Decalogue pronunciation is influenced by cantillation

[bdenckla]

It is somewhat well known that the Masoretes chose to preserve two cantillations for each of the Decalogues.

It is not as well known that there are many reader-friendly editions that present these cantillations in conveniently separated form. (BHS, like the great Masoretic manuscripts, presents both cantillations superimposed on a single set of letters. Because BHS and editions close to it dominate the market, alternatives are not well known.)

One might think that the superimposed presentation poses only a mild inconvenience if all you want to do is pronounce rather than chant a Decalogue. In other words, one might think that there are just more accents to ignore than usual.

While this is largely true, it is not 100% true. The few words for which it is not true remind us of the ways in which cantillation interacts with pronunciation in all of the Hebrew Bible, not just the Decalogues.

Perhaps the most obvious words varying in more than just accents are the two words varying in vowel marks. One is על־פני, which, in superimposed form, looks like this:



What the heck? That poor nun is burdened with both a qamats and a pataḥ?! The other such word is מתחת, whose tav is similarly burdened:



We could dismiss these words as, while differing in more than just accents, still not differing in pronunciation. That is another statement that is largely true but not 100% true. There are some remaining communities who pronounce qamats and a pataḥ distinctly.

A word whose pronunciation varies between cantillations in all communities I'm aware of is כל־מלאכתך, which, in superimposed form, looks like this:



Again (but for a different reason) I feign surprise by saying: what the heck? Assuming we even remember what rafeh normally means, what the heck does it mean when combined with dagesh?! Here it may help to see the cantillations separated rather than superimposed. I provide those below, along with some phonetic transcriptions using Jacobson's system:



(Those transcriptions are part of a wider project I am working on to provide transcriptions of the entire edition of MAM used in the Al-Hatorah Mikraot Gedolot. At the moment all that's available is the rest of Exodus 20.)

That's all I'll write for now on this topic; if it stirs up some interest I may present and discuss a few more Decalogue words whose pointing differs by more than just accents.

[ducky]

Hi Ben,

I'm looking at your transcriptions that you've made, and it is very impressive. Not just the hard work, but the punctiliousness as well.
I have on my computer something similar that I downloaded a long time ago - a Hebrew Transliteration of the Bible, but it is only the letters and without the vowels. And what you do is really about reading it and understanding.
(If it interests you, then here is a link for the PDF: http://rcbi-resources.yolasite.com/reso ... ration.pdf

***************

While reading it, I saw some points think about them, and some things that cause me to think that I didn't quite understand your method fully.
This is absolutely no criticism, because it is indeed great. And it is also not suggestions to change anything in your method.
Only some points to not ignore and to ask about them.

1. About the Dagesh.
You cut the Dagesh Forte to two and give each one to another syllable.
It is what expected to do.
But couldn't it be confusing with the reading?
Because word like וידבר (the first word in your file is transcript as: vay-dab-BÉR.
It is double B that is read as one B.
But a reader may be confused and read the letter B twice.
For example, the פן-נמות in Ex. 20:16 (in your file as well) is transcript as expected: pen-na-MUT.
The combination of the double N (pen-na) is the same as the combination of the Dagesh Forte.
But in this case, the N is really pronounced twice.

2. About the Sheva and the Hataph (the reduced vowels).
You gave a special mark for the Mobile Sheva, but you didn't do the same with the Hatap's.
The Hataph-Patah has the Mark of a regular Patah, and a Patah-Segol has the mark of a regular Segol.
A reduced vowel is a reduced vowel. and if you (rightfully) gave the Mobile Sheva a special mark, then the rest should get a special mark as well.
And by the way, The Hataph-Segol should get the same mark of the Mobile Sheva. And so, the first letter Aleph in the word אלהים should start with the same vowel of the Mobile Sheva (unless you want to be specific with every vowel. But I didn't see other vowels marked in a specific ways).

3. About Qamats and Patah.
I see that Qamats and Patah has the same mark.
Why?
If it is because they have the same sound, then why Tsere and Segol, who have the same sound, are marked differently.

4. Segol+silent-Y (is it called "plene Segol"?).
I saw that you differ the marking of Segol and Tsere for Tsere is a big vowel and Segol is a small vowel.
But then I saw Segol+silent-Y (like: אלהיך) and the mark of the letter H=ה is "small vowel".
But the Segol+silent-Y would be a big vowel.

5. Hiriq without Y (is it called "Deficient Hirik"?).
A word like יארכון (in your file) is written without the expected letter Y=י after the letter R=ר.
You marked the vowel as if it was a small vowel (because it comes without the Y), But even though it has no Y, it is still a big vowel.
(I don't know if I understood your method of how you choose to mark the vowel. It is about the way it is written or about the essence, or is it about how it sounds...)

6. Shuruq.
I see that you mark the Shuruq as a small vowel?
(There is no real Qibbuts in your file).

7. Dehiq.
I saw that you double the Dehiq with Maqaph in the second word (like: תַּֽעֲשֶׂה־לִּ֔י = ta-‘a-se-LLÍ).
What is your way when there is no Maqaph? (like לאמר in Ex. 6:10). Putting a double letter in the beginning of a word?

8. Consonants.
Last thing (hopefully) is about your decision to give two letters the same English letter.
For example:
כ-ק are both transcript as K.
ו-ב are both transcript as V.
ט-ת are both transcript as T.
ס-ש are both transcript as S.
Maybe, just like you are going for punctilious in the vowels and in the organization in the word's-letter, it would be nice to give another sign for each letter. Just maybe.

************************
I feel the need to say it again: Don't think of these words that I wrote as criticism. Because what you do is really great. It is really just points that i see confusing (at least for me).

[bdenckla]

1. About the Dagesh.
You cut the Dagesh Forte to two and give each one to another syllable.
It is what expected to do.
But couldn't it be confusing with the reading?
Because word like וידבר (the first word in your file is transcript as: vay-dab-BÉR.
It is double B that is read as one B.
But a reader may be confused and read the letter B twice.
For example, the פן-נמות in Ex. 20:16 (in your file as well) is transcript as expected: pen-na-MUT.
The combination of the double N (pen-na) is the same as the combination of the Dagesh Forte.
But in this case, the N is really pronounced twice.

Thanks for your thoughtful feedback. My short answer to all or most of the issues you raise is sort of a cop-out: I'm just using Jacobson's system, so these are not really my choices to explain the rationale behind. But let me dive down a little deeper than that cop-out.

I agree that splitting up doubled letters across the syllable boundary, although conventional, has its problems.

• As you point out, it introduces an ambiguity, as you point out with the example of פן-נמות becoming pen-na-MUT.
• Even setting aside this ambiguity, I do find it a little confusing, or at least not to my liking.

Perhaps one way to mitigate the ambiguity (without bucking convention) would be to use something other than plain old dash (hypen) for both maqaf and syllable boundaries. So, for example, I might use a different type of dash like "en dash" or "em dash" for maqaf, e.g.:

• pen–na-MUT (en dash) (may not be very different from hyphen)
• pen—na-MUT (em dash)

How different "en dash" is from hyphen is font-dependent; it seems that they are not very different in the default font used on this site. Yet, the "em dash" is perhaps so long as to be overkill. In Times New Roman, (at least the one I have), here's what the "en dash" option looks like:



Alternately, one might use mid-dot for syllable boundaries and continue using plain hyphen for maqaf, e.g.:

• pen-na·MUT

Or in (my) Times New Roman:



This use of hyphen and mid-dot is what appears in Hebrew World's Phonetic Bible, e.g.



(They also seem to have generous side-bearings around the hyphen.) (Note that unlike me, they transcribe sof pasuq (as a period).)

I'll respond to some of the other issues you raise in separate replies.

[bdenckla]

2. About the Sheva and the Hataph (the reduced vowels).
You gave a special mark for the Mobile Sheva, but you didn't do the same with the Hatap's.
The Hataph-Patah has the Mark of a regular Patah, and a Patah-Segol has the mark of a regular Segol.
A reduced vowel is a reduced vowel. and if you (rightfully) gave the Mobile Sheva a special mark, then the rest should get a special mark as well.
And by the way, The Hataph-Segol should get the same mark of the Mobile Sheva. And so, the first letter Aleph in the word אלהים should start with the same vowel of the Mobile Sheva (unless you want to be specific with every vowel. But I didn't see other vowels marked in a specific ways).

Here's what Jacobson says (page 243):

Syllables containing ultrashort [aka ḥataf] vowels are chanted more quickly than those containing normal [short] vowels. In our transliteration, we do not use a separate symbol to distinguish the ultrashort from the normal short vowels; the reader must consult the Hebrew spelling.

This doesn't really justify his choice to blur these distinctions, but at least it acknowledges said blurring.

[bdenckla]

3. About Qamats and Patah.
I see that Qamats and Patah has the same mark.
Why?
If it is because they have the same sound, then why Tsere and Segol, who have the same sound, are marked differently.

I guess Jacobson is targeting a different pronunciation (or at least a different ideal of pronunciation) that what you have in mind. From Page 242:

[bdenckla]

4. Segol+silent-Y (is it called "plene Segol"?).
I saw that you differ the marking of Segol and Tsere for Tsere is a big vowel and Segol is a small vowel.
But then I saw Segol+silent-Y (like: אלהיך) and the mark of the letter H=ה is "small vowel".
But the Segol+silent-Y would be a big vowel.

5. Hiriq without Y (is it called "Deficient Hirik"?).
A word like יארכון (in your file) is written without the expected letter Y=י after the letter R=ר.
You marked the vowel as if it was a small vowel (because it comes without the Y), But even though it has no Y, it is still a big vowel.
(I don't know if I understood your method of how you choose to mark the vowel. It is about the way it is written or about the essence, or is it about how it sounds...)

6. Shuruq.
I see that you mark the Shuruq as a small vowel?
(There is no real Qibbuts in your file).

See my response to your #3 above, i.e. see Jacobson's table from his page 242. If I understand your #4 and #5, they could be summarized like this:

• Should segol-yod be transcribed any different from segol without yod?
• Should ḥiriq without yod be transcribed the same as ḥiriq-yod, for a word that would normally be spelled with yod there?

It would be quite a challenge to transcribe ḥiriq without yod depending on whether there would normally be a yod there. But just because it would be hard does not mean it should not be done!

By the way, I don't think I made it clear that my transcription of Exodus 20 (and indeed of the whole Tanakh, though I have not posted it) is automated.

Being automated has the great advantage that I can easily produce transcriptions following different rules, e.g. giving segol-yod a different transcription from segol without yod. Or, for example, I have transcriptions (not yet posted) using the IPA (International Phonetic Alphabet) rather than Jacobson's choices.

Being automated has the disadvantage that it is difficult to make judgments like "even though this ḥiriq is without a yod, I'll transcribe it like ḥiriq-yod, because there would normally be a yod there." It is difficult, of course, because it would require annotating the whole Tanakh with judgments like that, e.g. "deficient spelling used here where full spelling is more common."

[bdenckla]

7. Dehiq.
I saw that you double the Dehiq with Maqaph in the second word (like: תַּֽעֲשֶׂה־לִּ֔י = ta-‘a-se-LLÍ).
What is your way when there is no Maqaph? (like לאמר in Ex. 6:10). Putting a double letter in the beginning of a word?

Indeed, though it is a little awkward-looking, I double such letters, e.g.

[ducky]

Hi Ben,

About the Dagesh.
I think this would be a good idea to represent the syllables cut by a mid-point.
It looks good and also not taking space causing the word seems more connected (as a word should).
And still, when the syllable is cut by a Dagesh, it would be confusing to the reader reading a word like "kidding" as "kid ding".
Maybe you should have a specific cut-sign when it comes to a case of a Dagesh, to specify that the letter is read only once.
Something like this sign: ~
So "kidding" would be "kid~ding"
(while all of the other syllables will be cut by a midpoint).

And while we're speaking about it, maybe it would be a good idea to give the Maqaph a "cut-sign" of its own: a true Maqaph (Like: ־).
Because if it would be just like the other regular sign (the mid-point), it will look as if it is a part of the word (and not been realized that it is actually another word that is connected).

(It depends on what is the purpose of your project. I assume it is for helping the reader who study Hebrew. And if so, then this would be the helpful way).

So, I think that if there would be three the cut-signs that would be nicer.
The regular sign - the mid-point.
The sign that represents a Dagesh forte cut - as ~ (that tells the reader to pronounce the letter only once).
The sign that represents a connection of two word by Maqaph (can be signed by Maqaph).

****
I hope I am not putting my nose to deep into your project.
I was very impressed by it, and I just want to help by suggestions to try improving.
If I am bugging, feel free to shut me down.

************************

About the reduced vowel.
His comment seems very ridiculous.
The Mobile Sheva is marked as a half-vowel, but the Hataphs as full vowels?
I think not marking reduced vowels differently from regular vowels is just amateurish.

***************************
About the Qamats-Patah

He said that the reason of similar mark for Shuruq-Qibbuts and Qamats-Patah is because they are pronouced the same.
He's right.
But what about Hiriq with Y and without Y? - aren't they pronoucne the same?
What about Tsere and Segol? - Aren't they pronounced the same?

In his examples he brings for Shuruq-Qibbuts he wrote שובו-שבו.
The two cases demand the letter V/W=ו after the SH=ש.
And so, the example of שבו is not a real example of Qibbuts (curious if he wrote another set of examples)

*******************

I understand that if it is automated than it can't be perfect because the "unregular" cases would be missed.
Or maybe that the different sign of the (I) is just comes to say what it is written (and not the essence of the vowel).

It is all depends on what is the purpose of your project.
If it is a matter of just let the reader reads.
Or get him a little bit inside the grammar.

************************
Anyway, no matter where you decide taking your project to, it is very good.

[bdenckla]

8. Consonants.
Last thing (hopefully) is about your decision to give two letters the same English letter.
For example:
כ-ק are both transcript as K.
ו-ב are both transcript as V.
ט-ת are both transcript as T.
ס-ש are both transcript as S.
Maybe, just like you are going for punctilious in the vowels and in the organization in the word's-letter, it would be nice to give another sign for each letter. Just maybe.

These are phonetic transcriptions and thus they need to use the same Latin letter to represent the same sound. Since I am targeting a pronunciation of Hebrew that gives the same sound to all of the Hebrew letter-pairs you mention, I have to use the same Latin letter for both Hebrew letters of each pair you mention.

The great thing about being automated is I could easily change things to target a different pronunciation, where, for example, ט and ת have different sounds. For example give Charles Loder's hebrewtransliteration.app a try with the "Tiberian" setting (based on the work of G. Khan). I don't know the details, but I think almost every letter and vowel-mark has a distinct sound (and thus a distinct transcription) in that system!

When I am "romanizing" (transcribing Hebrew to the Latin alphabet) by hand, for more than just phonetic purposes, I make some of these distinctions, particularly I like to make the כ-ק distinction by using q for ק and k for כּ (kh for כֿ).

But here the purpose is strictly phonetic. I imagine these transcriptions usually appearing alongside the Hebrew, which would allow homophones to be distinguished, if meaning is a concern.

[bdenckla]

I started a new topic over in a different (sub-) forum since nominally, the original topic here did have to do with pronunciation, but was quite narrowly concerned with some particular words whose pronunciation varies in the two cantillations of the Decalogues.