When I find a woman's married name and have not yet found her maiden name what is the best way to record her name? If I put nothing, I end up with many Jane or Mary names in my list and have no easy way to tell them apart. Should I put their married name in parenthesis to distinguish them or is there a better way?


The following list has been extracted from the responses provided below. You should read the complete answer to understand the full intent of each.

  • Add a year of birth to the name

  • Add the name and dates of the spouse

  • UNKNOWN (mar. name)

  • Put married name in parenthesis

  • Leave the surname blank

  • Married name in quotes or parenthesis with their first name

  • Use married name but add Mrs.

  • Use three underscores or ???

  • Use classical approaches of --?—or five underscores [I voted this the correct answer, but for now will personally use married name in parenthesis)

  • Use what works for you (some of the above solutions are in conflict with other solutions - but this one kind of covers them all)

  • Because you refer to the problem as being able to distinguish between people in your names "list," it seems likely your question is software specific. Personal experience indicates that genealogical software features vary greatly as to lists and sorting options. Assuming your question is software specific, then your question would be more valuable if you include the software and version number.
    – GeneJ
    Oct 13, 2012 at 15:32
  • P.S. Know that there are publishing guidelines for entering missing names. At least some of these publishing guidelines are well enough recognized to be considered themselves as standardized practices.
    – GeneJ
    Oct 13, 2012 at 15:35
  • 1
    For a well written, fun look at the practice of entering "LNU" (last name unknown), see Myra Vanderpool Gromley CG, RootsWeb Review 27 Aug 2002, for "In search of the Wild LNUs." Her article begins, "A long time ago in a kingdom far away there lived a maiden ..." ftp.rootsweb.ancestry.com/pub/review/20030827.txt
    – GeneJ
    Oct 13, 2012 at 18:40
  • The summary of responses added to this question is very comprehensive. My comment (NOT criticism) is that clearly some suggestions conflict with others.
    – Fortiter
    Nov 3, 2012 at 12:48

15 Answers 15


On top of other name complexities, it seems limiting to alter a name to make its entry in a list recognizable.* Given software has different limints, I hope some adding some classical approaches if of value.

Based on working with scholarly journals a few years back, there are some recognizable approaches to missing names.

  • Following NGSQ Style,[1] bracketed em-dashes surrounding a question mark or [--?--]
  • Following Register Style,[2] five underscores or _____

I know that if you leave the entry blank, some software will insert the appropriate symbolism for either; some software allows you to select or customize the symbolism.


[1] NGSQ Style or Quarterly Style refers to the style developed and used by editors of the National Genealogical Society Quarterly.

  • See current issues of the Quarterly for the most up to date information.
  • Board for Certification of Genealogists, The BCG Genealogical Standards Manual (Orem, Utah: Ancestry Publishing, 2000).
  • Joan F. Curran, Madilyn C. Crane, and John H. Wray, authors, edited by Elizabeth Shown Mills, Numbering Your Genealogy: Basic Systems, Complex Families, and International Kin, rev.ed. (Washington: NGS,2008).

[2] Register Style refers to the style developed and used by editors of The New England Historical and Genealogical Register.

  • See current issues of the Register for the most up to date information.
  • Board for Certification of Genealogists, The BCG Genealogical Standards Manual (Orem, Utah: Ancestry Publishing, 2000).
  • Michael Leclerc and Henry Hoff, editors, Genealogical Writing in the 21st Century: A Guide to Register Style and More, 2nd ed. (Boston, Mass.: New England Historic Genealogical Society, 2006).

*Consider the likely event that one might actually want to use the genealogical information for some higher purpose that the database--create a family group sheet or basic narrative. I guess I see a circumstance where solving the problem with a list might finding and re-writing all those names.


A lot will depend on the facilities of the software you are using. I use Family Historian, which allows me to customise the columns I see in a list, so I have added in the name and dates of the spouse to enable me to distinguish between the many Janes and Marys without recording misleading surnames.

An alternative approach might be to record a surname as \Married to JONES\, for example, which makes it clear that you're using a placeholder.


I use the method that was suggested in the original question – “put their married name in parenthesis”. This associates the name to the spouse and the names in parenthesis are all listed together so it still identifies it as a “loose end”. It seems a quick and easy way to do it, and works great for my purposes.

  • Are there some downsides to your method?
    – GeneJ
    Oct 17, 2012 at 18:46
  • IMHO the huge disadvantage in putting a women's married name in parentheses to distinguish it from her maiden name: that usage is the opposite of the newspaper convention of listing a woman with all her names with the maiden name in parens, e.g. Mary (MAIDEN) MURPHY SMITH JONES. With Family Historian you put the surname in between // so theoretically one could enter "Mary / / Murphy Smith Jones" if the maiden name was unknown. But think how much easier it would be to record data, especially from newspaper obituaries, if the standard and software agreed with the newspaper convention.
    – Jan Murphy
    Dec 3, 2013 at 22:38
  • @JanMurphy - I think genealogy software should optionally do this for you. I don't know why most don't.
    – lkessler
    May 17, 2015 at 9:19
  • @lkessler My comment was unclear because I ran out of room; I was referring to the way Family Historian allows the user to do data entry, not the way it displays data in the computer-generated narrative. For a screenshot see the screenshot next to Data Entry with the Property Box on the Tour page. If Charlotte's maiden name was unknown, the user could enter Charlotte // to leave the maiden name blank.
    – Jan Murphy
    May 17, 2015 at 19:04

This does not relate to women only.

In many places in the world, boys are named by their father/GF, and not always with a middle name: I, II, IV, Sr. Jr. etc.

Same for If I don't know the last name.

What I ended up doing was to add the year of birth to the name. It's not the best solution, but for men who do not have a maiden name it suits me.

So some trees for me look like:

  • John, 1890, Doe
  • John, 1912, Doe
  • John, 1915, Doe (can be a nephew or a 2nd child)

In the past, I have myself used __ (3 underscores). Since that follows "Z" in the character order, most programs tend to put them at the end of the surname index. Using ??? would put them at the beginning.

But I have since realized that the proper way is to leave the surname blank.

If your genealogical software will put a placeholder in for you for display purposes only (i.e. one of the many possible representations mentioned among the answers here, with "Unknown" being a software favorite), then that will help you identify those people in your tree. If the software developer was really thoughtful, there might even be a year range shown after the name, e.g. [1863-1942] to help distinguish this Mary from all your other Marys with unknown maiden names.

But it is always better to just leave the surname blank. The reason is if you ever export your data to GEDCOM for data exchange or for upload into some online shared tree, the artificial surname you've input will get treated as a real surname by almost all the reading programs.

As a result, too many online databases are full of people with surnames like: Unknown, Unk, Lnu, Ditto, --?-- and others. Tamura Jones has written articles about how these names have spread. See Tamura's articles: The Lnu Family Mystery and FNU LNU MNU UNK. Tamura says: "You simply should not enter made-up names into your genealogy. If you used these abbreviations already, remove them from your database." Only include real names: Unk is a Real Name

If you do enter made-up names as last names, you'll end up contributing to the junk in online trees that are populated with nonsense people such as this one on Geni:

enter image description here


@ColeValleyGirl said:

A lot will depend on the facilities of the software you are using.

Very true. I use Gramps, which gives a list of "todo" based on missing information. Because of this, I'll tend to leave the surname blank in this case. For most (all?) output that I deal with, it's clear how "Mary" or "Jane" is related.

If your software provides a decent search, @Andrew's suggestion of a convention of UNKNOWN (mar. name) is helpful, because you can pull up a list of all of the UNKNOWN when you are looking for your loose ends.

Also, I think most software that I've used has provided a way to record multiple names: "birth name", "married name", etc. A 19th century woman who was widowed and then remarried would have had three surnames during her life; it should be possible to record all of these names.


I use a variation of the husbands surname, by using

Jane /(Smith)/ 

The brackets show the surname is the married name, which makes it quick and easy to see them and groups them all at the top of my record list to remind me I need to sort them out. Like Cole Valley Girl, I use Family Historian and have the husband's surname in a column as well.


I put their married name in quotes or parenthesis with their first name and leave the surname blank: Mary "Smith" (blank) Lot easier to figure out needed research in a list of names. Also, I see a lot UNK (Unknown) for surname.


The GEDCOM Standard 5.5.1 makes it quite clear that an unknown name (given name or surname) should be left blank:

NAME_PERSONAL:= {Size=1:120}  
  <NAME_TEXT> |  
  /<NAME_TEXT>/ |  

The surname of an individual, if known, is enclosed between two slash (/) characters. The order of the name parts should be the order that the person would, by custom of their culture, have used when giving it to a recorder. Early versions of Personal Ancestral File ® and other products did not use the trailing slash when the surname was the last element of the name. If part of name is illegible, that part is indicated by an ellipsis (...). Capitalize the name of a person or place in the conventional manner— capitalize the first letter of each part and lowercase the other letters, unless conventional usage is otherwise. For example: McMurray.

William Lee (given name only or surname not known)
/Parry/ (surname only)

William Lee /Parry/
William Lee /Mac Parry/ (both parts (Mac and Parry) are surname parts
William /Lee/ Parry (surname imbedded in the name string)
William Lee /Pa.../

  • 2
    Yes, you are quite correct. +1
    – lkessler
    May 17, 2015 at 21:39

I've tended to use the variation of the convention you've suggested, of UNKNOWN (Married Name) as this allows at least a chance of identifying the right one.

Of course, if your package allows tables, as observed by @ColeValleyGirl then that helps too.

  • As long as you use a convention that you understand, anything will work. If you're publishing your work (on the web) you may need to be more explicit.
    – user104
    Oct 10, 2012 at 15:34

In Family Historian, which is the software I use, the user can customise the fields available in the Records View screen, which is the screen that looks like a spreadsheet. For a screenshot, see What's New in Family Historian 6 and look at the screenshot next to Quick Filtering in Lists.

I have added a Custom fact called "Who" where I can add a line of descriptive text that will allow me to quickly identify the person. It could be, like Rusty's example, a phrase like "Mrs. Charles R Jones", or "famous poet" or "newspaper proprietor" or anything else that allows me to distinguish them from someone who has the same name and lived in a similar time period.

I avoid putting anything in the surname field if I do not know the actual surname. for reasons which have been explained at great length by Tamura Jones on his site Modern Software Experience:

  • 1
    Jan: Your answer brought this question to the top of the list and got me to look at what I answered a few years ago. I realized my original answer was wrong and changed it. Interestingly, it was without seeing your answer, and I also included Tamura's articles as they were published since my original answer and were what convinced me as well that the surname should be left blank.
    – lkessler
    May 17, 2015 at 9:00
  • @lkessler I was trained to record what is in the record and no more -- comments like 'maiden name unknown' are recorded in research notes, not in the midst of one's data. So Tamura's articles convinced me I should be leaving the surname field blank, no matter what the software might encourage me to do. On Ancestry's online trees, for instance, I add a preferred name with a blank surname, no matter how annoying it is to have a long string of first-name-only women at the top of the all names list.
    – Jan Murphy
    May 17, 2015 at 18:56

I personally leave the married name until I locate the surname. [Mrs. Mary J Smith] [Mary J JOHNSON] It's important to put the Mrs. on a ladies married name and it shows the researcher that they haven't located the surname. Further, in the event of a family member having a similar name it will be less likely to confuse the two subjects.


I use 5 underscores. What's the difference if you have 20 Mary__'s or 20 Mary Smith's. It depends on the size of your database. If you want to add a married name, I would put Mary (Mrs. Charles R. Jones)__.


A couple of popular ones I see are SNU (surname unknown) and Mdnmunk or MdNmUnk (maiden name unknown). However, the points by the posters above about polluting the database with such placeholders are well-made; in fact most geneology sites including as ancestry dot com, MyHeritage dot com and familysearch dot org now think Mdnmunk is a legitimate surname. FamilySearch for example returns over 75,000 results. The pollution has spread to the point that at least one biographer has published a book in which he claims to have identified "after countless hours" the maiden name of an ancestor; one Christina Mdnmunk. He states that he "found" her maiden name via the family tree maker at geneology dot com (see Nine Generations: The Family History of Thomas Dell Lesnett III 1752-2008 by Scott Lesnett; p. 7).

Bottom line: blank is best. Anything else leads to database pollution and chaos.

  • 1
    Welcome to G&FH.SE! I encourage you to take the tour to learn more about how Stack Exchange works -- rather than having threads like discussion forums, SE uses a Q-and-A format. Your comment about "the posters above" to refer to people who answered earlier than you makes no sense in this context, since answers can be resorted by date, number of votes, and activity. If you want to refer to the people who answered before you chronologically, it is best to say that instead. You may edit your answer by using the edit link underneath your answer.
    – Jan Murphy
    Jul 20, 2015 at 22:03

FWIIW - If I don't know the maiden name, I put the last name as "() Married-name" The () reminds me that I have something missing. And they sort together. If I know the maiden name and married name, but not the spouse (e.g. from an obit), I will temporarily put "Maiden Married" as the last name.

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.