'Ayin is an epiglottal or pharyngeal consonant, part of a group of which only two survived into modern Hebrew (the other one is 'chet, which is pronounced deeper in the throat than the flatter ch sound of Germanic languages). Many more survive in Arabic, and were still pronounced by Jews from Middle Eastern extraction well into the modern era. The fact that they have atrophied from modern speech helps to explain a baffling and infuriating phenomenon for the student of Hebrew: the fact that it has so many seemingly redundant consonants, while other consonants do double duty as two different sounds, depending on the word and context:

  • The aforementioned 'ayin is interchangeable in speech with aleph, the first letter in the Hebrew alphabet and the equivalent of the Latin a. ‘Chet is usually pronounced the same way as chaf, which is sometimes a flat non-epiglottal ch and sometime a hard k. The other k sound in the alphabet, the kuf, used to be an epiglottal k, still extant in Arabic and usually transliterated as q, as in Qadi or Iraq.

  • Tsadik is now pronounced by Hebrew speakers like the zz sound in "pizza", but was originally an epiglottal sound and still survives today in Arabic words like Basra. It's the only one of these epiglottals that doesn't have a confusing doppelganger in the alphabet, and for that we thank it. As far as I am ware, the epiglottal g in words like Gaza and Bagdad didn't really have a Hebrew equivalent, although I could be wrong about that one.

  • Another group of sounds that are confusing to the newcomer are the t phonemes. The Hebrew letter tet was originally a voiced, semi-epiglottal th, a little deeper in the throat than the th in the English word "the". Taf was usually pronounced like the unvoiced th sound in "theatre", and a full stop version of the t phoneme didn't really exist in Hebrew; however that is the way both are pronounced these days.

  • There is also vav, which used to be a glide like w, but is no more - which helps to explain why bet is sometimes pronounced as a b and sometimes as an unstressed v. And lastly, gimel, which today is the equivalent of the first sound in "guile", used to have an unstressed version as in "gel" or "jury".

  • I think that's it; there is also shin, which as a stressed and unstressed version for s and sh sounds, but that is historically consistent since Biblical language times as far as we know.

The changes in the pronunciation of these letters is contingent on their extremely long history, and are no more sinister than the Great Vowel Shift in English, the elided consonants in French, or why so many European languages have k, ck, c & q, all of which sound the same. While it is true that the Sephardic, or Middle Eastern, Jewish communities retained more of the epiglottal sounds because of their close association with Arabic-speaking populations, other sounds (such as the voiced th) survived better in European communities.

Furthermore, there were other differences in accent and pronunciation between the two groups; while Sephardic populations had a mixed stress system such as English has, the Eastern European ones spoke almost entirely in words that had a stress on the final syllable, like French has. They also had a different vowel array which included diphthongs such as ey and ay, which are entirely absent from Arabic and from modern spoken Hebrew.

The point is, when these two large groups of Yiddish, Ladino, Moroccan, Russian and whatever else speakers all tried to speak the same language, they found that there were bits of it they simply couldn't pronounce correctly. And so modern Hebrew was gradually flattened to a middle ground that was relatively easy for everyone to get their tongue around; the epiglottals had to go, as did the diphthongs and the funny stresses of Yiddish. Broadly speaking, modern Hebrew is a mix of Ashkenazi consonants and Sephardic vowels.

Now this is a touchy subject in Israel, but the fact is that most if not all of the early immigration to Palestine was from Europe, especially Russia. Only in the fifties did significant numbers of people from Sephardic populations start immigrating to the new state if Israel, and they came to a country which was smack in the middle of the Middle East (boy, is it!), but had decidedly European values, culture, self-image etc., which dominated the elite. This, to put it very diplomatically, caused some friction. And yes, up until relatively recently pronouncing your 'ayins and 'chets was as likely as not to mark you out as working class and poorly educated, as well as summoning other broad stereotypes, such as being right-wing and anti-socialist.

But, and this is an important but, times move on. The fifties are a long time ago, Ehud Barak stupidly apologised to the Moroccan diaspora at large during his time as Prime Minister, and, most importantly of all, Israel has seen several large influxes of immigrants to look down upon. These days, not being able to pronounce your eiches properly - a phonemic signature of native Russian speakers - is just as evocative of a low social status, especially for women (single mother, prostitute, easy woman, peroxide blonde...). And then you have the Ethiopian Jews, whose skin colour broadcasts their "inferiority", to people who want to think in those kinds of terms, long before they ever open their mouths to speak.

In short, while it is true that an accent redolent of epiglottal consonants used to be seen as a lower class characteristic in Israeli society, this has little to do with the resurrection of the language and more to do with the contingency of immigration to the new state of Israel; and in any case, these stigmas are gradually sinking out of significance, and the "correct" pronunciation of these consonants is actually enjoying a renaissance in many quarters.