Aviv Keller is 91 years old. Hardly anything remains of the world of his childhood. In his house in Rosh Pina, which has housed six generations of his family since 1878, he sits and reminisces. Keller is the oldest man in Rosh Pina, the grandson of the founders of one of the first Zionist farming colonies in the country. He was born in 1918, when Ottoman rule ended in Palestine, and his childhood recollections are the last living testimonies of life during the first years of Zionist settlement. At the time, Rosh Pina comprised a small colony of observant Jews, who tended vineyards in the hills of the Galilee.
"My family had a vineyard and our wine was the best because of the topography. Some of us spoke Yiddish at home, others Arabic, but in school we spoke only Hebrew," he recalls.
Keller's language itself is a remnant of a distant and disappearing past. Indeed, he is apparently the last Hebrew speaker whose language still contains elements of a unique Galilee dialect - an odd form of modern Hebrew spoken in the communities in the Upper Galilee at the beginning of the last century, which gradually fell into disuse and all but disappeared.
"I think I am the last of the Mohicans," he says. "The young generation no longer speaks the way we did."
Even Keller himself uses a "milder" version of the original Galilean Hebrew: Unlike his teachers in Rosh Pina, he pronounces his name "Aviv" and not "Abib." Nevertheless, you can hear the difference between his speech and that of other Hebrew speakers of his generation.
Speakers of some languages can often immediately identify different accents or dialects of those languages. However, this does not happen among Hebrew speakers. The Book of Judges tells of how members of the Tribe of Ephraim had its own dialect, by which a rival group, associated with judge Yiftah Hagiladi, could identify them: The group would asked them to pronounce shibolet (stalk), and their interlocutors would say sibolet, thus revealing their tribal affiliation - and precipitating their deaths.
Identification of different linguistic strains doesn't happen with modern Hebrew, a young language with a short history. Indeed, most of the people who speak it lived, or live, in a limited, cramped area, and therefore separate dialects did not develop here. At most, one can identify old-time Jerusalemites because they pronounce the number 200 me-ah-tayim rather than matayim, or use a local term for "lollipop."
And yet, in the not-too-distant past, Hebrew did have another dialect. It was spoken in the Galilee, particularly in the older colonies of its northern part. The most noticeable sign of this dialect, as mentioned above, was the pronunciation of the letter bet: It was consistently pronounced "b" and never "v" - even when it appeared without a diacritic dot.
"Some people poked fun at this form of speech," Keller explains. "They would ridicule us for the way we would say 'zebubim lebanim mistobebim' instead of 'zevuvim levanim mistovevim' ("white flies are moving about"). But, we in the colonies ignored them."
But where was the Galilee dialect actually born? Unlike dialects in other languages, this form of pronunciation did not evolve gradually, but was actually "invented" by one of the first people who revived the ancient Hebrew language, Yitzhak Epstein. Epstein, born in 1863 near Minsk, settled in the Galilee in the 1880s and became the best-known and most influential educator in the region's Jewish communities. Beginning in the 1890s he taught his pupils Hebrew, but introduced a form of it that sounded different than the language spoken in central Palestine.
"First of all, he insisted that we always pronounce the letter bet as 'b,' whether it had the diacritic dot or not," says Keller. "In school they tried to make us say 'b,' and also 'p' [for the letter peh, which is usually pronounced 'f' when it has no diacritic dot]. I don't really know why. I think they wanted to give the people of the Galilee a special color. That's the way people spoke in the colonies and we still almost speak like that."
Another teacher, Simcha Wilkomitz, who followed in Epstein's footsteps, inculcated the unique pronunciation with no less enthusiasm, and it became a source of pride for youngsters in the distant, isolated Galilee.
Why did Epstein and Wilkomitz prefer such a pronunciation? Keller thinks it has to do with the influence of Arabic: "For many years we were the Arabs' neighbors, we spoke and lived with them and perhaps those conversations with them led to this. Perhaps it's not refined speech. It was not seen as something strange or out of the ordinary."
However, imitating Arabs was probably not Epstein's primary motive. Linguist Aharon Bar-Adon of the University of Texas claims he was trying to revive an ancient dialect of Hebrew that had been spoken in the Galilee. Iair Or, translator and language researcher, offers another reason for Dr. Epstein's caprice.
"In Epstein's days people admired what was called phonetic languages, which had a one-to-one correspondence between letters and sounds," says Or. "Esperanto, for example, was built on the basis of this principle and there were many suggestions for reforms in other languages, for example in the spellings of Russian and German. The idea that there would be only one way to pronounce the letter bet, and not two, derives from this principle, which no doubt guided Epstein."
In his own writings, Epstein provided rather odd explanations for his thinking. He maintained, for example, that the sound of the guttural khaf creates "frequent friction," which "could grate the refined layers" and cause physiological defects.
However, Or believes that Epstein's ideas concerning letters with two possible pronunciations was logical: "Linguistically, his solution was very elegant compared with what eventually prevailed, which sets rules for bet (versus vet), gimmel (versus jimmel), kaf (versus khaf,) peh (versus feh) and tav (versus thav). [Now there are] strict rules with a very long list of exceptions. Such complicated systems usually cannot be taught and do not survive, so the Galilean dialect at least had an advantage in that it was very simple."
Nevertheless, the Galilee dialect did not last long: By 1913 the Language Committee (the body that preceded the Academy of the Hebrew Language) was already campaigning against it. Legendary educator David Yellin, one of the committee's founders, argued that a different Galilee pronunciation should not be allowed because it could "ruin the whole idea of uniform pronunciation."
"The Galilee dialect is interesting in terms of being a linguistic experiment by educators at the time Hebrew was being revived," explains Or. "It is not a full-fledged dialect. When one refers to dialect, one usually speaks of differences in vocabulary and syntax - and not just of the matter of pronunciation. But had the phenomenon continued, and had a group of speakers in the Galilee continued to differentiate itself from other Hebrew speakers - it is reasonable to assume that over time it would have developed other characteristics that would justify calling it a dialect."
In subsequent decades, when there was more mobility between the north and the center of the country, the Galilee could no longer maintain its own form of speech. Its dialect gradually disappeared, and now, only a very few elderly people in Rosh Pina and Metula speak it.
For his part, Keller has a simple explanation for the dialect's disappearance. "Today people hardly speak Hebrew," he says. "Every conversation ends with 'okay' and 'bye' - and that's it, peace be upon Israel."