Missing Marriage certificate in FMP and FreeBMD.

[Stephen49.]

I would appreciate any thoughts. I found a marriage certificate on Ancestry for my cousin Robert Horton b 1864 Bethnal Green. The certificate I have shows him married to Louisa Matilda Cox b 1864 Bethnal Green. The certificate is dated 31 July 1887 St. Mark, Old Street, Middlesex. However, I can't find their marriage either on FMP or FreeBMD. On FMP I can find Robert in Shoreditch but married to another woman. At the same time I can find Louisa also married in Shoreditch but married to a different man. It seems strange to me that only Ancestry has this certificate, I wonder where they got it from.

 

[Helen7]

I can't see what the problem is here. I found this marriage of Robert Horton and Louisa Matilda Cox on both FMP and FreeBMD. Old Street is in the Shoreditch area so that would be the registration district. The only odd thing is that the FMP transcript lists the marriage as 24th July 1887, whereas the certificate on Ancestry has it on 31st July 1887 so why the one week discrepancy I don't know, but the other details match.

 

[Stuart]

On the right-hand side of the image on Ancestry, one of the icons (the one above the spanner) opens an information panel. That contains the source details, telling you this comes from a parish register held in the London Archives*. Ancestry have made an agreement with them to make these images and that gives them exclusive use, at least for a period.

So it isn't a certificate - that is a written transcript of the same information taken from the second register, which was filled in at the same time in the church and sent to the local registrar. For more recent marriages, ordering a copy from the GRO (on paper) gives you a true certificate; for older ones like this you get an image (though a rather inferior one) of their register.

FMP has no access to those images, but has two records. One they made from the GRO indexes published after the end of the quarter, merging the separate entries for the bride and groom. That has no date, just the quarter, as that's all the index tells you. The second record also has the date, and is credited to FamilySearch. This most likely comes from the transcripts made by volunteers over many years and sent in to go into the IGI. In principle it should contain the parish name too, which also is not in the GRO index. However, these FS/IGI records are very variable in completeness and reliability, as you'd expect given how they were collected.

* The London metropolitan Archives dropped Metropolitan from their name only last August, so I'm impressed that Ancestry have altered this source information so soon. Note, however, that the citation at the top of the panel still uses the old form! Perhaps they have a policy of not updating formal citations?

 

[Heather]

This marriage is also listed on Family search with a date 24 July 1887 . Robert and Louisa are to be found on census records for 1891, 1911 and 1921.

 

[David47]

The Family Search output says that the 24 July record is Banns, which fits with the actual marriage date.
But many indexes seem to refer to banns as marriages or mix them up.
View the original records if possible.

 

[Stuart]

Both Ancestry and FMP have records with the 24th July date, identified as sourced from FamilySearch, and both label it as "marriage date". So obviously all the banns got tipped into the "marriages" box early in the processing of the data set. And if you can't get hold of the original or its image, just remember that you may not have the actual marriage date. In almost all cases the marriage is a week or two later, in just a few cases it may never have happened.

 

[peter]

However, I can't find their marriage either on FMP or FreeBMD.

No problem finding it on FreeBMD:



FMP have only transcribed three of the entries correctly. As there are two females and only one male their MarriageFinder algorithm suggests that they both married the same man.

Robert Horton's entry has been confused with the one above:

 

[peter]

Understanding how errors occur is key to overcoming them. In this case the marriage was in a church, and the register was online: next time it might not be!

 

[Stephen49.]

Re: Marriage between Robert Horton and Louisa Matilda Cox. Thank you very much everyone for your kind help and support. Not finding this marriage in FreeBMD the first time was due to my poor research, I apologise for any disturbance within the forum. However, at 75 maybe I'm going a little 'gaga' upstairs. I go to FMP - Civil marriages - I type in Robert Horton - 1887 - England - Cox - No Results. I leave out Cox and put in Shoreditch - No Results. Seeing a marriage has been found, I must be doing something wrong but I don't know what.

 

[Stuart]

One register filled in at the church provides the details for the civil registration process. Until not long ago, all we could use to look those up was the published quarterly indexes. The other register is kept at the church and becomes a parish register, or at least can be thought of as one.

FMP have processed the quarterly indexes to provide their civil registration dataset for England (England & Wales marriages 1837-2008). They match entries to find couples, but as Peter has said in this case they have got it wrong and Robert Horton's name has gone missing from the Shoreditch entries. So putting in his name and Shoreditch and choosing "civil marriage & divorce" does not find anything. Putting Louisa's name does find her, but gives the the wrong bridegroom.

They have another dataset which they call "England Marriages 1538-1973", though they don't tell us anything about what it is! They do say "Index (c) IRI. Used by permission of FamilySearch Intl". Given the dates, this must be where they put parish register entries, but they also have some of the later register books, transcribed as if they are parish registers. That's why it has the exact date, and is chosen as "parish marriages". So if you choose "parish marriages" it will find either Robert or Louisa in that dataset.

 

[peter]

this must be where they put parish register entries

Where who puts them? Not Findmypast, it's a FamilySearch record set.

 

[Stuart]

Where who puts them? Not Findmypast, it's a FamilySearch record set.

Does that matter? I was talking about what you can find where.

I was thinking of a dataset as something bigger than what FMP call a record set. But in their hierarchy England Marriages 1538-1973 is record set, so I've mislabelled it as a dataset. Taking what I said as applying to parish marriages (a subcategory) it's true, if a bit obvious. But the point I was making is that some marriages appear in FMP as both civil and parish marriages.

Note, however, if you look up the England Marriages 1538-1973 record set, there is no attribution of origin or copyright to FamilySearch or another source. So if FMP are only giving that information per item, they could quite well have added some more items from somewhere else for all I know.

 

[peter]

Does that matter?

Yes, it does matter. Understanding how different providers segment their data, and knowing where it comes from is important.

some marriages appear in FMP as both civil and parish marriages

That's a slightly misleading comparison. Like other sites FMP has the GRO quarterly indexes, which they have transcribed, but the parish marriages that we've been discussing are transcribed entries which have been indexed.

That may sound like a minor difference, but consider all the confusion caused for Stephen by FMP matching the wrong bride and groom.

Note, however, if you look up the England Marriages 1538-1973 record set, there is no attribution of origin or copyright to FamilySearch or another source. So if FMP are only giving that information per item, they could quite well have added some more items from somewhere else for all I know.

That's possible, although I think it's unlikely. Both Ancestry and Findmypast usually keep FamilySearch records separate as they are free to access.

 

[CarolB08]

Coming to this discussion late, is it possible that the first marriage entry 24th July 1887 took place in a Register office and the second one in the church? Maybe the bride and groom were of different religions??

 

[peter]

Coming to this discussion late, is it possible that the first marriage entry 24th July 1887 took place in a Register office and the second one in the church? Maybe the bride and groom were of different religions??

There are certainly instances of couples who married twice, but it's unusual and possibly illegal. And if this were to happen post-1837 the marriage would be indexed twice by the GRO, as in the case I reported in the newsletter a couple of years ago.

Fortunately in this case the marriage register is online, so we can see other marriages in the same church on the same day.


All rights reserved. Held by The London Archives. Used by permission of Ancestry

Checking the transcriptions in the FamilySearch dataset for other marriages in the same marriage register shows that it is the transcriber who was at fault, and that they have used one of the banns dates in place of the date of the marriage. For example, the next marriage in the register also took place on 31st July, but has also been indexed as 24th July, whilst the one after that has been indexed as taking place on 10th July.

Although it is tempting to blame an incompetent transcriber, it's quite possible that the error occurred at a higher level in the transcription project.