METHODOLOGY:
THEORETICAL & EMPIRICAL METHODS

INTRODUCTION

A very useful tool for persons researching an ancestor when using their small set of known given names, would be a data base presenting all of the given names related to those already known and which could also have been used by the ancestor -- and which the researcher might be able to find in archival documents.  Best of all would be that these names include those the ancestor could have used in Europe as well as those that he might logically have adopted upon immigration to a foreign country.

For the researchers of all Jewish given names used by Jews in Europe during the period 1795-1925, the primary problem in developing families or sets of linked Jewish names has been the choice of "Root" or "Primary" names which unite all related Jewish names into single sets.  The Root or Primary name is the single index which defines the set or "family" of names.  The second main problem is how to link other "unclued" names to their proper Root or Primary name -- that is, how to place single Jewish names found in archival documents (names not having clear links to existing family sets) in their appropriate families/sets as defined by the Root or Primary name.  A particularly perplexing factor is that many Yiddish names or nicknames were used as independent, stand-alone names, but were also used in association with one or more different Root or Primary names.

In any scientific approach to solve these problems, there are two potential methodologies which can be used:  a Theoretical, or an Empirical approach.  In either case, the goal of the research must be pre-defined before beginning the research, and must be re-defined as needed during the course of the research.

The Theoretical approach has been used for several centuries.  It involves the creation of a theoretical system of analysis based on hypothetical rules for how groups of linked names might have descended over time from an ancestral Root name.  Then, single names found listed in archival sources of names are fitted into this framework.  This approach usually involves onomastic/phonetic/migrational studies.  Root names are selected by historical inference or by using a scheme used by a published source considered to be authoritative.  In a good methodology, the hypothetical rules must then be carefully tested with fresh experimental data to ascertain if they are valid.  The goal of this approach is to be able to explain the historical development of each set of linked names from an ancient root name.

On the other hand, the Empirical approach has been used for about 800 years.  It seeks and uses data sources which already contain small linked sets of the given names actually used by individual Jews;  by inductive reasoning, a number of these small name sets are grouped with one another to form one larger collection of linked names, and each such large group is assigned a Primary name which characterizes the group.  When all of the large groups have been formed, they are re-tested against new field data for validity.  When the data have been validated, the total set of name groups constitutes a data base which can be used as is.  Thus, the resulting data base of linked name groups is itself the goal of the research project, and not a verified theory of how names changed in sequence over time, or what theory can be deduced for why the the names in the various sets are linked together.

The Empirical methodology was adopted by European Ashkenazi rabbis involved in defining the Jewish laws for writing given names in Gitin (Jewish divorce documents).  Using contemporary and recent field data from actual divorce cases and the investigative results of divorce rabbis, they determined by statistical study and inductive reasoning what groups of linked names individual Jews actually chose for themselves.  Furthermore, they analyzed these data in the light of Jewish law to determine rules for defining the legal names to be written in Gitin.  The resulting Primary or legal name for use in a Get was usually the Hebrew or Yiddish name given at birth (since from a Jewish standpoint, this was the most important name that he/she would ever have) plus any additional kinuim (certain other Yiddish and/or secular names) needed for proper identification.  These data sources used by the rabbis may be considered to be of first quality from the point of view of name linkage within a family of names.

Besides names of people involved in divorces, other obvious empirical sources of linked name sets (not used by the rabbis but important today for this GNDB project) are gravestone readings and Chevra Kadisha lists for Jews -- these frequently list the Hebrew, Yiddish, and secular names of the person, but usually have not undergone as intensive an investigation as for the writing of a Get.  Names from such sources must therefor be considered to be of secondary quality.

The Given Names Data Bases methodology presented here is based and expands upon the rabbinic Empirical approach.  It begins with the tabular results expressed in Hilchot Gitin books, and then utilitizes extensions of the rabbinic rules to incorporate single names found in archival sources not available to the rabbis.

All books containing studies or lists of Jewish given names can be classified into three categories:

• Theoretical, onomastic/phonetic/migrational name studies
• Empirical name studies leading to Jewish divorce-law ("Hilchot Gitin") books
• Lists of names for various other purposes

Most Theoretical studies were usually conducted within the framework of scholarly research by specialists in onomastics, phonetics, migrations, or related fields.  Hilchot Gitin books are formal Jewish law books which guide divorce rabbis in writing the legal names of the parties to a Get in the divorce document.  And Lists of Jewish given names were prepared by persons with some specific audience and usage in mind:  Guideline books for use by government officials, possible names for newborn infants, guideline-names for rabbis in their congregational work, guideline-names for synagogue officials in calling men to the Torah in an aliya or for recording their official Hebrew names, etc.  Special name lists for various purposes will not be considered further in this study;  but see  Addendum:  Jewish-Name Lists in the US. 

THE SCIENTIFIC METHOD AND GIVEN NAMES RESEARCH

The progress of science is marked not only by an accumulation of facts but by the emergence of the Scientific Method and of the Scientific Attitude.

The revival of science during the late medieval period was stimulated by the recovery of ancient works, especially those of Aristotle.  However, progress was slow because investigators relied on ancient authorities and on deductions from the Scriptures as interpreted by the Church instead of on independent observations;  a notable exception was Roger Bacon.  Leaders in basing scientific concepts on observed data and on experiment include Copernicus, Kepler, and Galileo.  The value of inductive reasoning and of experiment in scientific work was formulated in the early 17^th century by Francis Bacon (1561-1626), who can be credited with generating the first concepts of the Scientific Method.  In the 17^th and 18^th centuries much scientific progress was made by men unconnected with universities;  scientific societies were the chief agencies in furthering and correlating research.  Universities began to play a major role in the 19^th century, and the 20^th century saw the rise of research institutes and of industrial research laboratories.

The Scientific Method was slowly incorporated into many of the natural sciences after the time of Francis Bacon.  This was true also for rabbinic works written on Hilchot Gitin.  While the early efforts to apply this method were imperfect, a steady improvement could be seen.  "Beit Shmuel", the seminal work of legal Jewish names printed in the late 17^th century, is a primitive example of the method, but later Hilchot Gitin books steadily improved in their adherence to the principles of the Scientific Method.  Finally, there came the publication of the quintessential early-20^th century book, "Get Mesudar", which was a fine example of the use of the Scientific Method in Hilchot Gitin research.  The scientific method has now been extended beyond the natural or "hard" sciences (e.g., physics) to embrace today some "soft" sciences (e.g., political science), including given names research.

The usual steps of the Scientific Method are:

1. Data are gathered by observation of phenomena.
2. Using inductive reasoning, hypotheses or preliminary generalizations are drawn from the data.
3. The validity of the deductions that follow logically from the generalizations is tested by further observations & experiments.
4. A verified generalization is then considered a scientific theory or law.
5. If contradictory facts arise, the theory must be modified or replaced by a new concept.

The second step (inductive reasoning) always involves the definition of assumptions and constraints on the scope of the hypotheses and generalizations.  These assumptions and constraints are always tested in the third (validity testing) step of the scientific method.

This process is much more difficult in the "soft" sciences because there are so many variables and factors which are not known or are not recognized, or which are hard to control and which must be defined statistically.  Accordingly, in cases where the the goals are set too high, or unsatisfactorily-stated or unstated assumptions are made, or the data constraints are not known or accounted for, the results are less than satisfactory.  If so, then the theory or "law" derived is controversial and never truly attains the status of an accepted law.  This appears to be the case for the pure Theoretical approach to given names research -- no universally-acceptable law of name development over time has yet been developed, and controvery has continued for hundreds of years.

However, in the Empirical approach, the goal has been set lower (to create a table of linked given names derived from the regional/then-current data collection), with no attempt made to show how current given names were developed over time from prior names.  This method provides an empirical framework the internal structure of which expresses the validity of the data.  No scientific theory or law of generalization is created -- the table or data base itself contains the unexpressed law.  There was continual agreement and cooperation among the rabbis to resolve regional differences or changes over time, and the results were anchored in Jewish law.

Theoretical onomastic/phonetic/migrational name studies may be considered to be "Top-Down", because they create and use hypothetical sets of linking-rules to define an overall theoretical framework into which individual given names are then inserted.  On the other hand, the empirical studies by the rabbis which resulted in the Hilchot Gitin books, were based on their prior examination of the actual linked sets of names used by men and women involved in divorce proceedings, their development of a consistent set of name groupings, and their determination of how the names were to be written in the Get in order to identify properly the parties to the divorce;  this "Bottom-Up" approach closely matches the objectives of our project.

The use of Hilchot Gitin books to formulate the Given Names Data Base structure was deemed the best approach, since our goal was to use empirical analysis to develop the families of linked names which were used by individuals.  However, in the later phases of this project, use was also made of other sources containing Jewish names, such as onomastic/phonetic studies books, lists of names, and archival sources, augmenting the basic framework data bases which were set up using Hilchot Gitin books.

DATA QUALITY

The Scientific Method properly places heavy emphasis on experimental data, and the quality of the data greatly impacts the success of the research which uses it.  Poor data are readily detected when working in the "hard" sciences, but are often not considered or not recognized by researchers in the "soft" sciences -- and in the latter case this causes difficult problems concerning the validity of the proposed "law".  For two main reasons, this is particularly true for the research of Jewish given names in Europe:  Most of the data available on Jewish given names is of poor quality, and in many cases researchers did not use proper care in applying the Scientific Method to the data available to them.

Most given names data in Europe are of poor quality for easily-understood reasons.  For example, names to be engraved on cemetery stones are always presented during a time of great stress and sorrow for the family members who must provide the information, and the names given to the engraver were frequently not representative, not current, in error, or otherwise improper.  Names appearing in civil documents were frequently presented under conditions of great stress on the parts of interviewer and interviewee (particularly in Eastern Europe) due to a certain underlying mistrust or hostility.  In both of these typical cases, unrepresentative or incorrect name sets were given.  Even in the case of documents originating in the Jewish community itself, the names recorded were frequently shortened versions of their complete name, or nicknames, or sometimes even names that had been abandoned by the bearer.

Name lists resulting from these types of sources must necessarily be considered to be of secondary rather than primary quality.

The only instance in which the total set of names borne by a Jewish individual was investigated with great care was in the case of a divorce. For here, the specter before the divorce rabbi was that he might misrepresent the legal name of one or both parties to the divorce, resulting later in a finding that the Get was invalid.  For Jews, this is a major trajedy in the case where the parties remarry and bear children.  The children of such a union could be declared to be bastards, with all of the stigmas and real disadvantages of such a status.  Because of G-d's commandment to "go forth and multiply", Jewish men tended much more than non-Jews to divorce their "infertile" wives (the divorce rate among Jews was twice as high as that for the secular population), greatly exacerbating an already difficult situation.

Accordingly, divorce rabbis went to great lengths to ascertain what was the correct, legal name of a man and woman who were partners to a divorce. They were obliged to do this by the laws embedded in the Hilchot Gitin books which they used as their guideline.  So they pursued quite detailed investigations to help them to decide what would be the true legal names of the pair.  Some of the binding factors which the authoritative rabbis required of them in the Hilchot Gitin books were:

• Frequency of use of his various names by others
• Which names were used for aliya to Tora, for signing documents, and by friends
• How names were used in different places
• When names were changed during illness

The result of this in-depth investigation was a name (or group of names) declared to be the legal name for use in the Get, plus other names which could not be used in the Get (names of endearment, nicknames, abandoned names, etc.).  By its very nature, this careful research led to a consistent, high-quality set of linked names -- ideal for use by the rabbis who wrote the results of their research in their Hilchot Gitin books.  And unfortunately, mostly unused by Theoretical names researchers to structure their linked-name groupings or to expand their lexicon of given names.

Accordingly, the linked name sets which appear in Hilchot Gitin books are of high quality, and this source is considered to be primary, particularly for structuring given names data bases.

THEORETICAL APPROACH

Up to modern times, onomastic studies which are presented in books have mostly begun by first choosing Root names and developing hypothetical rules of name linkage, then extracting given names from older books, and finally organizing the names in groups according to the rules.  These rules were generally based on a new process of logical analysis, Jewish migration patterns, and the corresponding appearance of new Jewish names in European immigration countries.  Usually, the analysis depended on how people would be expected to change name pronunciations and create new names from old, and it used logical, phonetic considerations, usually without using field data for rigorous proof since finding such linked groups was difficult.

One common working approach has been to begin with an ancient root name, to develop rules for how possible descendant names might logically result using phonetic and other considerations, and to use the resulting theoretical descendants list as a framework for grouping names extracted from prior books or other sources.  Root names were selected from both Jewish and non-Jewish sources.  Jewish Root names included Hebrew (Biblical) and Yiddish names, while non-Jewish names were secular European names.  Each Root name was always the earliest name which could be found for the family of names to be defined.

As expected, different researchers came to different conclusions about name inclusions and linkages within families of names, since each used his own sets of Root names and rules.  This approach may be called the "Theoretical" approach, since it starts with a theoretical analysis of what rules might be written to describe reasonable transitions from one given name to another, beginning with the Root names.  It embeds Jewish given names studies inside the more general secular onomastic studies. Its outstanding weakness is that in practice, a modification of the Scientific Method was used, i.e., no follow-up verification of the hypothetical rules created in the research was ever carried out using new field data.

JEWISH-LEGAL EMPIRICAL APPROACH

The Given Names Data Bases (GNDBs) project takes the new, different "Empirical" approach described above, which is based on the authoritative, legal books of Jewish divorce law written over many centuries by European rabbis -- legal guidelines for community rabbis who prepared divorce contracts.  In their own research and consultations with their colleagues and community rabbis, these experts co-ordinated new developments with those of prior rabbis and they created over time (beginning in the eleventh century) a single set of rules for associating names with one another and in spelling the names using Hebrew characters. For example, the rules took into account that many Yiddish names were pan-European and were spelled in a standard way, yet were pronounced differently in regions where the Yiddish dialects were not the same;  but they also recognized that many Yiddish and secular names were used in some regions but not in others.

The rabbis' rules were not based on the historical changes of names beginning with a single "ancestral root" name, and on a set of logical, phonetic rules of change producing "descendant" names;  rather, they reflected how ordinary people in the target population actually had chosen their names, and what were the linked-name combinations they ended up choosing, based on actual field data available to the rabbis.  In this empirical approach, a rabbinic team effort over an extended period of many centuries resulted in one agreed-upon set of rules which fit the field data obtained by examining each complete set of names used by many individuals undergoing divorces.  These rules could be applied to different European regions in different time periods in accord with natural differences.  The local exceptions not fitting the rules and the problems of names missing from the Hilchot Gitin books were to be resolved by the local community rabbi, within Hilchot Gitin book guidelines.

The end result was the definition of Primary names and their linked Subsidiary names (to be written in the Get), and other related names (like diminutives, terms of endearment, and childhood names, which are not to be written in the Get).

Thus, our GNDB project began by extracting from the Hilchot Gitin books applicable to each target region, these legal and vernacular (Yiddish, local, and pan-European secular) Jewish names defined by the rabbis, then entering them in a basic, framework data base for each region.  These data bases contained the legal single and multiple names, the Yiddish and diminutive names and the names of endearment defined by the rabbis, as well as the corresponding local and European secular names also defined by them.  In these framework data bases were then embedded additional given names which we extracted from regional European archival documents (e.g. Chevra Kadisha and revision lists) and other sources, based on numerous towns throughout the region to achieve uniform coverage.  To these name groupings we added the statistically-linked foreign vernacular names adopted by Jews in their new countries of immigration, extracting the data from foreign-country data bases for immigrants.  This is called the "Jewish-Legal Empirical" approach since it takes a purely Jewish legal-document approach to given names, relying only on Jewish rabbinic sources to define the method and framework.

The archival sources may be categorized as follows (in decreasing order of quality):

1. Hebrew & Kinuim & Old and New Loazi names (from Hilchot Gitin books)
2. Hebrew & Kinuim & Loazi (Yiddish plus secular) names (from other Jewish sources, e.g. Chevra Kadisha lists)
3. Hebrew & Kinuim names (from other sources)
4. Hebrew &/or Kinuim &/or linked Loazi names (from other sources)
5. Hebrew or Kinuim names (from other sources)
6. Loazi names of known linkage to Legal names (from other sources)
7. Loazi names of unknown linkage to Legal names (from other sources)

STRENGTHS AND WEAKNESSES

The Theoretical and Empirical approaches have inherent strengths and weaknesses which reflect their methodologies.  The Theoretical method has been applied by researchers who usually lived many years after the period of their research interest, and frequently not in the European sub-region being researched.  Their information sources were of two kinds:  Primary data (e.g., Hilchot Gitin books), and Secondary data -- gravestone readings, and names recorded in books by prior researchers.  The Empirical method was used by rabbis who lived during the time period of their research and in the European sub-region itself.  Their information sources were of two kinds:  Primary data (direct information from divorce rabbis in their sub-region preparing the names for a Get, and by rabbis currently researching their own Hilchot Gitin books for other regions), and Secondary data -- names recorded in Hilchot Gitin books by previous rabbis inside and outside of their region.

Unfortunately, the reading of Hilchot Gitin books is not easy since they are Jewish law books written in the traditional dense style of such books. Most researchers using the Theoretical method either did not consult most such books at all, or if they did, they took the easier path of simply extracting names from the summary section of the book.  They therefor missed much essential detail because they did not attempt to understand the essential sidebar explanations provided by the author of the works. Another problem was that advantage was not taken of the automatic internal structuring of name groups based on experimental data.  The result was that these high-quality souces were either not used or were improperly used, and the Theoretical research suffered accordingly.

Accordingly, the work of Theoretical researchers had the disadvantages of distance in both time and place from the subject of their research, and the partial or complete absence of high-quality Primary data sources from their experimental data set.  This is reflected in the necessary use of assumptions based on the theoretically-based (but experimentally unsubstantiated) hypotheses which they had to make during their research. On the other hand, the work of Empirical researchers had the advantages of closeness in both time and place to the subject of their research, and the use of high-quality data sources.  This, together with the approach itself and the more modest goal of the research, minimized the need for assumptions.  These two facts had an important impact on the strengths and weaknesses of these two major methodologies.

The strengths of the Theoretical approach are:

• Jewish given names study is embedded in the broader set of onomastic names study, providing a more general picture.
• Schemes of name development are proposed for the extended period from the eleventh to the twentieth centuries.
• Lists of names are presented for major cities in various European regions in an orderly manner, sometimes along with citation of sources (book, date, pages).
• An extensive body of onomastic literature has been developed which is useful for further research by onomasticians, as well as for use by Jewish genealogists.

For Jewish genealogists, the weaknesses of the Theoretical approach are:

• The hypothetical rules of name linkage remain largely unproven and the resulting linkages are questionable.
• Natural disagreements exist between various onomastic researchers, leading to confusion as to how Jewish genealogists should interpret the listed data.
• The name presentations in books are not in a convenient format for the research of their European ancestors from the last few centuries, since the books require manual extraction of sought-after names embedded in long lists of different types of information.
• The listings do not make clear the natural linkages between various names in a given region.
• Classification of names as Hebrew, Yiddish, or secular are usually not made or are unclear.
• There is a tendency to de-emphasize the more intimate names (e.g., names of endearment) used inside Jewish homes.
• Relatively few large cities and very few small towns in European regions are represented.
• No foreign-name information is presented for Jews who emigrated to foreign countries.

The strengths of the Jewish-Legal Empirical approach for Jewish genealogists are:

• The structure of the linked name sets is derived from actual field data, reflecting actual usage.
• The methodology places emphasis on the Jewish religious and cultural aspects of Jewish given names, including intimate names used within the family.
• The name groupings are based on Jewish divorce law and therefor have legal status.
• The fifteen regional data bases are easy to use, displaying search results on a computer monitor in easy to understand groupings, without other distracting information.
• Search results show clearly the name classifications within each family of names.
• Linked foreign vernacular names used by European Jews who immigrated to foreign countries are shown with the European names.
• A uniform coverage of large and small towns in each European region is presented.

For Jewish genealogists, the weaknesses of the Jewish-Legal Empirical approach are:

• The data organization relates only loosely to the historical development of Jewish names over time in Europe and is not explicit.
• The applicable time period is only 1795-1925, and earlier period-spans will need to be developed separately.
• No information is presented for specific cities or towns, only a region-wide average.

SUMMARY

The two methodologies are both Scientific.  The differences between the Theoretical and the Empirical approaches resemble the different approaches to studies in other Jewish areas such as religion, Jewish history, Jewish culture, Jewish art, and so on, and it is to be expected that these two different approaches would arise.  These differences enrich Judaism by broadening our understanding of things Jewish.

In conclusion, the main uses for given names collections based on the Theoretical approach are by onomasticians doing research in this field and by Jewish genealogists who wish to find out how given names may have evolved over time in their region of interest.  The main use for the Given Names Data Bases using the Empirical approach is by Jewish genealogists researching their European ancestors in the period 1795-1925; onomasticians will have lesser interest in the GNDBs.  Thus, the first approach is mainly oriented toward onomasticians, while the second is oriented towards Jewish genealogy researchers.