The "official" statistics that most people quote, based on the "National Jewish Population Survey -- 1990" put the American Jewish population at 5.5 million. A new study, "American Jewish Identity Survey (AJIS), 2001" conducted (as was the first) by respected demographers and social scientists and published last February, says that this number has declined to less than 5.3 million. The "National Jewish Population Survey (NJPS) -- 2000-01" says 5.2 million. The big question is: 5.2 million WHAT?

The number of 5.2 million is supposed to count what the authors of the studies call the "Core Jewish Population" of the United States. This group includes, the authors of the Identity study (Egon Mayer, Barry A. Kosmin and Ariela Keysar, of the Institute for Jewish Studies of the Graduate Center of the City University of New York) write, "those whom most Jewish communal bodies accept without qualification as potential members of their communities."

A fuller analysis of who exactly is included in this group leaves one wondering whether this is merely a description of the current situation or a prescription for whom the authors think should be included. However, it is clear that religious Jewish communal bodies do not regard much of this group as their constituency, as will soon be shown. (A report of the study is available electronically at: WWW.GC.CUNY.EDU/STUDIES/STUDIES_INDEX.HTM)

This core Jewish population is divided by the authors into three major groups: BJR (Born into the Jewish Religion), that is people born into the Jewish religion and still there, JBC (Jews by Choice), meaning those who have converted or "otherwise" become committed to being Jewish, JNR (Jews with No Religion), those who said they had an ethnic Jewish background but were not at all religious. A major group not included in the core population is JOR (Jew of Other Religions) that is those born Jewish but now professing another religion.

It is very important to realize that these groups are self-defined. Thus, survey respondents are asked questions and their answers are simply written down and counted, without any attempt to challenge or verify them. If a respondent says that both his or her parents were Jewish, for example, he or she is marked as a person born of two Jewish parents. There is not even a follow-up question asking if either of them converted or not.

This is standard for academic surveys, and gives a result that is valid for their purposes. Most social groups are simply made up that way. If you ask someone what his native tongue is and he answers "Russian," you do not ask him to say "comrade" in Russian to prove it. If someone says that she uses Detergent X, you do not ask to see her ring-around-the- collar. You just take people's word for it.

For most concepts and groups, this works fine. They are well-defined and there is no uncertainty about who is in and who is out. With regard to other social questions, often it is the attitude itself that is of interest and being measured ("What is your opinion on the death penalty?"), or else, they are groups in which you are what you say you are (more or less) unless you are lying. If you say you are a Democrat, you are a Democrat. If you say you are a Methodist, you are a Methodist.

The situation with Jews is much more complex.

Nonetheless, in times past this approach gave a reasonable approximation of those who were actually Jewish. If someone said they were Jewish, they probably were. But in these assimilationist times, and after three to four previous assimilationist generations, that is no longer so.

The meaning of "Jew" is not agreed upon. Some people "feel Jewish" but may not be. It depends on what you mean by "Jewish." Is it someone born of a Jewish mother? Or also someone born of a Jewish father? Or also someone who identifies with the Jewish people?

In the previous study in 1990, no questions were asked about parentage. Last year in the new study, those interviewed were asked about their parents and their answers were recorded. The results are interesting indeed.

Continues . . .