Quaker Marriage Certificates: Using Witness Lists in Genealogical Research

From: "Stewart Baldwin"
Subject: Re: [Q-R] signatures on marriage certificate
Date: Wed, 14 May 2014 11:24:48 -0500

Source: http://archiver.rootsweb.ancestry.com/th/read/QUAKER-ROOTS/2014-05/1400084688 Tom Kirk wrote:

> Is it correct that members of immediate families sign on the lower right hand side of the document? Can one assume that these then represent family members?

I wrote an article on this phenomenon in 1997, which included a detailed statistical analysis of two early Pennsylvania Monthly Meetings (Falls and Chester) [Stewart Baldwin, "Quaker Marriage Certificates: Using Witness Lists in Genealogical Research," The American Genealogist 72 (1997): 225-43]. As Dan has already pointed out, the word "assume" is too strong. Nevertheless, early Quaker marriage certificates show a strong tendency to list relatives in the far right column. How strong a tendency? Well, that depends on a number of factors, as practices varied with respect to time and geography.

The most important thing to realize is that only a small percentage of original Quaker marriage certificates have survived from the earlier period. Those were kept by the family, and what we almost always see in the records are the information from the certificates as they were copied into the various Monthly Meeting records. The degree to which the arrangement of the witness lists (or the lists themselves) were faithfully copied varies widely. In my statistical study of the early Falls MM marriage records, I found evidence that the arrangement was less faithfully copied if the copier was forced to start a new page. In some of the North Carolina meetings, it was common practice to copy exactly twelve witness names into the MM records, regardless of how many actually witnessed the marriage. There is the same problem in published records, which, in order to save space, do not always accurately publish the arrangement of the lists, even if they do list all of the witnesses.

I have never seen a copy the original of an early English Quaker marriage certificate, so my experience there is solely with the various MM records themselves, mainly from Lancashire, Yorkshire, Cheshire, and Worcestershire, and occasionally other English counties. I have seen enough cases where the same marriage was copied differently into more than one record to say that the copies did not always list all witnesses and did not always keep the arrangement of the signatures when they did list them all. Many of these early marriage records list very few witnesses, which probably means that only some of the witnesses were listed by the copiers. In most cases, the lack of accurate genealogical information makes it difficult to judge whether or not there was a separate "relatives list" in the certificates, but there are a handful of marriage records from the Marsden MM in Lancashire in which certain individuals are explicitly labeled as "relations" (in a couple of cases even separated into parents, siblings, and others). However, even in that Monthly Meeting, most of the witness lists lack such a designation.

One feature that is very common in Quaker marriages, as they were copied into MM records in the 1600's in both England and America, is to have separate lists for the men and women, with a very strong tendency for relatives to sign at the beginning of each list. For early English Quaker records, my general impression is that longer lists tend to follow this pattern, while shorter lists probably did not list all of the witnesses. However, even in the earliest American Quaker records, many marriages show the pattern of relatives signing in the right-hand column, and some certificates from the Chester MM even have that column explicitly labeled as "relations." By around 1700, the vast majority of marriage cerificates recorded in MM records in Pennsylvania, Delaware, and New Jersey seem to be using the "standard" relatives-in-the-right-column format, with the arrangement of witnesses usually copied relatively faithfully into the MM records, but with additional caution required whenever the witness list spills over to the next page. There are some variations. Some used the right two columns when many relatives were signing. The overwhelming majority of ORIGINAL marriage certificates that I have seen (not that many) follow this pattern through the early 1800's, less frequently in the late 1800's. Some of these original certificates can be found in the records just put online at Ancestry.com. This seems to be true even in the small number of original certificates that I have examined from Virginia, North Carolina, Ohio, and Indiana. However, the copies of witness lists that I have seen in the MM records for these states do not conform to this pattern in general, most likely because less effort was made in those states to record the names of witnesses as they appeared on the certificates, unless the small sample of original certificates that I have seen is not representative. Thus, in general, the witness lists of early Quaker marriage records from Pennsylvania and neighboring states can be very useful in giving clues about relationships, while the records from North Carolina, Ohio, and Indiana are much less useful in that regard. I should emphasize that, although I have quite a bit of experience with original MM records from New Jersey, Delaware, and North Carolina through the early 1800's, my experience with Pennsylvania MM records is much more extensive than it is for the other states, and I do not have much experience with original Quaker records from the mid-1800's and later, although I expect that to change now that so many original records from other states are now available online.

Some basic tips for using witness lists:

1. Except for the rare cases in which the right-hand column is explicitly labeled as "relations" or something similar, the presence of a name in the right-hand column cannot be regarded as proof of a relatiosnhip.

2. If most of the signers of the right-hand column can be proven as relatives by other evidence, that increases the chances that the other in that column were also relations.

3. If an individual appears in the right-hand column in several different marriages of the same family, then the liklihood of a relationship is strong, but the exact relationship still needs to be proved by other evidence.

4. Remeber that the word "relative" is itself an ambiguous term. We are all related if you go far enough back, and since I have a large amount of Quaker ancestry, many of the readers of this list are related to me within ten or fewer generations, but we wouldn't generally refer to ourselves as "relatives" on a day-to-day basis. Even when there was a relatives list on the marriage certificate, there was no hard and fast rule about how closely related someone had to be in order to sign that list. In most cases, parents, siblings, step-parents, step-siblings, uncles, and aunts would sign a relatives list. For first cousins, it varies widely. One thing I have noticed is that if either the bride or groom was widowed, then relatives of the deceased first spouse would sometimes sign the "relatives list," so the term relative was sometimes applied rather loosely.

5. Most importantly, don't jump to conclusions. Electronic databases on Quaker families are littered with falsehoods that arose because somebody used a witness on a Quaker marriage as "evidence" but did not follow it up carefully enough. Forming experimental hypotheses from witness lists is often a good way to proceed on a difficult problem, but it is important to follow up such a hypothesis with careful research to see if it can be confirmed, and you need to be prepared to abandon the hypothesis (or at least set it aside until more evidence is available) if the proof isn't there. It is easy to get attached to pet theories. I have made a large number of tentative theories which I later had to give up on (sometimes reluctantly) because they turned out to be false or the evidence just wasn't there. [I only report the ones that turn out to be right, or at least highly probable. :-) ]

Stewart Baldwin