When I first started this blog, one of the first articles I wrote was an explanation of the rule on pronouncing the Hebrew vowel qametz as o as in hole as opposed to its more usual pronunciation of a as in father. I chose this topic then because as a Hebrew teacher I have noticed that students find it perplexing. This week that discussion returned as a colleague asked me to comment on the word qodashim found in Leviticus 6:18 and repeated at 6:22. There we find it as part of the term qodesh qodashim.
This happens to be the term associated with the inner sanctum of the Temple and in such context it is often rendered “The Holy of Holies.” In this case, however, the meaning is more prosaic–the author is indicating that a particular foodstuff of offering has the highest degree of sacred status and in that context the translation is better rendered most holy or most sacred.
The meaning of the text is reasonably clear, but the pronunciation of the first vowel in qodashim is anything but.
When we first looked at the text, both the version of the Hebrew Bible we used and my electronic version contained the same spelling of the text. The vowel was a qametz followed by a diacritical mark called a metheg. Under normal circumstances, this indicates an a pronunciation, so the word would be vocalized: qadashim. The problem is that no one pronounces it that way. And that is the reason that I have been spelling the word qodashim rather than qadashim.
The View from Ashkenaz: What’s All the Fuss About?
At this point I need to explain that there is a significant part of the Hebrew-aware community that doesn’t see any problem here. The community of Jews who speak with the dialect of the Eastern European Hebrew heritage (known by the Hebrew designation Ashkenaz) pronounce the vowel qametz something akin to the aw in claw. They are consistent in this pronunciation: each sign that looks like any sort of qametz is pronounced the same.
Speakers of Israeli Hebrew, Jews from the Mediterranean and Islamic regions of the world use the variants on qametz that we are discussing and this has become the dominant pronunciation. But it should be conceded that one reason there may be problems here is that the group of scholars who invented the sign qametz almost certainly spoke something more akin to Ashkenaz and therefore did not distinguish among the various forms of qametz. How do we know this? Because they used the same sign for all the variants. They could just as easily have chosen other signs. So Ashkenazi Jews are not without a serious argument when they ask what the fuss is about here. I’ll return to this concept at the end of our journey.
The reason is that the Tiberian Masoretes were not the only group of scholars annotating the text of the Bible. There were at least two other groups (deemed Babylonian and Palestinian in most handbooks) whose diacritical marks suggest that they had a pronunciation more akin to the non-Ashkenazi Jewish community. And even without that evidence, we can’t simply ignore the fact that there is a substantial, important and lettered community of people who have faithfully transmitted a dialect of Bibilical Hebrew which has three separate phonemes for the sign qametz.
Back to Basics: The Segholate Nouns
Notice that the word preceding qodashim is qodesh. The grammatical form of qodesh is well understood. It belongs to a category of nouns termed segholate. The segholates are among the oldest base forms in Hebrew. What I mean by this is that a number of core, common Hebrew words are in the category. For example: yeled (child), melekh (king), shemesh (sun). All of these words began as monosyllabic words containing an a vowel. The original Semitic word for child was yald. King was malk. Sun was shamsh.
In the course of time speakers of Hebrew developed a dislike of consonant clusters. Notice that each of the Hebrew words above ended in two consonants with no vowel between them. Hebrew speakers began to insert just a bit of a helper vowel between the two consonants. Malk became malik. Shamsh became shamish. Eventually, the 2nd (newly created) vowel started to exert some influence over the first, original vowel. People liked to homogenize the two vowels. So in time (and in fact by the era of our Biblical texts) yalid became YEled, malik became MElekh, and shamish became SHEmesh. The reason I capitalized the letters was to show you where to put the accent or stress on the word.
This placement of the stress seems to violate the rules of Hebrew pronunciation, which prefers the accent on the last syllable. But in this case, by retaining the accent on the first syllable, the Hebrew speaker was holding on to the original word. If you think about it, where was the stress on the word malk? You see, all these words had only one syllable, so there was only one place for the stress!
One of the interesting aspects of segholate nouns is that they have a particular pattern for forming plurals. Lets look at a few of these: yeled becomes y’ladim; melekh becomes m’lakhim; regel becomes r’galim. In other words, the eh/eh vowel sequence changes to shva followed by an a vowel in the second syllable. But the phenomenon I’m pointing to is that the very first vowel shrivels up to sh’va. The reason for this is simple: in the plural, the natural inclination of Hebrew to accent the final syllable overpowers that first syllable. As the speaker rushes to get to the third syllable, the first vowel is slurred over.
We can now turn out attention to the word qodashim. What we expect based on the examples above is q’dashim with that first syllable shrinking to sh’va. Part of the explanation for why this does not happen lies in the fact that there are actually three types of segholate nouns as we mentioned above. The most common type is the eh/eh we demonstrated. All of these are based on ancient groundforms which began with the a vowel. Some segholates originate in a word that featured the i vowel. And what we have here is a word whose origin goes back to an o or u vowel.
One phenomenon we see demonstrated here is that it is more difficult to slur over the o/u vowel than the a vowel. People remember that o sound and continue to express it.
A second issue here is that the qof (quf) is one of the more striking consonants in the Hebrew language. Although modern speakers do not differentiate the quf from plain k sound of the kaf. But the original sound of the qof was produced deep in the throat–a kind of clicking sound. When a consonant is produced in the throat, most speakers will retain a vowel to help them pronounce that sound. In other words, it is more difficult to slur a throaty consonant than others.
So we have two reasons to preserve the original sound of the vowel. First, the “stickiness” of o vowels and second, the need to assist the speaker in pronouncing a throat consonant.
For these reasons we expect to see the word qodashim. But when we look at text, we do not see that. We see what appears to be qadashim. Beneath the qof we find the qamatz. Some knowledgeable speakers might say that this is no problem. We have the qamatz qaton, a vowel that is written exactly like a regular qamatz, but is pronounced o as in hole. The trouble with that explanation is that the qamatz qaton has a specific spelling convention and this instance does not agree with that convention. Next to the qamatz we see a diacritical mark called a metheg. That mark, shows that the qa is an open syllable. A qamatz in an open syllable is an ordinary qamatz, not a qamatz qaton.
What is going on here? Should we ignore our inclinations and pronounce the word qadashim rather than qodashim?
At roughly this stage of the game I decided to have a look at my copy of Biblia Hebraica, the most commonly used text for Biblical Hebrew scholars. This text has undergone something of an evolution. One of the earlier versions was prepared by Rodolph Kittel, who became infamous by assuming a position at a German University after ensuring that his Jewish forebear was dispatched to a concentration camp. As you might imagine, Jewish scholars have never been enamored of consulting that version, and so it came as a great relief when a new version, vastly improved by consultation to original manuscripts, was published as Biblia Hebraica Stuttgartensia (BHS) in the 1960s. This is the version that I consulted.
In Lev 6:10 we have the first of three occurrences of the term qodesh qodashim in the chapter and I found qametz followed by metheg. But in 6:18 and 6:22 something different! In both those verses, the qametz hatuf (aka hataf qametz) is printed. This vowel would have been very easy to explain and I happily reported to my colleague that the mystery was solved. The qametz qaton is a simply slurring of the o vowel which would be completely normal for a segholate plural. The first instance must be a case of misprint or perhaps a misreading of the manuscript–after all, these printed versions are all the result of someone staring at a manuscript–and often a poor quality photocopy of a manuscript.
My giddiness at having solved the puzzle was short lived.
One of my study partners immediately informed me that her version of BHS has the qametz/metheg spelling. What’s going on here? This part is not so complicated. The BHS is the premiere text of the Hebrew Bible for most of the scholarly community. As such, there are new editions which correct errors from time to time. My edition happens to be the first (produced in the mid-1960s) and there have been several since. In the more recent editions, the text we are examining was corrected to qametz/metheg.
Wanting just a bit more evidence I turned to the Qoren edition of the Hebrew Bible. The Qoren was the first Hebrew Bible entirely typeset in Israel and it was meticulously prepared from new photographs of the most complete Biblical text, namely the Leningrad Codex. In another article I will discuss the available manuscripts of the Hebrew Bible, but for now, all you need to know is that the Qoren text also has qametz followed by metheg.
Bottom line: my simple and elegant solution was pulverized.
Is there another solution? As it turns out, not really. In fact, the great grammarian of the classical Hebrew language, Gesenius, lists this spelling and simply finds it to be an exception. Grammatically it appears to be a qametz that should be pronounced a as in father. But no one would pronounce it that way.
As I mentioned earlier, Ashkenazi Jews would pronounce it aw. Among non-Ashkenazi Jews, it is essentially universally pronounced o. And there is a good reason for this whatever the spelling: the close proximity of this word to the base form qodesh. When seeing the phrase qodesh q?dashim, it is almost impossible to pronounce it any other way!
I have a personal theory as to why this spelling never bothered the Masoretes. I believe that the pronunciation of the Tiberian Masoretes was closer to the Ashkenazi than the Sephardi pronunciation. If so, they would have heard little difference–they would say qodesh qawdashim and that would have worked well for them. Some day I hope to be able to examine this passage in a manuscript vocalized in the Babylonian system. But that will have to wait for another day.