Why is it important to understand the distinction between original and derivative sources?

+74 votes

As we continue to merge duplicate profiles and clean up the results, discrepancies in dates and relations are bound to arise-- especially on profiles of people older than 200 years. This is due in large part because of the "diversity" of sources available. Profiles of early colonists are particularly vulnerable to misinterpretation of these various sources. And, as the field of genealogy/family history research has matured-- especially in the last half of the 20th century-- results of more recent and more thorough research have been published. Therefore, when looking at "evidence" for a particular piece of information-- a birth date, the maiden name of a wife, etc.-- it's crucial to understand the source of that information and learn how to judge its reliability.

The first question I ask is: Is this source ORIGINAL or DERIVATIVE? An example of an original source is the 1650 church register, in which a pastor recorded the births, marriages and deaths of parishioners as they happened.  A derivative is an extracted version of that, where someone, at a (usually much) later date, transcribed the original, and perhaps re-ordered it by surname. When transcriptions are made and edited, room for error increases. In New England, there are many such books upon which we rely that published vital records from colonial times through 1850. (And aren't we also lucky that most of these have been digitized and made available online!)  Other derivative sources frequently used as evidence include:

  • The IGI at www.familysearch.org. Quality varies-- some records are extractions from original church records; others are entries without specific source information
  • WorldConnect (at Rootsweb) / Ancestry.com Family Trees - these are user-contributed family trees of various quality. It appears that the bulk of these entries do not include any source information at all. I have heard more than a couple of people say: "but so many people claim she was born in Salem, so it must be true." Ancestry.com, in particular, makes it really easy to download/copy an entire family tree. So in this case, quantity of a fact does not make it more reliable.
  • The Millennium File. This looks like an official source; it was compiled by Ancestry.com from a variety of sources, including user-contributed family trees, so please use with caution.

While it might be obvious that an original source is more valuable than derivatives, they both have their value-- we just need to review all the information we have in light of which type of source we used. 

Another aspect to pay attention to is WHEN the source was created. Family genealogies published in 1890s (many were) typically do not include their own sources, so it's difficult to judge how reliable they are. Family genealogies published after, say, 1930, tend to follow better documentation processes. 

THE definitive work about use of sources is Elizabeth Shown Mills' Evidence Explained. She also runs a Facebook page where she posts daily tips: https://www.facebook.com/evidenceexplained

Thomas Jones just published a new book that I haven't read yet, but might also be of interest: Mastering Genealogical Proof. It comes out in late May.

in Policy and Style by Jillaine Smith G2G6 Pilot (771k points)
edited by Robin Lee
Phew, that's a change (for the worse) in FindAGrave's policy!
My experience, too, Darlene.  I also contact photographers to request permission to use their photos on Wikitree.  Almost all grant permission, and I also keep their emails authorizing this use archived.
Fred, the usage agreement may have changed since Ancestry purchased FindAGrave. While photographers may own their photos, the license might revert to Ancestry once it's uploaded.
Photographers who uploaded photos earlier would have a good reason to object to a retroactive change in Findagrave's license for content.

Let 'em sue me.  To hell with Ancestry's money grubbing, I have the creator's permission.  Ancestry doesn't like it? - See you in court.
When Ancestry bought Rootsweb, folks who had pages posted there were told that everything hosted on Rootsweb would belong to Ancestry from that point forward. A cousin who had posted more than 10,000 pages of his original research and analysis removed it all - lost overnight (and none of the pages can be recovered via the WayBack Machine either).

But my question now is... How do I protest that Ancestry is absconding with photographs that I took/uploaded to WikiTree and which since have showed up in FindAGrave without my permission (and without crediting me either)? I've also seen thumbnails of my photos appear in Google search returns as being in Ancestry.
It's amusing and frightening that we are all still discussing something that we started in 2013!  Anyway, my question is this:  how can ancestry (via find a grave) simply abscond with original copyrighted photos?  The person who took or owns the photos has the copyright. When I added photos to profiles on FAG, I did not sign anything relinquishing my ownership of those photos that I know of and there was certainly no retroactive notice of "now we're stealing your photos whether you like it or not" when ancestry purchased. If this is true, I will never again put a photo on FAG for any reason. I've gone out and taken some in our area when people asked, but I will not propagate the greedy habits of ancestry and FAG. I have to keep a membership because it's a much easier place to begin searches; however their greed is beginning to marginalize any good that one gets from their sites. That's sad, especially when it is all run by a religious organization.
If I'm not mistaken, it's people with individual family trees who post findagrave photos to their ancestry.com trees.  I don't think that ancestry.com is profiting in any way; they seem to be bowing to the inevitable(?) and allowing people to do what they were doing anyway.  Not that it's right, but it does seem to be a different can of worms.

p.s.  I have always asked for permission before posting findagrave gravestone photos to wikitree profiles of my ancestors, and I almost always got it.  :-)
Why would they make the statement that someone here so generously attached a couple of years ago about only using it on ancestry sites?  The statement itself implies ownership when they tell you where you can and cannot use your photo. As I stated here in 2016, I always get and archive permission from the owner of the photo before I post it on Wiki. I'm not sure what you mean, John. Bowing to the inevitable doesn't keep one from being concerned over their actual choice of wording and legality. Do you think they own those photos? I know that you are very thoughtful about these things and have probably come to some personal decisions.
I have no problem taking pictures for FAG and anyone is allowed to copy those pictures to your heart's content. I am a firm believer that genealogy research is improved by sharing rather than hoarding. Barring some kind of fraud or identity theft, anything posted to a public website should be considered public. Privacy is the antithesis of posting to the web. The best way to maintain "ownership" of virtual property is to create your own webpage and only provide access to content based on people agreeing to your terms of service agreement. I am going to continue to assume that any data not otherwise proscribed is open to the public...and you are welcome to consider my data in the same way.

1 Answer

+34 votes
Best answer

I'm finding that a lot of people are using the following as their primary source for the information they add here-- especially those of us working in the colonial New England space. I want to comment on a few of them:


  1. The Millennium File From the source information about this file:
    "The Millennium File is a database created by the Institute of Family Research to track the records of its clients and the results of its professional research... The Millennium File is a compiled source and is similar in form to other linked databases, such as Ancestry World Tree."

    This means it's very derivative-- several steps removed from the original. Such sources are great for clues for searching, but I strongly encourage anyone not to use it as a definitive source for information-- i.e., I would not trust anything in it without confirmation from at least one other (and closer-to-original) source.
  2. Edmund West, comp.. Family Data Collection' [database on-line]. Provo, UT, USA: Ancestry.com Operations Inc, 2001. As bad or worse than the Millennium File. Errors abound. Created while gathering genealogical data for use in the study of human genetics and disease. Compiling data for genetic research does not require the same type of documentation as traditional genealogical research ... Millions of individual records were created from birth, marriage and death records; obituaries; probate records; books of remembrance; family histories; genealogies; family group sheets; pedigree charts; and other sources. 
  3. U.S. and International Marriage Records, 1560-1900. This one is almost worse than the Millennium File. Ancestry's description of it says: "This unique collection of records was extracted from a variety of sources including family group sheets and electronic databases. Originally, the information was derived from an array of materials including pedigree charts, family history articles, querie[s]." I.e., it is a derivative compiled from other derivatives. Great for research clues, but not great for relying on as a source.
  4. Passenger and Immigration Lists Index, 1500s-1900s
    This is a decent source; as an index, its derivative, but through it you should be able to find and go to the actual passenger list --  Ancestry.com has both the index and individual passenger lists -- and see the original image of the list itself. I find it preferable to use the images of the original because in most cases they group people travelling together. In the case of one of my maternal ancestors, I was able to identify (and later confirm through other sources) siblings and in-laws who were travelling together but with different surnames. 
  5. U.S., New England Marriages Prior to 1700, by Clarence A. Torry. Another derivative, although with a bit more authority than the Millennium File. However, subsequent researchers have found errors in it, so it's always good to find an additional, closer-to-original source to support what you find here.

As I review all these, I am struck by the fact that the bulk of Ancestry.com's recommended sources for individuals are derivative-- and in many cases, HIGHLY derivative. 

It also occurs to me that because these collections are on Ancestry.com, that some people may believe Ancestry has vetted or checked their validity. I have seen absolutely no evidence that Ancestry.com engages in such practice. So just because you read it on Ancestry.com does not make it so.

I strongly recommend membership in the New England Historic Genealogical Society (www.americanancestors.org). Their digital collection includes images of far more original records, or closer-to-original derivatives, as well as journals of reviewed and well-cited research. While initially focused on New England, their resources have expanded to other parts of the U.S. and with great resources related to finding links back to (mostly) England. And no, the NEHGS does not send me a free toaster every time I refer a new member. 

NOTE: Their advanced search engine is imperfect. I get poorer results (and often NO results) when I add much more than first and last name. I especially avoid use of the "Keywords" field which appears to be wholly useless.

(That last comment should really assure you that they won't send me a free toaster.)




by Jillaine Smith G2G6 Pilot (771k points)
edited by Jillaine Smith

yes Very nice, Jillaine. I second the recommendation to join NEHGS.

With regard to the NEHGS online database (not ancestry.com, which does not include references) for U.S., New England Marriages Prior to 1700, by Clarence A. Torry: for each marriage, Torrey's entry is annotated with the references from which he (apparently) drew his conclusion. That list can provide excellent hints for further research, not only for the marriage, but for the individuals and associated families. N.B., most of those references are themselves derivative sources. Thus, Torrey is, in large part, a list of derivative sources from which he drew (sometimes erroneous) conclusions.

From: Torrey’s New England Marriages Prior to 1700. (Online database. AmericanAncestors.org. New England Historic Genealogical Society, 2008.), entry 33912:

[VEAZY] William (-1681, ae 65) & Eleanor/?Helen [THOMPSON] (1627-1711), m/2 John FRENCH 1683; by 1644(5?); Braintree {Reg. 9:152, 15:113, 55:380, 66:353; Dedham Hist. Reg. 4:61; Braintree Ch. Rec. 22; Crane 23; Tingley-Meyers 103; GDMNH 682}

Information on the cited references can be found in a downloadable PDF: Key to References in New England Marriages Prior to 1800, the link to which is on each entry page.

I did re-join NEHGS, and my research and sourcing is so much easier and better!
Vic, I know that you are involved in many of the original settlers and NEHGS would certainly be of help. Do you suggest it for those of us who are not part of that specific project? I have found some very helpful articles attributed to their magazine, but it was when I was actually working on my very early part of the tree. Your group now has that sewn up so we can't work on those people. Is it still worthwhile to have NEHGS membership?

Darlene, I don't know what you mean by "sewn up."  As far as I know, all profiles in the PGM project are Open privacy, and anyone can edit them, so they are still available for people to work on.  If you mean that we have "done everything,' I know fer sure that that ain't so. devil  i haven't tried to use NEHGS for anything other than early American profiles, so I can't make a recommendation about its usefulness in other areas.


a) what Vic just said. ;-)

b) NEHGS/Americanancestors.org used to be focused predominantly on New England settlers-- both their (mostly) British origins and their subsequent descendants. But, as the rename of their web site reflects (they used to be newenglandancestors.org), they are moving toward broader American ancestry. For example, they some Virginia genealogical journals online.  I haven't tried to use it for much beyond New England ancestors, so can't speak to HOW far afield they've gone; I am aware they are adding more resources regularly. In my experience, the quality and reliability of their content far exceeds most of what's found on Ancestry.com.

Thank you Jillaine and Vc, I'll look into it some more and see how affordable it is. I do know that what I have used from them in the past has been excellent. And Jillaine, there are some good references and books on ancestry.com too. It all depends on the user, I believe. Lots of lazy researching out there. Or maybe I should say "passing the time" research. Even when I started years ago, I wasn't sure how to go about it or what was reliable and what was not reliable. Sometimes when I look at my early profiles, I gasp in horror! Heaven knows I have still a lot to learn. It takes time to learn and people like you two do help a lot. Thank you.

For members who cannot afford or do not wish to spend money on genealogy, there is much more help here on WikiTree than meets the eye.  We have so many resource pages that can connect you to good sources.

And don't forget what Jillaine said - list the source you use for your data. It helps a lot to know where it came from even if it is not the perfect source.

Since we are a collaborative worldwide tree, others will expand on your work if they know where to start.  

Thanks Jillaine for helping us understand sources!!

After reading this, I feel the need to go back and remove everything I have added. Because even census would then be considered derivative. Every source available, except first hand knowledge, has been touched by the human hand and we all know that too can be sketchy and open to errors. I have seen plenty of censuses with errors yet they are acceptable sourcing. I look for consistency. if every record I can find, says basically the same thing, then I feel validated with the information. If I can get a headstone photo to match that information. I am even more comfortable. But, by looking at this list, I just wasted some of my time and and wikis. too.. because even if all of the dates match with other documents, these are still considered bad sources.. on that note, I bid you all a good night. Or a good morning..

Charlotte, derivative sources in and of themselves are not bad. But if you've seen one derivative source you've seen only one derivative source.  They're all different. That point is to understand what you're looking at. (And to be honest, I was also really trying to discourage the use of Millennium File and the various "U.S. / International collections" at ancestry.com which are pretty bad.

The census itself is not a derivative source. It was an original record made at a specific time.  

So be sure to understand the difference between a source and the information contained in it. A source is either original (like a census record) or derivative (a typed transcript of a census page or an index of that page).

The information on the source is either primary or secondary.

This page explains it all well:


Bottom line, you did not waste any time. Unless you relied on the Millennium File.

Gravestones, particularly, as you mention, an image, not a transcription, is a good source. Just be sure that other information that's consistent with it doesn't, itself, derive from that gravestone! ;-)

Consistency is, in general, a good indication, Charlotte, provided that consistency is among multiple sources or references that are independent of one another. In many cases, such as ancestral files, trees on ancestry.com, Millennium File(s), family data collections, etc., the dates and relationships may well be all from one original source.

That is, one person may guess that John Doe is the son of Joe Doe, or John Doe was born on such-and-such a date. Someone else copies that info, which gets copied by yet someone else. After that information propagates, it appears everyone agrees that the original guess, now lost in time and space, looks like consistent agreement.

On the other hand, consider a birth record, a marriage record and a death record that all indicate John Doe was the son of Joe Doe (or born on such-and-such a date). Because the information in those records likely came from different people at different times, that consistency is more likely closer to the "truth."

[ninja'd by Jillaine]

Oh yes, I agree. But, if the Millennium File is consistent with an actual birth record, and certificates, will, etc.. then I consider it as reliable. If not, I do not. I have found them to be the least accurate as well and understand the ideal is to not use them. I wasn't trying to encourage anyone to consider them other than what they are, and apologize if I did. Having taken over 3k photos of headstones and then researching my photos for any other information, spouse, etc, when adding to Find A Grave, I know many are not correct. I try to always notate the correct name, or spelling when entering in the notes or bio section.. I hope people understand that per Find A Grave rules, we have to enter what is on the actual stone. A lot of us, who do research our memorials, hope a family member will come along and help with errors. Or take control of the memorial. But, I never consider it a source in itself, only if it is consistent. Maybe I am doing this wrong?? I don't know. Good night everyone.

I wouldn't say, "...the ideal is to not use them."  As Jillaine has said, the important thing is to know what you're looking at.  If you are aware that Millenium File (for example) is not a highly reliable source, you know that a more definitive source is needed.  On the other hand, if an unreliable source is all you can find, you note that the data from that source is uncertain, and hope that you, or someone, finds something more reliable at a later date.

As I've mentioned before, a source is simply where you got the information on a profile.  If we all understand what rock solid sources, and conversely shaky sources are, we know how good the support is, and where better sources are needed.

Related questions

+2 votes
1 answer
+11 votes
3 answers
177 views asked Aug 30, 2015 in Policy and Style by Michael Stills G2G6 Pilot (397k points)
+36 votes
13 answers
+33 votes
8 answers
+29 votes
2 answers
+21 votes
5 answers
+26 votes
7 answers
+4 votes
2 answers
161 views asked Dec 23, 2017 in Policy and Style by Richard Shelley G2G6 Pilot (203k points)

WikiTree  ~  About  ~  Help Help  ~  Search Person Search  ~  Surname:

disclaimer - terms - copyright