The Power of Extracts

Lauren B. Eisenberg Davis

This article originally appeared in the Kielce-Radom SIG Journal, Volume I, Number 1 (Winter 1997), pages 25-29.

The unfortunate reality of our genealogy quests is that we must work with documents we don't necessarily understand. The Kingdom of Poland's Napoleanic format of vital civil registrations, which records statistic events, is not columnar with headings. They can appear formidable, with extensive text in either Polish or Russian, depending on the year. Don't be intimidated; anyone can master these records. I did. The complexity of this paragraph style format may appear to be a curse, but it is a blessing in disguise: a great deal of information is buried in these records, mysteries that are not apparent from only looking at the indices. The challenge is not so much how to decipher individual records, but to be able to tie records together to form a family.

Format of the Records

Each original record book contains birth, marriage and/or death records for a particular town for a specific span of years. Most years were indexed by the original clerk. The reels of microfilms are typically arranged as follows: for a specific year, birth records are followed by the birth index, marriage records followed by the marriage index, then death records followed by the death index. Occasionally I've noticed that an index is out of place. The years are usually delimited by some type of header frame on microfilms of the original record books made by the Church of Jesus Christ of Latter-day Saints (LDS). Typically a marriage record will fill a page; birth and death records are usually two per page, although I have seen as many as seven.

Over the course of time, I learned that a standard method of genealogy research was insufficient to produce a comprehensive picture. Searching index lists of vital event registrations, and then examining those records, was simply not good enough. Too many important pieces of the puzzle were slipping through the cracks.

Under the tutelage of Warren Blatt, I'd learned the importance of extracting basic facts from every record for an entire town in order to study the pre-surname period (1810 to 1820, and often beyond). I began to realize that this was a necessary technique for a much wider range of years: the full scope of the LDS microfilms of the Jewish vital records for the nineteenth century.

My original research in Krasnystaw, Poland (Lublin gubernia) yielded a handful of ancestors for my tree. Once I studied every record for the whole town, I realized that nearly ten percent of all vital events registered in Krasnystaw were for my family! Many of the connections were difficult to discern. A lengthy analysis process slowly helped me to identify large undiscovered factions of my family.

I expanded my extractions to encompass Radom, Turobin and Izbica as well (I had previously worked on Checiny with Warren Blatt). In doing so, I was able to make some meaningful observations about certain aspects of life in my ancestral towns: juvenile deaths, occupation distributions, naming patterns, cousin marriages, literacy rates. I could see how the trends and patterns differed from town to town.

What Are Extracts?

Vital record extractions are the isolation of crucial facts from vital statistic records. Often, the fields extracted include name, parents, ages, occupations and towns. It is very useful to have this summary of a whole town's records available for genealogical research. A full set of extracts can reveal solutions to mysteries that might otherwise remain buried. Often the only way to get such extractions is to perform them yourself. A cooperative research group such as the Kielce-Radom SIG is a good means of pooling our efforts.

Why Do Extracts?

Why should we perform vital record extractions when we can simply access the index of civil registrations in the LDS microfilms for our towns? The answer is many faceted, and often complex.

On the most basic level, many films are not completely indexed. Some events/years have no index at all. Sometimes marriages are indexed only by groom. The alphabetical order of an index can be misleading. Most indices are semi-alphabetical: alphabetical by the first letter of the surname, then chronological within that letter. However, some indices are not alphabetized at all, but are chronological, and those that are alphabetical may be ordered by given name rather than surname, or a mix of both. A few items may be misplaced within an alphabetical listing, such as the name Goldberg in the midst of the "M" section, or a few 'leftovers' appended to the end of the alphabetized index.

The index is often not an accurate reflection of the simple facts transcribed from the records. The very name of the person whose event is the subject of the registration may be misidentified on the index due to clerical error made over 100 years ago. An index is simply not a primary source document.


The most crucial reason to perform complete vital record extractions is to be able to use data from the pre-surname period. Prior to the Russian surname mandate in the Kingdom of Poland, and on a limited basis for several years beyond 1821, most Jews registered only by patronymic names (identification by father's given name). The Polish notations for patronymics are the following suffixes:

It is impossible to identify your ancestors from a patronymic entry on an index, unless the given names are such an unusual pairing, such as perhaps "Sumer Elkon Tanchumowicz", and even such an unusual combination of names in one town could be quite common in another. There are simply too many "Abram Szmulowicz"s to identify which is yours without further study. In some towns, patronymics are used as part of a full name even after the surname mandate. Hence you may see "Abram Szmulowicz Cukierman". Here "Szmulowicz" is neither a middle name, nor part of a multiple surname. It simply means that Abram Cukierman was the son of Szmul.

A confusing issue in working with pre-surname records is the way in which the patronymics were recorded. Sometimes the patronymic used in a registration was the father's name, and sometimes the grandfather's name. Naturally, this may be inconsistent across multiple registrations in a nuclear family. Consider the following family:

Child: Jankiel
Father: Icyk
Child's paternal grandfather: Berek
Father's paternal grandfather: Szlama

The index entry for this birth might read Ickowicz, Jankiel or Berkowicz, Jankiel. Similarly, the record itself may identify the father as either Icyk Berkowicz or Icyk Szlamowicz. It is impossible to analyze the value of these records at a glance.

Technically, when a girl or woman is identified as "Hana Dawidowiczowna", it should mean Hana, daughter of the son of Dawid, making Dawid Hana's paternal grandfather. In reality, this expression was often used erroneously as a synonym for Hana Dawidowna in vital records. Similarly, a married woman identified as "Sura Moskowiczowa" should be Sura, wife of the son of Moszek; Moszek should be Sura's father-in-law. Again, examples can be found for the incorrect use of this phrase to denote Sura, wife of Moszek. You must consider multiple records of a family to unravel this situation.

Additionally, many records contain names that are difficult to discern between surnames and patronymics (or sometimes even occupations), such as Berkowicz and Abramowicz. In Checiny, for example, "Manelowicz" represents both the surname Manela and a patronymic for the Gnat family, whose patriarch had the given name Manela. Too hasty a decision might cause a researcher to connect these two distinct families without realizing the different connotations of Manelowicz.

Maiden names are not usually reflected on the index. There are a few exceptions to this, but very few. If you are tracing the descendants of a woman whose marriage you cannot find, you will not know her married name, and hence will not be able to find the birth and death records of her descendants, nor even her own death record.

Often, towns do not list women by maiden surname, but by patronymic. This clouds the issue even further. By extracting the records and studying the extractions, it is possible to assign maiden surnames to women by identifying other records in their families: their birth and/or marriage records which would indicate the formal family surname. Once a woman's birth family is identified, her ancestors (and lateral relatives as well) may be traced.

Conjectural Surnames

Analyzing record extracts can result in assignments of conjectural surnames. This means that it is possible to determine what surname a family later acquired, although they were not identified by a surname at the time of the earlier records. By comparing families for similar characteristics, such as names, patronymics, ages and occupations, it is often possible to identify later records for the same family: records for events registered after the family adoption of a formal surname.

For example, an 1820 birth record for a Liba Laia, daughter of Herszek Fiszlowicz, cap maker, and Marya Perecow[na] might be paired with an 1822 death record for a Liba Laia Zaiac, age 2, daughter of Herszek Zaiac, cap maker, and Marya Rozenwald. Hence, a researcher could conjecture that Herszek Fiszlowicz took on the surname Zaiac, and that Marya Perecowna's father and/or brothers acquired the surname Rozenwald. This is not a 100% guarantee. It is always possible that the death record is from another family of similar characteristics. Unusual names and occupations make conjectures more probable. Ages, if available, should be taken into account, but realize that ages were often inexact as reported at the time of registration, and that spelling of all names varied from record to record, or even within a single document. It is conventional for conjectural surnames to appear in square brackets in published extracts. These names do not appear in the original records.

Surname Mandates

A very confusing issue, when working with 19th century Jewish records from this region, is that even once surnames were adopted, they were not fixed or permanent in the way we think of them today. See the article "Alternate Surnames in Russian Poland" in the Summer 1996 issue of Avotaynu: The International Review of Jewish Genealogy (XII:2). Perhaps your family name changed over time. This was fairly common early in the process of surname adoption. This is a major consideration for the Kielce and Radom gubernias, which were once included in West Galicia. The Russian surname mandate was imposed on the Jews of the Kingdom of Poland in 1821, but these Jews had experienced a prior surname edict in 1805, under the auspices of the Austrian controlled West Galicia. From 1810, when the Russians gained control, through part of 1821, surnames were not required at all in this region. Consequently, at the time of the 1821 mandate, many families chose different surnames than they'd had earlier, or different branches of a family would make diverse decisions regarding retention of the 1805 surname versus adopting a new one. Siblings may have acquired different surnames. You can identify the multiple surnames used by a single family from studying a large set of records. Extraction makes this possible.

Town of origin is an important fact in extractions. Towns are usually listed on marriage registrations and occasionally on birth and death records as well. A marriage record also identifies towns where the bride and groom were born or resided, as well as the towns where the banns were read. Banns are documents of intent to marry. Be careful in your interpretation of banns, if they are available (banns were often not microfilmed for many towns), in the absence of other supporting evidence of the marriage. On rare occasion, the marriage never took place.

Typically there were two banns per marriage. When the bride and groom were from different towns, banns were written in each town. Hence, if the marriage took place in another town, there will be no corresponding marriage registration in one of the towns where the banns were registered. Banns usually contain the same basic name, age, parent and residence information as the marriage record. Consequently, it offers another opportunity to verify such facts if the information is illegible on the marriage registration itself. Additionally, if one party was from a small village and the marriage record doesn't indicate where that village is, the place where the banns were written is a good indication of the location of that village.

The towns embedded in the registrations are a crucial clue if your family came from a different town than the one you are studying. An unusual and poignant example is the high percentage of deaths at the Radom hospital in the 1848 cholera epidemic, where many of the victims were not only from other towns, but also from other powiat (districts) and even other gubernias (provinces).

What makes the toil of extracting worthwhile? You may argue that you can peruse all the records for your town without the additional work of extracting records that bear no resemblance to your family. However, in order to be thoroughly analyzed, the data must be studied, not perused, and studied repeatedly. While this may seem like an inordinate amount of work to reap a few extra benefits, consider the key pieces of information that you may uncover, or the sheer volume of relatives that may surface.


Following are a few examples of situations in which full town extractions can solve puzzling mysteries:

Primary vs. Secondary Source Data

When dealing with extraction documents, it is crucial to bear in mind the nature of the material we are dealing with. There are differing styles of record extraction. It is possible to create an extraction document simply from names and parents. Some researchers extract as much information as possible from each record, including house number (for early records which include such a field), date of the event, time of day, witnesses, occupations, and who signed the document.

Some of the data is subject to error, either in the clerk's original transcription, or interpretation of the handwriting from the poor quality of the microfilms and original record books. Many of the ambiguities may be resolved as the researcher becomes familiar with the residents of the town. As always, it is prudent for the researcher, when using secondary source data such as extractions, to examine the primary source data (original records) for final verification. It is always best for the genealogist to view the actual records pertaining to his/her family to verify the interpretation and glean additional facts.

Full extraction was an important jump for me. My research had taken on new dimensions, new proportions of importance. I not only knew who my ancestors were, but had gained a clearer picture of their way of life. Town extracts provide a window through which to look beyond the simple confines of one's own family, to life in the shtetl as a whole, the very environment in which our ancestors lived. Only when considering as many clues as possible can we realistically compile a cohesive and comprehensive family tree.

Copyright ©1997, 1998 by the Kielce-Radom SIG and Lauren B. Eisenberg Davis.

Kielce-Radom SIG Home Page