The historic sound of the sixth consonant of the Hebrew Language
One of the more controversial questions on Hebrew phonetics in traditional circles is on the pronunciation of the letter ו: was ו pronounced as a V or W? This debate has been around for a long time, and each side stands mostly on the grounds of Modern Hebrew and traditional pronunciations. In many cases, the debates leave the world of academic study and end up becoming a theological argument.
Due to this, I must stress the following point:
Before discussing anything, it is important to define the parameters of the question. The subject being looked at here is not about vowel points and is not about the state of the language in medieval or modern times. What is being examined is the question of “is there evidence for the existence of one specific sound or the other in earlier stages of the Hebrew language?”. By saying this I am trying to avoid two problematic issues:
- Basing oneself on late and modern pronunciations attested in Jewish sources and tradition.
- Avoiding the miss-association some make between the way some words are said based on the consonants vs. vowels. It has become common to try and argue that if one says W, then automatically it must be a specific vowel setting for the name YHWH (specifically Yahweh vs. Yehovah). Based on this, the discussion about vowel points has nothing to do with the sounds of the consonants and should not be confused with the phonetics of the consonants.
To determine an answer to the above question, there is a need to examine multiple areas of study which can help determine if there is a clear answer on the state of the language in earlier stages.
The overall reality
In a basic done on the subject of the sound of ו conducted in many Jewish communities, it is common to assume that in Arabian countries where there were Jews the sound of the consonant ו could be W or V. In European countries, the common attested sound is of a V. The pronunciation of V is the common one used in modern Hebrew, with some words that use the W sound. Modern Hebrew by most bases its consonantal practice on the used in Europe, while the vowels are based on the Sephardic/North African practice.
Masoretic Hebrew attested in 10th-11th centuries show that in Tiberius most people used the V sound, however there were exceptions due to specific phonetic functions. When ו was a consonant it was a v, but when it functioned within a vowel it was a W. On the other end of the Jewish world we find that Jews of Iraq contested this practice and only used a W sound (Khan, 2019). Another thing to consider is the evolution of sounds. For example: in Ancient Greek there were two periods- in the Mycenaean ear the sound was W. However, in time it changed and in classic and modern Greek it is a V. The pretense of a V sound in Hebrew maybe a result of exposer to Greek of Aramaic.
Phonetics and Comparative Semitics
Traditionally ו is assigned a lip vocalization together with ב, מ, פ (B, M, and P). If ו was a V then it should be mostly seen as a dental sound, and less a lip. The fact is that the V sound is considered to be a dental sound or dental with partial lip, but not a lip alone. Because the assigned related group of sounds for ו are not of a dental but lip origins, it is difficult to accept the partial dental sound as the dominant one for ו. Another point to consider is that ו does not belong to any of the dental sound shifts attested in Semitic languages, hence a dental sound V is improbable[1].
In comparative linguistics for the Semitic family, both sounds are attested in related languages, however, W is considered to be the most common and historical sound between the two. This is the reason why most scholars would say the W sound is the older historical sound used for ו[2].
Could there have been a dual sound?
It is a well-known fact that Hebrew can use the same symbol for more than one sound. This is a reality even today with the dual sounds for ב, כ, פ. By adding a Dagesh (small dot) in these consonants, one can change the phonetics of the consonants from a soft V, Kh, and F into B, K, and P. The same goes for dual sounding consonants which did not use a Dagesh in the past but still had a double sound. This is, for example, the case of ח, ע which had two glottal sounds.
It is worth considering the theoretical possibility that under certain rules ו would have been sounded as W, and under other rules, as V. In some reading traditions from North Africa, we find similar rules when ו is a conjunction and it is sounded as a W.
There are, however, several points to consider about this section:
It has been stated many times in early Hebrew grammars written by Jewish scholars that there cannot be duplicate sounds between different consonants[3]. This means that the V sound can only belong to ב or ו. Because this statement usually appears in medieval writings, a time where the V sound is already known as being a sound for ב, one can conclude that these writers are saying that ו was a W sound.
Another side of this is the fact that we also find in Medieval Hebrew a switch between ב-פ in places where the sound should be soft. Such is the case of the switching of הפקר-הבקר (Hefqer-Hevqer) in Mishnaic Hebrew, and there is no evidence that both ו and a soft פ were both interchangeable with ב when without a Dagesh. It seems more logical that when the switch between ב-פ happened it was between V-F which are both dental, while ו-ב would be between W-B which are both lip[4].
A similar exchange can be found in the Arad letters where sometimes ב and פ switch (B-P). At this stage of the language, it is uncertain whether or not there were soft sounds, making it highly plausible that the switch being between two lip sounds[5]. Whatever the case may be, the writers associated these two sounds B-P or V-F (assuming the softer sound existed at the time) as closely related, however not ו which seems to belong to a slightly different group. As a result, it would seem for now that a W sound would fit this category of lip sounds.
Diphthongs
One of the more intriguing issues with Hebrew phonetics is the discussion on the creation of the o sound in Hebrew. It is understood that all Semitic languages have three basic vowels: a, u, and i. Each language in time adopted other sounds by contracting diphthongs. In the case of the vowel o, it is understood that it came to be by a contracting of the diphthong AW. If the ו was a V sound such a contraction would probably not allow the creation of the o sound. We also find such a reality to this very day when ו is used as a conjunction with Shuruq. Instead of sounding a V with an o sound, the ו moves into a U sound. This U might be a hint to a pre-existing W sound[6].
An interesting example of such a case is the German tradition of creating a diphthong in a select number of words. A specific case I have personally heard is with the word עולם-Olam. When this word is used in the German tradition the tendency is to create a diphthong OW-OwLam. This could be a remnant of the AW diphthong, though the evidence is inconclusive at the moment.
ו and י
There are two aspects of the issue Hebrew has with both consonants. The first is a graphic problem, while the other, which more central to our discussion is the phonetic one. The graphic problem stems from the similarity in form caused by epigraphic practices. Though this is less an issue for the phonetic discussion, the fact that both consonants are commonly confused in MSS, and especially in the DSS (w y) where sometimes the scribe gave a the י a longer body. This gives place to the possibility that the creators of the square Alphabet paid attention to the phonetics of the consonants, which gave, in turn, the graphic similarity between the consonants.
In the phonetics of Hebrew, it has been recognized that ו and י share a half sound property. Being a half sound (or semi-vowel) means that they do not create a distinct cut on the speech organs. Both consonants only brush on the area they originate from, making them soft and partial. This can only be achieved if ו is a W sound that only partially touches the lips. In a dental sound, we have a full cut which is not what ו is supposed to do[7].
Another side to this issue is the fact that both consonants tend to interchange in Hebrew when compared to most North-West Semitic languages when appearing at the beginning of the word, demonstrating the strong link between the sounds[8]. This is also common in Masoretic text cases where one of the consonants changes into the other based on the vowel point settings[9]. This relationship is so vast that many if not most of the cases which appear in the MT are of the ו-י type[10].
Another testimony to the weakness of these two consonants can also found in both Hebrew and Aramaic use of them. In both languages, we find that the partial sound created by ו and י led them to a point where they lost their position as a consonant, and shifted in many cases to functions as matres lecations-vowel consonants. This process is more distinct in these two constants than the cases of א and ה which by most lengthen the sounds, but do not create them. Though אand ה do represent specific vowels, they are employed differently by not specifically assimilating into their diphthongs to produce vowels. The shift to a vowel-consonant can only be achieved if the consonant, to begin with, was already partway a vowel[11].
Both sounds were and are very close and originate from almost the same place, but have a slight distinction between them. This is demonstrative of the use of close and similar sounds in the text, but still distinctive of one another. Hence, the switch between ב-ו is not necessarily between sounds that are exactly the same.
The Cases of ב-ו interchanging
It is important to first note that the cases of ב-ו changes in the Biblical Hebrew text are extremely rare (3 cases in total), and in my opinion, should be used with caution as evidence of the state of the language in earlier times, or during the time of the writing of the Aleppo codex[12].
It has been argued (Gordon N.in personal communication) that in the case of the word גוך/גבך in Ezekiel (10:16 גבהם, 23:35 גוך) we can find a switch between וand ב which indicates that Vav was a V[13]. This argument is based on the rule of Beged Kefet mentioned earlier where ב, ג, ד, כ, פ, ת can switch from a soft sound to a hard sound by inserting a Dagesh. The problem with this argument is that the spelling might not reflect such a reality.
In cases where we find the ו there are two issues:
- The consonant in all appearances of the word (1Kings 14:9, Ezekiel 23:35, Nehemiah 9:26) has a Dagesh in it which may indicate a harder sound for both the ו and ב. This is known as Dagesh Forte and is used to increase the sound of the consonant. In this case, it seems two ו were assimilated into one, creating a heavier sound that is closer to the B sound. This same reality is expressed in the ב which in such a case has a heavier sound and would resemble a W sound and not a V.
- In the compared cases of ב with suffixes, the ב is always the harder sound B and not V. Hence a heavy W would be closer in sound to a B and not a V.
Another problem that stands before us is the sources the Masoretes used for their text. It is possible that they had different texts, and made decisions based on different rules which allowed them to choose from variants of the text. Hence the interchanging ב-ו could be a result of different texts which at the time of their writing did not have different sounds for these consonants, which meant that either one could be used. As a result, the Masoretes did not or could not allow themselves to change the text even if they knew it was a mistake. It also does not seem they had variants in front of them, and the text actually contained this switch. If there was a difference in version this would have resulted in a Qeri and Ketiv situation with a note telling us it is spelled one way, but read another.
An interesting case of interchanging similar sounds which can be also found in Ezekiel is the switch between כ-ק. In the case of the word כובע/קובע (hat or helmet) we find the same word spelled differently in the same book (Ezekiel 23:10&38:5 use כ, 23:24 uses ק). The difference between the two consonants is in the precise location of the K/Q sounds. כ is a uvular consonant, while ק is velar. When considering the understanding in historical linguistics that the softer sound of כ is a later development, and that the glottal KH is not original for כ, but only the K, it becomes clear that these are similar consonants[14]. Hence any changes in pronunciation (and as a result spelling) would keep the changes limited to similar sounds in the same group.
Later stages
In later cases, we do find writings from the Roman and medieval ages where ב and ו interchange in cases where it is clear the writer already has a V sound for the ב. The probability is very high that the sound depended on regional accents, and there was no official way of pronouncing the letter. “In later stages of Hebrew, 2nd-4th century, we find words which had phonetic “mistakes” that show variant spellings confusing ב/ו, which might indicate a V sound and not a W[15]. However, this probably should be seen as a guess of what happened due to the usual division of the consonants to different parts of the mouth tongue, and throat.
As stated above, we find variations of word spelling where ב and ו are clearly interchangeable due to them sounding the same way. It would seem to me that the fact is that at this point we do find a V sound for ו, and this might be the stage in time where there was a shift in sound. However, it is impossible to say this with absolute certainty due to the lack of evidence for other stages. As far as I know, there are no cases of vocalization in Greek or other transliterations where one can find a ו.
Did Arabic influence Hebrew
A very common argument in regards to the W sound is that it was derived from Arabic which influenced Hebrew. The only clear case where we can find a direct influence on Hebrew by Arabic is in Yemen where also ג and ק ended up sounding like Arabic. However, in many Arabic-speaking countries, we find the local Jewish population uses a V. On the other hand, in countries where there was almost no Arabic, or that there was no Arabic at all we can find W. This is true to areas in the former USSR, Persia (where Farsi is spoken), and in a very small way in Germany under very specific rules.
In a book called Sifte Kohen, published by Ben-Tziyon Hakohen in 1987, one can find a long list of Jewish writers from both Arab and non-Arab countries. In many cases, we can find evidence or open confessions that the sound used was a W. This evidence shows that the Arabic influence argument is incorrect[16].
I will note that I have seen the opposite argument being made that the V sound is actually an influence of European languages on Jews living in Europe. However, it is difficult to make such a statement for the same reasons the Arabic claim needs to be rejected. One historical possibility is that the V sound spread via Spanish Jews who immigrated to different parts of the world, bringing with them the Spanish tradition.
Conclusion
It is my personal opinion that there is no doubt that both V and W existed in post 2nd temple Hebrew, and that V became more popular than W. The question at hand is “was V the original sound”? The case for W has been demonstrated as much more complex than reading traditions, and it is something that is embedded in ancient morphology and phonetics. Even if we do not have a recording of the common speech dating to the bronze and iron age, the stressing of words in Hebrew which created the morphology of today shows that it was indeed a W sound and not a V.
I will not be surprised if some will find other reasons to argue a V sound, and I will address them if they are presented and update this article. However, the overall information we have on Semitic languages and the history of Hebrew morphology indicate that the W sound was the earlier sound of the ו.
[1] Morag Shlomo, Torat Hahegeh V’Hatsurah Shel Ha’ivrit, Jerusalem 2009, pp.27-28
[2]. Ravin Hayim, Safot Shemiyot, Jerusalem 1991, pp.52-53. This is the common transliteration in every comparative linguistic book written.
[3] In modern Semitics studies, other consonants such as ש, צ, ס are given different sounds from what is used today. This is based on the comparative method which finds the related consonants and compares them to find the majority of cases.
[4] Sharvit S., Prakim Betoldot Halashon HaIvrit, Vol.3, Tel-Aviv 2004, p.40
[5] Ahituv Shemuel, Haktav V’Hamikhtav, Jerusalem, 1992, p.4
[6] See Bergsträsser G., Hebräische Grammatik, Hebrew version by M. Ben Asher, Jerusalem 1982, p.74
[7] Morag, 2009, p.21
[8] See for example the case of ילד-ולד. Hebrew uses both, but for most cases Hebrew would use Yeled while Arabic uses Walad.
[9] Blau Y., Torat Hahegeh V’Hatsurah Shel Leshon Hamiqra, Jerusalem 2010, pp.84-90
[10] See Cohen Maimon, HaKtiv V’Haqri Shbamiqra, Jerusalem 2007
[11] See Qimron Elisha, Biblical Aramaic, Jerusalem 2007, p.16-17
[12] In truth, the spelling pre-dates the vowels as used by the Masorates, and might reflect a different way of speaking. It is possible that the exchange was between B and W which belong to the same sound group. This by no means indicates that someone made up the vowels in the Aleppo codex. The vowels as used in the Aleppo should be seen as a latter version of what was used earlier, and that it reflects the preservation of the original meaning of the words. What is in question here is the exact way people spoke, and not the meaning of the words. Even if a vowel point changes its sound, it is still the same vowel point.
[13] There is no doubt that the two spellings are referencing the same word “back”. The context and idiomatic use are very clear in the texts.
[14] Blau, 2010, pp. 53, 77-82&91
[15] Sharvit S., Prakim Betoldot Halashon HaIvrit, Vol.3, Tel-Aviv 2004, p.40. This principle can also be seen in the Kaufman MS of the Mishnah.
[16] Kohen Ben-Tziyon, Siftey Kohen, Jerusalem 1987, pp.44-68, In some cases, one can argue that the writer is trying hard to prove that it is a W, but in most cases, it is very clear that the testimonies do say it was a W.
