSaf/BaSaf (patach), Sal-HaSal (patach), Taf-HaTaf (patach), Dak-KaDak (patach), and many more, where in each of those cases it ONLY takes a pausal form under a true pause."
Please ALSO provide examples demonstrating that the plural manifests this extra letter. The point, even in the original post above, is that we see rbb and hrr and tll in the plural. do we see sll in the plural?

there is also the difference between the general trend and occasional exceptions to the rule. you would not say that the existence of Hanofeles as opposed to Hanofales proves that there is no such thing as a pausal form.

your "proof" regarding hatal relies on the pausal form including zakef katon, which generally is NOT where we find pausal forms, and one counter-example.

kol tuv,

Tally said...

The plural is irrelevent; it's something you made up to justify this rule. Regardless, we don't find GNN in the plural either, which is the "proof" you brought to your own rule.

Again, Hanofeles vs. Hanofales is not relevant. You are grasping here. It is well-established that one cannot bring proof from exceptions. We know a pausal form concept exists, clearly. One exception in Tanach does not disprove a rule. But in your case, we are talking about so many exceptions that the rule is no longer a rule. The very word we are dealing with is found in The Torah exactly as you say it shouldn't!

Zakef Katon can certainly create a pausal form; this is also well-established. In fact this is what some give as the reason why we say HaTal - kamatz - in davening, since there might be a slight, zakef-level pause there.

However, your original post does not say that. You state that HaTal is always with a kamatz, because of a rule. This is patently false.

I'm still not clear how you have answered the fact that we find HaTal-Patach on a non-pausal trup in the Torah.

Perhaps you should restate your piece to say that HaTal-Kamatz is only found with a zakef or greater level of pause, and Goshem is only found with an Etnachta or greater level of pause. Thus one can say HaGeshem and HaTal-Kamatz. This would be true, and consistent with what others have said.

Lastly, I'm not the one who needs clarification from Dr. Steiner.

joshwaxman said...

"The plural is irrelevent; it's something you made up to justify this rule."

it was stated in the very beginning, in terms of the rule. i didn't make it up ex post facto!

"It is well-established that one cannot bring proof from exceptions."
my point exactly!

"we are talking about so many exceptions that the rule is no longer a rule"
based on your exclusion of something that I stated initially, but that you saw fit to disregard.

"I'm not the one who needs clarification from Dr. Steiner."
I don't feel like I *need* clarification either, because you are unconvinced. This is probably a character flaw of shyness, but I seldom bug people in real life. (Besides, I already devote too much time to blog matters.) So maybe I will get around to it. Check back in a few months, or weeks. Or, bli neder, if i do clarify, i'll post on the leining group.

In the meantime, if you are unconvinced, please attribute errors to me, and me alone.

kol tuv,

Tally said...

To summarize your faulty logic:

Fact: We find HaTal with a patach under the Tes, never with a kamatz, in Tanach, under non-pausal trup, and we never find the word Tal with a kamatz except under a zakef-level pause or greater. This is indisputable.

You are saying that since we do find SOMETIMES, a few OTHER XYY forms with a kamatz, even though the word itself under question is NEVER found this way, we can say that where we do find this very word under question with a kamatz it can still be non-pausal. This is very thin ice, at best. Certainly the form in which we find this word *itself* in Tanach should shed light on how it behaves! Ignoring that fact is ridiculous.

We should certainly be able to agree there there is no such "rule".
We are not talking about a few exceptions. And as I have already said, the root GNN itself, from which you bring proof, does not fit into your plural scheme. You have not answered that either.

TLL contains no gutterals (unlike RVV and HRR), and is never found the way you claim it should be found; how can you possibly claim to be using the data to generate rules?

This will be my last post. Either you are after the truth, or you aren't.

joshwaxman said...

good. perhaps i will post a follow-up later, with a different summary.

it is not that i am after the truth. i don't agree with you, but that is not the same thing.

in the meantime:
hahar (harerei)
ha'am (?)
hagan (ginnah ?)
harav (rabbah)

a closed set of words, which we see instances of not usual pausal instances yet them assuming a seeming pausal form. for many, they manifest this second letter, or are geminate with the final heh.

a nice single theory would account for all this. and we could say that this is "tashlum dagesh" for the missing dagesh at the end of word. the same sort of compensatory lengthening we find by gutturals, where we ALSO cannot have the dagesh.

am i trying to "prove" everything from the word "tal" in particular, rather than a surprising pattern across a few words which resemble one another? no. will i persuade you? probably not. is this a simplistic "rule"? more like a consistent way to understand deeply the manifesting phenomena.

i understand your frustration. but that i am not giving in doesn't mean that i am not a truth seeker.

kol tuv,


Blog Widget by LinkWithin