Monday, July 31, 2017
When do you rely on an index? Illusion or Reality?
Accumulations of information eventually reach the saturation point and finding any one item becomes nearly impossible. Linear searches, before computers, were extremely time-consuming. Let's think about this in terms of genealogical research. Inevitably, as we do more in depth research into historical records, we depend more and more on the organization and classification of those records.
Historically, organizing voluminous records required maintain manually created indexes or lists of the records in some order. The utility of these organizational methods depends on the depth of classification imposed on the information. For example, if we are searching for a specific name, the utility of the index, for us, depends on whether or not the index included the individual being searched.
For example, let's suppose we are research county court records. The particular organization used by the county may be chronological or by case number or by the names of the parties. The county may or may not have created an index of the records listing the names of the parties. Even if an index exists, we may be searching for someone who was not a "party" to the litigation and therefore not included in the index. However, there may still be extensive information about our target individual in the file. If we are forced to do a manual search, the only method that will discover the information we are seeking is to do a word by word search through the entire corpus of information.
When I am searching unindexed deeds, for another example, I have to read each of the deeds and look for the names of the individuals who may have signed the deeds as parties or witnesses. The parties (grantor/grantee) may be indexed but the witnesses are not. In this case, a grantor/grantee index does little to help me with my search. I may have to search hundreds or even thousands of deeds to find one name on one deed.
As computers became more and more common, we entered an age where we depend more heavily on indexes. As long as the information stored by the computers is in "text" format, computer programs can search massive amounts of data in seconds and look at every word. This ability gives researchers an illusion of complete searches with or without indexes. However, we need to remember that much of what we are searching as genealogists is locked up in its original format, i.e. handwritten records. Until we develop a reliable and extensive ability of handwriting recognition, we are still heavily dependent on manual indexes.
The good news is that there are a lot of manual indexes. The bad news is that the indexers still select only certain "fields" to include in their indexes. For example, the deeds I wrote about above. I have yet to see a manual index of the names on deeds that included all of the people named in the deeds. If we want to find the information we need, we still have to do our own manual search. Of course, it may be easier to search digital images than either original records or microfilmed copies, but the existences of an index often gives the impression that we are searching the records when what we are really doing is searching the index. This illusion is pervasive. In many instances, the large online database programs provide only an index of records and do not provide access to the original record images. We are essentially locked out of finding the information we need under the guise of doing an adequate search because we rely on the information selected for inclusion in the index.
In every case where we rely solely on an index for our information, we must make every effort possible to examine the original record set in detail before concluding that the information is not available in those records.