User Tools

Site Tools


diacritics

Diacritics

The so-called “Masoretic Bible” is the worlds oldest intact Hebrew scripture. Older versions do exist, but only in very fragmentary forms. The Dead Sea Scrolls are an example of these older versions. It is important to note that these older forms are substantially different from the Masoretic texts, particularly in one very important regard.

The Masoretic spellings of every Hebrew word differ from older Hebrew forms in that they contain additional markings called “diacritics” (a.k.a. “cantillation marks”), which do not exist in the earlier texts. That is to say, the original texts did not include these small additional markings. They were added into the text relatively late - beginning sometime around 500 CE, by the scribes who created the Masoretic version by copying the text from an older version and adding their own diacritics. These versions ended up being adopted by a few highly influential Christian translators, but not by the Hebrew community at large. In fact, modern Torah scrolls used in Jewish synagogues today do not include any diacritics.

Why these scribes added the extra markings is not entirely clear, although there are widely accepted theories. What is clear is that it was a momentous decision, due to the fact that the additional markings that they added were adopted into many of the later versions. Thus the decisions of these scribes forever altered the spelling of almost every word in the future Christian translations of the Hebrew Bible. The diacritics also made their way into the modern Hebrew language (but not the actual scriptures) and have become a standard part of the Modern Hebrew grammar.

Many scholars are fond of pointing out that the ancient Hebrew written language had no vowels. They claim that the diacritics were added in order to clarify which vowel sounds belong within each word. This is patently false for two reasons:

  1. the ancient Hebrew, once diacritics are removed, DOES contain vowels.
  2. the diacritics and the vowel sounds do not match up.

The following sections will discuss this further and back up these two claims with evidence.

The Ancient Hebrew Vowels

It is often stated that ancient Hebrew contained no vowels. This is quite incorrect. The following four vowels are found widespread throughout the ancient Hebrew lexicon.

Hebrew Vowel Rough Pronunciation Diacritic Dependency Rough English Equivalent
א various “a” to “eh” sounds as in “bar”, “bare”, “bear”, “behr” Partial A, E
עvarious “a” to “eh” sounds Partial A, E, Y
י “i” to “ee” sound No I, Y
ו “o” to “oo” sound No O, U, V

In English, we are accustomed to using five primary vowels; A, E, I, O, U. The four Hebrew vowels do not quite align with ours, but there do exist some parallels as seen in the above table.

The most straightforward of the Hebrew vowels is י (named 'yud'). It is essentially the same as the English letter 'i', though its sound leans closer toward an “ee” or “eey” sound.

The letter ו (named 'vav') represents the consonant “v” in modern Hebrew but also represents the vowels sounds “u” and “o”.1) Even so, these patterns should not be thought of as rules and it is quite easy to find examples which “break the rules”.

Take the word “Elohim” for example. Based on the well-known pronunciation, we should expect to find a ו in the word, to make the “o” sound. In a minority of instances this does in fact occur, but in the vast majority of instances, “elohim” is spelled אלהים leaving out the ו, so it technically should be pronounced “el-him”. These kind of spelling and pronunciation problems occur all throughout the texts, affecting the majority of Hebrew words, and often foiling the work of the translator who seeks a consistent application of meaning to a consistent base of words. See Hebrew Spelling Inconsistencies

The letters א (named 'aleph') and ע (named 'ayin') represent “a” and “eh” sounds, but the manner that they do so is somewhat inconsistent and depends on diacritics for clarification. In actual fact, the differences between these two sounds may be less important than one might imagine, due to the distinction only having arisen after the Biblical period, as discussed at https://en.wikipedia.org/wiki/Biblical_Hebrew#Phonology, and particularly as affected by the so-called “Canaanite Shift” which occurred roughly contemporaneous with the Patriarchal to Mosaic periods. Essentially, what this means is that there are many cases where the sounds are interchangeable as far as semantics goes. It is essentially different accent but does not alter the meaning. See a vs e

Besides these four vowel letters, there are many cases in Hebrew where a vowel sound is implied between two consonant sounds but there is no letter to indicate it. As already discussed, “elohim” is spelled אלהים is an example of such a word, but there are literally thousands of others.

Thus it can be seen that the ancient Hebrew language does contain vowels, but perhaps not as many as we would like. This situation may indeed leave one wondering whether other grammatical features might also add to the pronunciation to assist with the “missing vowels”?

Diacritic Vowel Mismatch

Do diacritics determine the vowel pronunciation of missing vowels? Consider this example:

In this example we have a word made of two consonants without a letter specifically indicating a vowel, but where some kind of vowel sound obviously is needed. In the ancient Hebrew documents all six of these instances would have read שם, and most likely would have been pronounced 'shem'.

Yet, the first three entries are typically translated as 'sham', and assigned a meaning of “there”, whereas the other three are translated as 'shem', meaning “name”. This apparent distinction is somewhat problematic, given that the diacritic-based difference in spelling and pronunciation did not appear in the texts until they were injected during the early Medieval period, at least a thousand years after they were originally written.

The extemporaneous nature of these tiny little diacritic marks that the Masoretes added is highly suspicious. They were supposedly added to aid as a pronunciation guide. Could this claim be true? What evidence might they have had for some supposed pronunciation differences one thousand years previous in a long-extinct verbal dialect?2) But we need not rely on conjecture. Let us examine the diacritics as they appear in situ, and see if they actually do create any discernible variability in the pronunciations.

Immediately we may note that we are given six different diacritic combinations, but only three pronunciations. Clearly, if the diacritics affect pronunciation, the six different combinations should theoretically yield six different pronunciations. But they don't.

Stripping away the ineffectual parts, we do uncover a bit of a pattern here, but we have to ignore half the diacritics to see it. The 'sham' instances have tiny little T-shaped diacritic (named 'kamatz'), and the 'shem' instances have a tiny little pair of dots (named 'tzayray') instead. We need to ignore all the dots on top and the other little swoopy jots.

There appears to be a correlation between an 'e' vs. 'a' vowel, and a tiny T-shaped object and a pair of dots. We have to ignore the other seven little diacritics to make any sort of correlation at all.

So does it hold true that throughout the Hebrew language a little T-shape makes an “a” sound, and a little pair of dots makes an “e” sound? That question is going to take a lot more analysis. FIXME - Let's circle back after more data analysis, as the project continues.

One thing is clear though; out of the nine diacritics examined in this example, seven of the nine had no impact whatsoever on the pronunciation (nor the meaning) leaving only two that might influence pronunciation. This is a far cry from the claim that in general “diacritics determine pronunciation”. That statement, is at best only 2/9ths true. In other words it scores only 22% on the test. In our books, that is a FAIL.

One final note on the subject of diacritics, it has been claimed by some theologians that Jesus referenced these diacritics in Mathew 5:18. Jesus states that “Not one jot or tittle shall pass away from the law.” They claim states that the “jot and tittle” spoken of by Jesus refers to the diacritics found in the Hebrew scriptures which Jesus calls “the law” in reference to “the Law of Moses”. The problem with this claim is this. At the time of Jesus, the Hebrew scriptures did not contain diacritics. They would not be added until at least five hundred years later. Unfortunately, Jesus did not give us the answer as to whether we should attempt to make any sense of the diacritics, since he himself never lived to see their introduction.

Diacritic Tonality

It has been shown that the diacritics generally do not effect which vowel belongs in a given word. So what do the diacritics represent?

The mainstream view is not one hundred percent wrong. The diacritics probably do influence the pronunciation; just not in the way they think. The diacritics quite possibly served the purpose of affecting not the vowel sound itself, but rather the pitch of its vocalization as well as volumetric and other types of accentuation. These are critical aspects of the Jewish tradition of memorization and recitation, that are still in use by conservative Jews to this day.

The diacritics can be compared to the musical notation that goes along with the lyrics. This also appears to be one of the main roles of Hebrew affixes.

Diacritics in no way affect the meanings of the words and only moderately affect the basic pronunciation in spoken language.

Who Cares?

This is only worth bringing up because in today's Christian understanding of the Hebrew Bible, many of the word meanings come from erroneous distinctions made long ago, and substantiated via Strong's Concordance and other exegetical tools. These erroneous distinctions are in large part due to, or at least supposedly backed by, the concept that diacritics change the vowels contain within a word, forming a “different” word and thus allowing spurious semantics to be thrust upon the word. However, in many cases the case for a different word is very weak. In many cases, the root words are ignored, and new root words are proposed out of thin air in order to justify a meaning that better fits the translators presumptions. To some degree, this is likely to be inevitable. All translators have preconceived notions which influence their work. The same is true in the case of our own translation work. This is why we are going out of our way to present transparency in our translation processes, and hence why we feel it is important to discuss openly even technical details which may seem trivial or boring.

Our Process

So how ought we to treat the presence of diacritics in the text? Firstly, considering the late addition of the diacritics by the Masoretes, we acknowledge that the diacritics do not exist in the ancient (original) versions of the Hebrew texts, therefore we are strongly inclined to remove them. However, we also acknowledge that occasionally the diacritics may be useful in determining questions of pronunciation, particularly between “a” and “e” sounds, and that it is at least theoretically possible for two words to exist which are differentiated only by this vowel choice, and therefore there could be a semantic differentiation that we would be remiss to overlook. Therefore, the following metric is employed to yield a standardized simplified syllabic pronunciation rubric.

Hebrew Vowel Bara Transliteration Suggested Pronunciation
י i Usually, “i” as in “ski” or sometimes as in “kick”
ו o “o” as in “snow”, or sometimes “ov” when a consonant is implied (see tehovim for example)
א a usually “a”
ע a usually “a”, “occasionally “yeh” when a consonant is implied
none e implied between consonants

This work has has since been further developed into a transliterative alphabet

2)
Granted, the same question should be asked of this current translation project. One difference is that we are using large-scale data analytics in an iterative fashion to enable exhaustive contextual and etymological analysis with rapid iterations of error correcting algorithms.
diacritics.txt · Last modified: 2024/09/25 22:04 (external edit)

Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki