Responsa Project wins Israel Prize

Creator recounts academic opposition

By MATTHEW WAGNER
Imagine taking an obscure corpus of religious legal decisions that spans nearly two millennia, four continents and a half a million documents that is written in an ancient Semitic language, and making every single detail accessible within seconds via computer search. And imagine doing this in the 1960s while fighting an uphill battle against an academic community that was skeptical at best, and often downright antagonistic. That is precisely what Prof. Aviezri Fraenkel of the Weizmann Institute did. The result of his labors, the Responsa Project [Proyeht Hashut] won this year's Israel Award for the Works of Judaism category. Although most of the world has never even heard of it, Responsa literature is for many Jews the heart of Jewish practice. It compromises the creative output of Jewry's most outstanding rabbinical minds throughout centuries of exile. It comments on every conceivable aspect of a Jew's life including, but not restricted to, theology, marriage, torts, jurisprudence, war ethics, burial rites and the afterlife. Fraenkel, an Orthodox Jew and computer scientist, got involved with the computerization of rabbinic literature more than three decades ago. It all began with an innocent memorandum that he sent to the Hebrew University of Jerusalem. In the memorandum, Fraenkel criticized the manual-index method that Hebrew University's researchers were using to make Responsa literature more accessible. He offered a method of his own that became the model for what has become known as the Responsa Project. Fraenkel argued that it was impossible to capture the essence of a Responsum by using a few key words and descriptions as conventional information retrieval (IR) methodology posited at the time. Instead, he said, the entire document should be copied into computer storage. Needed excerpts from the text could then be retrieved by a computerized search system. But leading experts in the IR field rejected Fraenkel's criticism. In 1962, MIT's Prof. Yehoshua Bar-Hillel wrote: "Any scheme of directly comparing a request formulation with a straightforward one-to-one encoding of the original document must be regarded as wholly utopian and unsubstantiated." In the mid-Sixties, the Academy of Science in Jerusalem rejected Fraenkel's proposal. "One senior member even attacked me for daring to submit such a proposal while Hebrew University is doing their own project," Fraenkel told The Jerusalem Post. Prof. Ya'acov Choueka, a specialist in computational linguistics who believed in Fraenkel and joined him in the Responsa Project in the mid-Sixties, recounted the animosity and opposition he and Fraenkel encountered in the academic community. "He [a leading expert in IR] got so furious and angry, claiming that the idea is dumb and absurd and that it can never and will never be successful," wrote Choueka in a 2001 edition of the Journal of Combinatorics, which was dedicated to Fraenkel's 70th birthday. "I had no intention of getting involved in this project," Fraenkel said. "But when I encountered so much opposition to my idea, I realized that I had no other choice but to do it on my own." Today, Fraenkel's method for IR is used by Google, AltaVista, Lycos and other search engines, while the manual-index method preferred by Hebrew University is a relic of the past. "Very often I fell into despair," Fraenkel said. "I never had a doubt that I was right. No one ever shook my faith. But I was afraid that I would not have the means to continue." Ironically, a project that was so quintessentially Jewish and geared to a Jewish audience ended up being funded at the outset by gentiles. The National Bureau of Standards, which provided Fraenkel with a research grant for work he was doing on numbers theory, allowed him to use some the money for the Responsa Project. Also, Temple University Prof. Aaron Shreiber convinced America's National Endowment for the Humanities to provide matching funds. But Bar-Ilan University, which later adopted Fraenkel's project, and the Hebrew University gave him a cold shoulder in the formative years. In 1991, Bar-Ilan released the first CD version of the Responsa Project. This Pessah the 15th version will be released. Bar-Ilan has also launched an Internet version that is accessible on a subscription basis. The Responsa Project incorporates all the major Jewish legal texts and tens of thousands of responsa. But Fraenkel is concerned about its future. "Rabbi Ya'acov Weinberger, the present head of the project, is doing wonders with a minimal budget," he said. "But we have no research and development budget whatsoever. As a result, we are falling behind in areas such as user interface, and more improved retrieval methods such as direct retrieval from a compressed text. "The Jewish people should have the best tools available to access the texts that make up their religion and culture," Fraenkel said.