I'm devising a complex search strategy to identify the birth and parents of my great-grandmother Mary Ann Harper. She's already discussed here: Where was Mary Ann Harper in the 1881 England census? and here: Locating an elusive UK birth record (Mary Ann Harper born circa 1865 possibly Dudley, England)? and I hope I've taken full account of the answers to those questions in the formulation of my strategy.

As it's based on a number of assumptions, I need to assess what flaws exist in the assumptions and correct them before I implement the search strategy. (I'll ask a separate question about the search strategy, as the two elements combined produced a long unwieldy post, and I want to make sure my assumptions are the best they can be before finalising the search strategy).


I'm asking:

  • Are my assumptions justifiable, given the evidence currently available?
  • How should I amend them, and why?



In the majority the records in which she has been identified (from her marriage certificate in 1883 to her death certificate in 1932, including the birth certificates and marriage register entries for several of her children, the censuses from 1891 to 1911 and electoral registers from 1918-1930) she is identified as Mary Ann. In a small minority of the records she is identified only as Mary.

I am assuming she was known as Mary Ann (by choice) or Mary at all points in her life , but will also consider variants such as Mary Annie, Mary Anne and Maryann and possible mis-spellings/mis-transcriptions of those variants, as well as Polly or Molly as a childhood name.

My search will also include other names beginning with Mary such as Mary Alice or Mary Elizabeth, for the purposes of elimination (as they may also appear in records as simple Mary).

I think variants such as Marie, Maria, Marisa and a whole host of others are unlikely (except as transcription errors), as are Ann or Nancy (based on the fact that she used neither in later life).


In all the records I have seen where this is recorded, it's shown as Harper but I'll consider variant spellings as well, primarily Harpur, Harpour and Harp; and be alert for mis-transcriptions (Hasper and Harker seem to be repeat offences).

More importantly, I'll consider three hypotheses for her surname:

  1. She was born Harper, either legitimate (father named Harper) or illegitimate (mother named Harper, no father named). She remained a Harper until she married.
  2. She was born a Harper but her mother remarried. Mary Ann adopted her step-father's surname (or was erroneously recorded with her step-father's surname in censuses) but used her birth name on her marriage certificate.
  3. She was born with another surname, but her mother subsequently married a Harper. Mary Ann adopted his surname and married under it.

In case 1 and 3, I'd expect to see her recorded as a Harper in the 1881 census (2 years before her marriage) and possibly earlier censuses as well. In case 2, she could be recorded as anything.

Birth year:

Her age at marriage in 1883 was given as 18, suggesting an birth year of 1864/1865, which is consistent with her ages in the 1891 and 1901 census.

In 1911, she is shown as 40 years old, which would give a birth date of 1870/1871, but I think it highly unlikely she was married at 12/13 years old (although it would have been legal at 13) and having her first child 4 months later.

At her death in 1932 she was aged 66, suggesting 1865/66 for her birth.

Her known children were born between 1883 and 1902, which doesn't constrain the likely birth range any further.

I am therefore assuming she was born between 1860 and 1870; and that it's most likely but not certain she was born in the middle of that range (1864 to 1866).

Birth place:

The censuses from 1891 to 1901 (which are the only evidence so far for place of birth) consistently state that she was born in Dudley (which was at various times in Staffordshire or Worcestershire, but Staffordshire around the expected time of her birth). However, I'm not relying on this, as it might simply be the first place she remembered living.

At her marriage she was living in the poorest part of central Birmingham. Her children were born variously in or near Burton-on-Trent (including her first child), or in or around Birmingham.

I am making no assumptions about her birth place, as there isn't enough evidence to support any.


Although I'm making no assumptions about her birth place, I think it likely but not certain that she was resident in England before her marriage in 1883, long enough to meet her future husband and fall pregnant in late 1882.

Parents names:

I have no evidence for her mother's name.

Her father is named as James Harper (deceased) in her marriage certificate, but that could be:

  1. The name of her birth father or step-father.
  2. A fiction concocted to diguise the fact that she was illegitimate.

I am making no assumptions about her parents' names.

  • Is misspelling of surnames in 1890s England so a common that they must be taken in account when searching? I've seen this problem in other countries with illiterate people and minorised languages, but it seems surprising to me that that could happen in England in late 19th century.
    – Pere
    Commented Feb 27, 2017 at 20:45
  • 1
    @Pere, Yes, spelling was still variable, particularly where people were writing down what others said, perhaps in a regional accent that might not be their own. Also, Harper and Harpur and Harpour seem to have been distinct surnames that people mistook for each other.
    – user104
    Commented Feb 28, 2017 at 6:48

1 Answer 1


I can't flaw your assumptions. All I would add:


To surname assumptions I would add that you also have evidence from her children's birth registrations. Even without seeing the certificates we know that she appears as Harper on all eleven of them:

enter image description here

In cases where a woman went by multiple surnames prior to marriage, I have yet to see a case where the mother's maiden surname is consistent for every birth registration. For example, if she sometimes used a stepfather's name, that might appear on a couple of the registrations. The fact that these are all so consistent (and there are so many to compare!) suggests to me that whether or not Harper was her birth surname, it was very likely the surname she had always used up until marriage.

Although likely cost prohibitive at this point, it would be worth checking every birth registration to see exactly how Mary Ann's name is written, and who the informant was in each case.

Birth year

I have a couple of instances in my family tree where a pregnant 16 or 17 year old married and was listed as 18 or 'full age'. I therefore would not be surprised if Mary Ann was born between 1864-1868. Any later seems unlikely; marriages at 13, while legal, were incredibly uncommon. Based on the other evidence you have I also think it is unlikely she was much more than 18 at the time of her marriage.

Birth place

The 1911 census also lists her birth place as Dudley – you did not mention that one.

I agree with your hesitancy to make any assumptions about birth place. You can't, for example, rule out that she was born in Scotland and just never knew it herself. It is nevertheless by far most likely that she was born in the Dudley area.


With the only piece of information being a father's name listed on a marriage certificate, you rightly shouldn't make any assumptions about this. I have lost count of the number of times an incorrect father's name on a marriage certificate has sent me on a wild goose chase. Also do not assume that because it says 'deceased' that her father was actually so. It could have been anything from a clerical error to an intentional untruth.

  • 1
    Excellent points, especially the evidence from the registration of her children, and the possibility that her father was not actually deceased in 1883.
    – user104
    Commented Feb 27, 2017 at 17:11
  • The entries above in the GRO birth index refer to her maiden name, which is consistent as HARPER. But a woman's maiden name is not necessarily her birth name, it is the "name in which she first contracted a marriage". So your scenarios in which she was born in a different name but later used the name HARPER, and then married in that name would still be valid.
    – AntonyM
    Commented Mar 1, 2017 at 15:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.