Pashta vs. Qadma

152 views
Skip to first unread message

Art Roth

unread,
Aug 24, 2017, 1:43:46 AM8/24/17
to leining
Common sense based on context says that the trope on b"shaleixakha (D"varim 15:18) is a qadma rather than a pashta, i.e., it makes no sense for there to be even a small pause between b"shaleixakha and oto.  But is there any way to tell this visually?  Since the word is milra` and the last syllable consists of just a single letter, both qadma and pashta would be on that letter.  And since the previous trope is a r"vi`i, both qadma and pashta make sense with respect to the trope sequence in the pasuq (ignoring context).  When I first thought of this question, my first impulse was that maybe on such words, care would be taken to place the trope near the beginning of the letter (for a qadma) or the end of the letter (for a pashta) --- and in my edition of Tiqun Simanim, the trope is indeed near the rightmost portion of the khaf (sofit).  But in all of my other xumashim and tiqunim, the trope is inconspicuously located near the middle of the khaf.  Can anyone think of any other such words that have a qadma or pashta, so that we can look to see whether Tiqun Simanim makes this distinction consistently or whether this particular occurrence was just coincidence?  If this is indeed the appropriate way to indicate the trope in such situations, are the other xumashim remiss in this respect?

Art    

Mark Symons

unread,
Aug 24, 2017, 2:30:06 AM8/24/17
to lei...@googlegroups.com
The convention has generally been that the pashta is on the very last letter and protrudes beyond the word - I have heard this as one explanation why it is called pashta, as it is "posheit" beyond the word - whereas a kadma is placed slightly back.

On 24 August 2017 at 15:43, Art Roth <artj...@gmail.com> wrote:
Common sense based on context says that the trope on b"shaleixakha (D"varim 15:18) is a qadma rather than a pashta, i.e., it makes no sense for there to be even a small pause between b"shaleixakha and oto.  But is there any way to tell this visually?  Since the word is milra` and the last syllable consists of just a single letter, both qadma and pashta would be on that letter.  And since the previous trope is a r"vi`i, both qadma and pashta make sense with respect to the trope sequence in the pasuq (ignoring context).  When I first thought of this question, my first impulse was that maybe on such words, care would be taken to place the trope near the beginning of the letter (for a qadma) or the end of the letter (for a pashta) --- and in my edition of Tiqun Simanim, the trope is indeed near the rightmost portion of the khaf (sofit).  But in all of my other xumashim and tiqunim, the trope is inconspicuously located near the middle of the khaf.  Can anyone think of any other such words that have a qadma or pashta, so that we can look to see whether Tiqun Simanim makes this distinction consistently or whether this particular occurrence was just coincidence?  If this is indeed the appropriate way to indicate the trope in such situations, are the other xumashim remiss in this respect?

Art    

--
You received this message because you are subscribed to the Google Groups "leining" group.
To unsubscribe from this group and stop receiving emails from it, send an email to leining+unsubscribe@googlegroups.com.
To post to this group, send email to lei...@googlegroups.com.
Visit this group at https://groups.google.com/group/leining.
For more options, visit https://groups.google.com/d/optout.

jecg...@aol.com

unread,
Aug 24, 2017, 5:24:59 AM8/24/17
to lei...@googlegroups.com
Most -- but by no means all-- chumashim do differentiate between kadma and pashta when it is on a final one-letter syllable.  Exactly where they are placed varies, generally the pashta sticks out, but sometimes the pashta is at the left end without sticking out and the kadma a bit further to the right. 
Ephraim
 


To unsubscribe from this group and stop receiving emails from it, send an email to leining+u...@googlegroups.com.

jaybr...@gmail.com

unread,
Aug 24, 2017, 8:27:34 AM8/24/17
to leining
In this week's parasha, v'yasafta (19:9) has a qadma on the last letter.

Koren has distinctive symbols for qadma and pashta, as well as mahpakh and y'tiv.

j

Avram Herzog

unread,
Aug 24, 2017, 8:41:54 AM8/24/17
to lei...@googlegroups.com
Hi,

Art, this doesn't help resolve consistency/inconsistency within Simanim on kadma/pashta, but Koren does distinguish slightly between the two symbols in general.  I didn't realize this until it was pointed out to me several years ago.  The Koren pashta is a bit wider on top and narrower at the bottom (it widens as it rises), and the kadma is the opposite.  (I realize you may be aware of this already.)

This is not to suggest that Koren and Simanim are in harmony re. b'shaleichacha, as on several occasions they are not in accordance with one another.

KT,
Avi H

On Thu, Aug 24, 2017 at 1:43 AM, Art Roth <artj...@gmail.com> wrote:
Common sense based on context says that the trope on b"shaleixakha (D"varim 15:18) is a qadma rather than a pashta, i.e., it makes no sense for there to be even a small pause between b"shaleixakha and oto.  But is there any way to tell this visually?  Since the word is milra` and the last syllable consists of just a single letter, both qadma and pashta would be on that letter.  And since the previous trope is a r"vi`i, both qadma and pashta make sense with respect to the trope sequence in the pasuq (ignoring context).  When I first thought of this question, my first impulse was that maybe on such words, care would be taken to place the trope near the beginning of the letter (for a qadma) or the end of the letter (for a pashta) --- and in my edition of Tiqun Simanim, the trope is indeed near the rightmost portion of the khaf (sofit).  But in all of my other xumashim and tiqunim, the trope is inconspicuously located near the middle of the khaf.  Can anyone think of any other such words that have a qadma or pashta, so that we can look to see whether Tiqun Simanim makes this distinction consistently or whether this particular occurrence was just coincidence?  If this is indeed the appropriate way to indicate the trope in such situations, are the other xumashim remiss in this respect?

Art    

--

MG

unread,
Aug 24, 2017, 11:54:29 AM8/24/17
to leining
Numbers 31:28 - VaHareimota Meches...  Kadma on last letter.  Although there is clearly no confusion about whether it is a pashta, you can at least see what Simanim does there.

It's my understanding that Kadma is placed on the middle portion of the letter containing the accent and the pashta on the very left of the last letter of the word. So even if the accent is on the last letter, you should be able to tell.

Art Roth

unread,
Aug 25, 2017, 1:07:53 AM8/25/17
to leining
Avi, thanks for pointing out that Tiqun Simanim uses a somewhat different symbol for pashta and qadma.  I'm amazed that I never noticed that before --- but maybe I shouldn't be so amazed considering that I've been conditioned for so many years to believe that these tropes are identical in appearance, so there was no reason for my brain to internalize the difference.  Each time I saw either one, I "recognized" the symbol as the one that (in my mind) represented both qadma and pashta, so there was no need to do any further processing.

We all know that when a pashta appears on a mil`eil word, virtually all xumshim have two pashtas --- the first one to indicate where the accent is and the second one (at the end of the word) to indicate that it's a pashta rather than a qadma.  Well, once there are two different symbols, there would be (at least in theory) no need for the second pashta.  Of course, even Tiqun Simanim doesn't go to this extreme --- it would be too confusing to those ignorant people (like me before today!) who are oblivious to the use of the two different symbols.  However, I noticed something weird --- in such words, the first "pashta" is displayed in Tiqun Simanim using the symbol for qadma!  I wonder why --- it seems to me that two pashtas would be more appropriate than a qadma and a pashta.  But I'll propose a possible answer in what follows.

Thanks to Jay for pointing out v"yasafta (D"varim 19:9) and to MG for pointing out vahareimota (B"midbar 31:28).  On top of that, I realized that there are probably many examples of both pashta and qadma on the word l"kha.  I certainly didn't look for lots of them --- just thought of one of each off the top of my head, namely asei l"kha teivat atzei gofer (pashta --- B"reishit 6:14) and v"haya l"kha l" 'ot al yad"kha (qadma --- Sh"mot 13:9).  It turns out that Tiqun Simanim places two of the qadmas (D"varim 15:18 and 19:9) on the extreme right hand side of the last letter of the word and the other two qadmas (B"midbar 31:28 and Sh"mot 13:9) near the middle of the last letter of the word.  At this point, I thought that Tiqun Simanim was inconsistent --- but I was wrong.  When I looked up the pashta in B"reishit 6:14, I discovered that this pashta and all others (i.e., even the ones where the last syllable consists of more than one letter) is placed on the far left side of the last letter of the word, so that a small part of it is "posheit" (as someone has already pointed out) beyond the entire word.

At this point, I realized two things.
     (a) Simanim is perfectly consistent in distinguishing between qadma and pashta when the last syllable contains just one letter --- pashta is on the extreme left side of the letter, while qadma is sometimes on the extreme right side of the letter and sometimes in the middle of the letter.  It's strange that there's not more consistency between different qadmas --- but there's a perfectly ironclad rule for distinguishing pashta from qadma based purely on location (even if you ignore the use of the two different symbols).
     (b) Perhaps the desire to restrict the pashta symbol only to the very end of a word (and beyond) is the reason why they chose to use the qadma symbol for the first "pashta" in words that require two of them.  Nevertheless, I would personally prefer to see a pashta symbol in the middle of a word that has two of them than to see a qadma symbol on a word that clearly has no qadma associated with it.

Art

Avram Herzog

unread,
Aug 25, 2017, 1:16:13 AM8/25/17
to lei...@googlegroups.com
Hi Art,

It's Koren, not Simanim, that uses a slightly different symbol for pashta and kadma.
Avi H

Avram Herzog

unread,
Aug 25, 2017, 1:21:20 AM8/25/17
to lei...@googlegroups.com
Let me rephrase what I just wrote.  I should have written that I was originally referring to Koren, not Simanim.  Indeed, Simanim does as well but in a different way--Simanim's pashta is longer and more curved than the kadma.

Avi H 

jaybr...@gmail.com

unread,
Aug 25, 2017, 7:45:24 AM8/25/17
to leining
> in such words, the first "pashta" is displayed in Tiqun Simanim using the symbol for qadma!  I wonder why --- it seems to me that two pashtas would be more appropriate than a qadma and a pashta.

In my Simanim, the names of the trope are presented for S'pharadim, Askenazim, and Teimanim. What you and I call a pashta on a mil'&el word is called t'rei-qadmin by S'pharadim (and fishtin by Teimanim).  It's still a mafsiq, and clearly Simanim wants to distinguish the mafsiq from the m'sharet.  But since Simanim reflects S'pharadi usage, the use of qadma for the first mark of t'rei-qadmin makes sense.

j

jaybr...@gmail.com

unread,
Sep 3, 2017, 12:55:54 AM9/3/17
to leining
In yesterday's reading, 23:15 contains the word לְהַצִּֽילְךָ֙ which, in context, could be a qadma or pashta, as it appears before mahpakh, and there is a preceding r'via& (without which it could not be pashta).  Koren marks it as pashta, which, as pointed out above, uses a slightly different symbol than qadma in Koren publications. 

j

Art Roth

unread,
Sep 3, 2017, 3:29:15 PM9/3/17
to leining
This is another good example of the phenomenon under discussion.  Simanim also indicates a pashta here.  I agree with Jay that without a special symbol, the context of the trope would be consistent with either a qadma or pashta --- so the special symbol is indeed useful.  However, I personally think that it would be a bit of a stretch (although not impossible) to justify a qadma here based on the context of the words, i.e., a mafsiq on this particular word seems to make the pasuq flow much more naturally to me than a m"shareit.  That's not a criticism of Jay's point --- just an additional comment from the peanut gallery.

Art 

jaybr...@gmail.com

unread,
Sep 3, 2017, 5:41:22 PM9/3/17
to leining
I agree, Art.  In fact, this was almost a counter-example to your original example, which was far more clear-cut.

Like you, I could go either way on this; the one indication that mafsiq is more fitting (and, admittedly, a retrospective indication) is that the word l'fanekha only goes with v'latet oy'vekha.  It could not go the l'hatzil'kha, as there is (to my knowledge) no expression l'hatzil'kha l'fanekha.

j
Reply all
Reply to author
Forward
0 new messages