Killing a Dead Language

A Case against Emphasizing Vowel Pointing when Teaching Biblical Hebrew

William P. Griffin

This is a simplified webpage version of the original of this article, which is found at:

The typical Hebrew grammar devotes a great deal of space to the intricacies of Masoretic pointing, and many who teach elementary Hebrew have similar emphases. Students are taught propretonic reduction, the rule of shewa, and countless other rules of vowel patterns, each with many exceptions. "Why does this have a patach instead of a qametz?" They then move to verb paradigms, where they encounter further lists of exceptions, when to include the dagesh, variations for gutturals, and so forth. And they have only encountered the Qal. Our students, who by and large are anything but linguists, encounter statements like, "The verb 'rr differs from sbb in one way. In the second and first position forms, singular and plural, the Geminate consonant rejects the Daghesh Forte and the Pathach under the [aleph] becomes Qamets due to compensatory lengthening."[1] "When there is an unaccented syllable followed by a . . . ," and so forth. Next comes the Piel with its multitude of variations. These students are not merely expected to know that such vowel patterns exist, but they are expected to be able to reproduce them as well. We then wonder why people regard Hebrew as a "hard language" and why attrition rates are so high.

The answer is this: We are not teaching one language, but two; if we emphasize the accents, make that two and a half. We demand that our students understand and replicate an elaborate cellophane overlay that is more complicated than the language it attempts to clarify.

I propose that we peel back this layer and focus on what is underneath. To many, the notion of de-emphasizing Masoretic pointing probably sounds foolish at best or heretical at worst, and one suspects that this has never even occurred to others. Yet I would urge that we consider doing so for the following reasons: first, vowel pointing, historically, postdates the latest of the original texts by at least a thousand years — thus Tiberian pointing, while possibly in the ball park, represents one pronunciation system out of many, and one that cannot be demonstrated to be the "biblical" pronunciation. In fact, there is much evidence that Tiberian pointing, from the standpoint of the history of the language, contains enough flaws to call into question its accuracy. Second, students can read biblical Hebrew with comprehension and translate with accuracy without learning the intricacies of Tiberian pointing. This simplifies and speeds up the learning process.

Historical Reasons for De-Emphasizing Tiberian Pointing

"Biblical Hebrew" is the name applied to the language(s) contained in a collection of texts that were written over a period of about a thousand years, spanning the second and first millennia B.C.E., by people in diverse regions and settings. Originally, it was written in what is called "Paleohebrew," a script much different from current Hebrew texts.[2] Dots were employed as word separators,[3] and there were no distinctions between medial and final forms. The language was written without vowels. From about the sixth century B.C.E. on, many consonants were added to represent vowels (aleph, he, vav, and yodh),[4] known as matres lectionis, "mothers of the reading."[5] Sometime after the Babylonian captivity, a cursive form of the Aramaic script was adopted that by and large replaced the Paleohebrew script[6] and forms the basis of what we see as the Hebrew alphabet today (with the addition of final letters for kaph, mem, nun, peh, and tsadeh).[7] Most of the so-called Dead Sea Scrolls are written in this square script, although some are in Paleohebrew, and still others are in the square script but use Paleohebrew for the divine name. Many of these Qumran texts add even more consonants as vowel indicators, especially the aforementioned yodh, vav, he, and sometimes aleph.[8]

During the second half of the first millennium C.E., a much larger revision was undertaken: the use of indicators around consonants to represent vowels, known as "vowel points."[9] At least three vowel pointing systems emerged: the Tiberian (ben Asher), the Babylonian, and the Palestinian.[10] These new pointing systems preserved pronunciation as understood by various groups at that time, as well as interpretations where ambiguities and debate existed. Accents were likewise added, aiding in interpretation and the liturgical reading of the text.

Some readily acknowledged that these vowel pointing systems were neither original nor historically accurate. Others viewed pointing as another revelation given by God on Mount Sinai. One way or another, Tiberian pointing, as represented by the Masoretic text, appears to be the most popular as early as the tenth century C.E. and won the day by the thirteenth century C.E.[11] Versions of this text have been enshrined in printed Hebrew Bibles since.

The point of the above summary of the early history of Hebrew is this: There are a variety of reasons, based on the history of the language, to call into question a rigid adherence to Tiberian pointing:

1. Strictly speaking, classical biblical Hebrew is a dead language. It is as dead as Akkadian, Ugaritic, Sumerian, or koiné Greek. We are dealing with a fixed set of texts,[12] not the Hebrew spoken in the nation of Israel today. Israeli Hebrew has significant differences from biblical Hebrew. While the verb forms are by and large biblical, "not only were all the archaic forms of BH rejected . . . but also the consecutive tenses, the cohortative, the infinitive absolute more or less, and the infinitive construct (except for the plus [lamedh] and plus [bet] infinitive forms."[13] The syntax is heavily influenced by a combination of Mishnaic Hebrew and European languages;[14] some biblical vocabulary is either rejected or substantially modified as to its usage; and "it has been calculated that some ten percent of the words in Hebrew dictionaries might be of foreign, usually Western, origin."[15] Further, the pronunciation system is basically Sephardic, not Ashkenazic or Yemenite.[16]

2. However well the Masoretes preserved a pronunciation and interpretive tradition, vowel points are not original. This should be important, whether one views the Hebrew Bible as Holy Scripture (it is only the consonantal text that was canonized) or sees the text as a source of information for reconstructing the history of the religion of Israel. People wrote, read, and learned Hebrew in ancient times up through the latter part of the first millennium C.E. without vowel pointing.

3. Similarly, Masoretic pointing should be treated with a skeptical eye, whether one uses it for pronunciation or interpretation. The Masoretes applied a fairly consistent pronunciation system to a collection of texts that cover (depending upon your perspective) up to a thousand years. It would be difficult to argue that any language maintained the same pronunciation over a thousand year period (or even five hundred years), let alone Hebrew. Further, even consonantal indicators in the Hebrew Bible, such as sibboleth/shibboleth (samech/sheen [Jdg 12:6]) show that pronunciation differences existed at the same time in different regions.

4. Even if one concludes that Tiberian pointing is in the ball park, some Greek sources provide evidence that some ancient vowel patterns do not match that of the MT.[17] First, the vocalization of proper names in the LXX sometimes reflects different pronunciations (e.g., often a for i; o for a or u; monosyllabic pronunciations of segolates;[18] Jerusalem is pronounced "-em" not "-ayim"; "baithel" instead of "beth-el"; "melkisedek" not "melek-tsedek"). Second, remains of the second column of Origen's Hexapla (third century C.E.) show pronunciation differences (e.g., e for i; o for u).[19] Finally, while Jerome's transliterations (fourth to fifth centuries C.E.) reflect a tradition closer to the MT, there nevertheless are significant differences.[20]

5. Masoretic pointing preserves a liturgical reading for a text that, by and large, was not intended to be read liturgically. This includes odd features such as "in pause"; i.e., where pronunciation changes because of a word's position in a verse (e.g., lak for leka [which looks like a feminine form being used in place of a masculine one]).

6. Once again, the Tiberian pointing in the MT is only one of three pointing systems (which indicates that there was disagreement over what was the "correct" pronunciation). Just because it happened to win broad acceptance does not mean that it should be uncritically accepted as authoritative.[21]

Thus, I find it baffling that many graduate schools renowned for their critical scholarship nevertheless demand a thorough knowledge of Tiberian pointing.

Practical Reasons for De-Empahsizing Tiberian Pointing

I was trained in the pointed Hebrew of standard grammars. My first encounter with unpointed Hebrew was when, while working on my M.Div., I took a doctoral seminar on Isaiah. My assignment was to compare the MT to 1QIsaa. The bareness of the text was a surprise to me, as well as the consonantal spelling differences. My next encounter was while working on my Ph.D. I did all my translating for a seminar on Hosea, Amos, and Micah from an unpointed text that I had on computer. I laid out the text in a logical fashion, and it was common for me to come to class with a different reading than that of the other students. (In a number of instances, it was apparent, in light of parallelisms, that Tiberian scribes made mistakes.) A while later, I did reconstruction and translation of biblical Dead Sea Scrolls, which, of course, are all unpointed. I frequently assign unpointed Hebrew to my students on all levels (elementary, intermediate, and advanced). While it tends to look odd to anyone whose first year was entirely with pointed texts, many get so that they prefer it. I now regularly read unpointed Hebrew and encourage my students to do the same.

Here are some practical benefits of reducing an emphasis on vowel pointing:

1. The learning process speeds up tremendously. I can use the same grammar and teach it at almost twice the pace as in past years.

2. The confusion level is greatly reduced without sacrificing translation precision.

3. Rules for noun plurals are greatly simplified. From a consonantal perspective, most changes involve affixing -im and -ot. Students must also know changes such as the loss of the final he or the shift from he to tav, but this is considerably simpler than learning propretonic reduction and the host of other vowel changes.

4. One of the greatest benefits might concern the conflation of the multitude of verb paradigms into a few. For example, Lambdin's grammar[22] contains at least sixteen different Qal perfect paradigms. However, there are only six different consonantal patterns; if one notes that monosyllabic verbs tend to have two consonants, this could be viewed as only 5 patterns. As for the Qal imperfect, Lambdin lists at least twenty-seven different paradigms, but with only ten different consonantal patterns (one of these, the first aleph, has only one form that is different — the 1cs). Most differences with paradigms involve the disappearance of nun, yodh, he, and the loss of the duplicate consonant in geminate verbs. (As for the example of the geminate verbs that I cited above, 'rr, unpointed, does not differ at all from sbb.) By emphasizing unpointed patterns, students have to learn only sixteen patterns instead of forty-three. In addition, if one compares all Qal forms to all Piel forms, the only consonantal difference is the mem prefixed to the Piel active participle. Since verbs tend to prefer either the Qal or the Piel (with both feeding into the Hiphil), translation confusion between the Qal and the Piel, while possible, is probably rare. All this greatly reduces the amount of information students must learn, without sacrificing reading comprehension or translation accuracy.

5 . My experience has shown that students more easily recognize roots. Similarly, a consonantal focus seems to help students who have learned one form of a word (e.g., the noun) to recognize another one that they have not yet learned (e.g., the verb).

6. Students can easily learn to read unpointed Hebrew if they are exposed to it from the beginning. While it is more awkward if they become dependent upon pointing, I have found that students can make the transition fairly quickly at a later time.

7. Students will be more prepared to read Qumran texts, rabbinic Hebrew, or modern Hebrew, if and when they make those transitions.

In sum, the greater efficiency is likely to attract more students to the language while keeping more marginal students in our classes as well.

Excursus: Why do People Take Hebrew?

I was struck by Charles Isbell's 2005 article in The SBL Forum, "The Hebrew Teacher: Guru, Drill Instructor, or Role Model?" He notes that there are a variety of reasons people take Hebrew and that these differ greatly for Jews and Christians. For Jews, they are mainly interested in using the Hebrew Bible in worship services, and they want to leave the door open for learning modern Hebrew. Christians, on the other hand, learn Hebrew primarily because they want "to translate and interpret the Hebrew Bible so as to buttress their understanding of their own faith" and are rarely interested in learning the modern version.[23]

Isbell's comments provoked me to examine what my students expect to do with the language. His observation that Christians "want to learn Hebrew in order to do something else that they consider more important — interpret the first part of their Bible,"[24] can be broken down into at least three motivational subgroups. First, there are students who wish to prepare for graduate school, who plan to acquire other languages as well. Second, there are those whose primary goal is to be more informed when they prepare sermons and teachings. These two motivations dominate most professors' approaches to teaching the language. However, Isbell noted a third reason, which I (and, I would argue, the bulk of gentile Hebrew professors) have neglected: for the enjoyment of being able to read the Word of God in the original language. These three goals are not mutually exclusive, but their motivations should influence how we teach the language. Neither the "for preaching" or the "for enjoyment" groups are greatly benefited by an emphasis on pointing. The "preparation for graduate school" group does not need to emphasize a knowledge of pointing unless they expect to attend a school that emphasizes pointing. From a practical standpoint, these students will be at a level of Hebrew expertise and motivation where they can go through a grammar (such as Lambdin or Seow) during the summer before matriculation and learn all the fine points of pointing.


My oft-stated goal for my students is that I want them to read with comprehention, translate with accuracy. With that outcome in mind, I have radically changed my approach to teaching elementary Hebrew. I am engaged in what I will call a deductive-immersion approach. It is deductive in that I expect them to know vocabulary, consonantal forms, syntax, and grammar (vowels aside). It is immersion in that I have them translate a lot more Hebrew than I used to have them translate (and this includes unpointed Hebrew).

For pragmatic reasons, I teach my elementary students how to pronounce the Tiberian vowels. My students learn a modern pronunciation, which, while preferred by many teachers, is also not strictly Masoretic.[25] For example, of the of the BGDKPT letters, modern Hebrew has only hard pronunciations for gimel, dalet, and tav.

I sift a lot of details out of the grammars I have used in recent years. I am not making them learn all the details of propretonic reduction, the rule of shewa, and the like. Instead, I tell students that vowels tend to change when prefixes and suffixes are added to words, and often in unpredictable ways. I also regularly give them unpointed Hebrew to translate with almost every lesson, and they do so with the same ease and precision as pointed Hebrew. The translation on their exams is half pointed and half unpointed, and I have not detected a significant comprehension difference. Many of these elementary students prefer to work from unpointed texts.

My students are required to learn the vowels for forms such as the strong verb and some pronominal suffixes. This is mainly for pronunciation purposes. For most other verb forms, they must know consonantal changes, but if the consonantal forms are the same as the so-called strong verb, I note that and we move on. I require students to learn consonantal paradigms for verbs that lose or replace consonants (such as third he verbs and ntn).

In comparison to when I emphasized pointing, there is less confusion in class, students are translating better, and their pronunciation is probably about as good at this point as it was back then. The average grades are much higher, and the bottom students are in the low C realm instead of failing. Their vocabulary knowledge is better, and about a quarter of the class learned the Paleohebrew alphabet for extra credit. And in all of this, they are doing less homework than I used to assign.


Students do not need a thorough knowledge of Tiberian pointing if the goal is to read with comprehension and translate with precision. Instead, the use of Tiberian pointing should be limited to aiding students with their pronunciation, since "proper" vowel pronunciation is not necessary for translation accuracy. Finally, Tiberian pointing should be seen in a similar light as modern commentaries; i.e., as an informed opinion that, while valuable, should always be viewed with skepticism, especially when ambiguities exist in the consonantal text of the Hebrew Bible.[26]

William P. Griffin, Evangel University

[1] Gary D. Pratico and Miles Van Pelt, Basics of Biblical Hebrew Grammar (Grand Rapids: Zondervan, 2001), 157.

[2] Many authors use this term to describe the ancient script, but Mark D. McLean ("Hebrew Scripts," ABD 3:95) reserves this term for later use of what he calls "the Hebrew script."

[3] See. inscriptional evidence from Tel Dan, Mesha Inscription (Moabite Stone), and the Siloam Inscription. See also Emanuel Tov, Textual Criticism of the Hebrew Bible (Minneapolis: Fortress, 1992), 208-9.

[4] Frank Moore Cross and David Noel Freedman, Early Hebrew Orthography (New Haven: American Oriental Society, 1952), 58; see especially their excursus, "Orthographic Problems in the Massoretic Text," 65-70.

[5] Instances where one would expect these fuller forms, but without them occurring, have been labled "defective," when in reality they are more original spellings.

[6] "The change from paleoHebrew to the so-called square script in writing the Torah is dated to the time of Ezra." (M. Greenberg, "The Stablization of the Text of the Hebrew Bible, Reviewed in the Light of the Biblical Materials from the Judaean Desert," in The Canon and Masorah of the Hebrew Bible [ed. Sid Leiman; New York: Ktav, 1974], 301).

[7] Tov says, "During the Persian period final forms of letters . . . gradually developed but were not used consistently." A lack of consistency in the use of these letters is also reflected in several biblical and non-biblical texts from Qumran (Emanuel Tov, The Text-Critical Use of the Septuagint in Biblical Research [2d ed.; Jerusalem: Simor, 1997], 145).

[8] For a discussion of orthographic characteristics of the DSS, see M. Greenberg, "The Stabilization of the Text of the Hebrew Bible, Reviewed in the Light of the Biblical Materials from the Judaean Desert," in The Canon and Masorah of the Hebrew Bible (ed. Sid Leiman; New York: Ktav, 1974), 298-326.

[9] "The various systems of accentuation and vocalization introduced in to the text of the Bible by the Masoretes had started to develop by the about the sixth or seventh century CE" (Angel Saenz-Badillos, A History of the Hebrew Language [trans. John Elwolde; Cambridge: Cambridge University Press, 1993], 77).

[10] Bruno Chiesa, The Emergence of Hebrew Biblical Pointing: The Indirect Sources (Frankfurt: Lang, 1979), 9-16.

[11] Chiesa, Emergence, 5-8.

[12] This is true even if new texts from that time period are discovered.

[13] Eduard Kutscher, A History of the Hebrew Language (Leiden: E.J. Brill, 1982), 197.

[14] Kutscher, History, 196, 222.

[15] Saenz-Badillos, History, 286.

[16] Kutscher, History, 226.

[17] Cross and Freedman (Orthography, 53) state that "many scholars (including the Massoretes) have generalized a particular vocalization, artificially forcing the common speech and the elevated language into the same linguistic mold, and incidentally playing havoc with recognized orthographic principles."

[18] Kutscher, History, 106-7.

[19] Ibid., 143; but Kutscher goes on to say, "Although the vocalization of the Masoretes is known to us only from a period about 600 years later than that of the transliterations, it faithfully preserves older forms." He refers to the pronunciation of the LXX and the Hexapla as "substandard," which is difficult to prove.

[20] Ibid., 145.

[21] As Harry M. Orlinsky has said, "But the question asks itself: What is there inherently in the Masoretic work of the Ben Asher school that gives it greater authority than that of the Ben Naftali school? Why should the vowels, the dagesh, the maqqef, the raphe, the metheg-ga`ya, the accents, the hataf, and the like, as used by Ben Asher's school be more acceptable to an editor of 'the' masoretic text than their use by Ben Naftali's school?" ("The Masoretic Text: A Critical Evaluation," in The Canon and Masorah of the Hebrew Bible [ed. Sid Leiman; New York: Ktav, 1974], 862.)

[22] Thomas O. Lambdin, Introduction to Biblical Hebrew (New York: Charles Scribner's Sons, 1971).

[23] Charles David Isbell, "The Hebrew Teacher: Guru, Drill Instructor, or Role Model?" SBL Forum, 8-9 [posted 10/20/2005 - 11/30/2005]. Online

[24] Isbell, "Hebrew Teacher," 9.

[25] A modern pronunciation will help students if they wish to learn modern Israeli Hebrew. I also teach them Hebrew cursive script. However, a thorough knowledge of pointing is also unnecessary for learning modern Hebrew — my wife took a year and a half of the language without ever learning vowel pointing.

[26] I would like to express my appreciation to Dr. Wave Nunnally for his valuable insights and critiques on this issue.