Proof or Doubt?

[Henry]

This posting is prompted by the thread “What Standard of Proof do you Apply”, but it does not really contribute much to the thread as it is currently running. My point is a general one and relates to the nature of FH investigation. I hope this is the correct forum to be posting it. I shall be brief.
Suppose we take our model of “proof” from the Law Courts. I have to declare that I have no expertise whatsoever in law and I stand to be corrected, but it seems to me that in a court case there are two major influences which are brought to bear on the jury: evidence and argument.
First, the assumptions have to be that the evidence before the court is as full as it reasonably could be; that it is not mistaken or inaccurate; that it has not been falsified or rigged. Then, on the basis of such a set of understandings the members of the jury feel they can rely on the evidence when coming to their verdict.
Second, the success or failure of the prosecution and of the defence often depends upon the persuasive arguments of the respective barristers. The arguments must not be specious, but be logical and not import information which is contrary to the evidence. But they should not lack imagination or flair; they should be insightful; and that they should make connections which the jury might not be expected to make for itself.
Then at some stage the jury has to decide on the verdict. Now here there are phrases such as “the balance of probabilities” and “reasonable doubt”. One can only approximate to the notion of truth, here; all one can have is a feeling of confidence that what one decides is just and in accord with the evidence as it has been presented. Not only that, but if the jury have got the verdict wrong, can they live with the consequences of that? We know of cases where innocent people have been incarcerated or hanged. There was neither truth nor proof in those cases, there could not have been. They had got the verdict wrong. There had been merely a sense of confidence in the jury that they had got it right.
And this is, I think, as much as we should ask of ourselves when carrying out FH research. Having accumulated as much evidence as we reasonably can and having given order and shape to that evidence and set it in the context of other evidence, the question we should ask ourselves is not “Is it true; is it proven?”, but “what doubts do I have or what level of doubt do I have that I have got it right? And can I live with the consequences if I have got it wrong?” If the answer to that second question is "no" then we must search for more evidence or perhaps look at the evidence we do have in a different way.
Fortunately, unlike our day in court, we can constantly ask those questions of ourselves and can review our decisions in the light of new evidence and our own persuasive arguments. Henry

[Olde Crone Holden]

Henry

In many proven cases of wrongful conviction of a crime, it transpires that evidence was suppressed or even falsified.

In Family history research, you have to ask yourself not "Is this right?" but "Could this be wrong if a, b, c etc".

I often see inexperienced researchers saying something like

"He is the only Joe Bloggs on the census, so he must be my great grandfather".

That is not proof and it is not evidence either.

Certificates often contain errors, half truths and downright lies, but that is not really a problem of proof, if it is the correct person.

In other words, if you are following Joe Bloggs who married Annie Arbuckle, then you have proved the existence of your great grandparents and you know you are following the correct bloodline, even if they did lie about their ages, their marital status, their parents and their occupation! If you don't get a certificate at all, then you could well be doing someone else's tree, not your own.

SoG suggest three pieces of primary evidence as proof. Primary evidence includes certificates of course (but not census, which are secondary evidence) and surprisingly, things like family correspondence, family bibles etc.

The aim is firstly to prove that Joe Bloggs existed at all, and then to prove that he was YOUR Joe Bloggs and not someone else's Joe Bloggs.

However, the more evidence you can amass from any primary or secondary source about an individual, then the more confident you may be that Joe Bloggs is your man.

You can never rule out the deliberately concealed - a non-paternity event (what a mealy-mouthed expression, lol) or a false identity etc, but family history is not a game of certainties, it is a game of "beyond reasonable doubt".

OC

[Peter Evans]

OC,

I agree with much of what you say but I feel that the requirement for three pieces of primary (or even secondary) evidence is unrealistic for most of our ancestors in the C19, let alone earlier. It may be technically true that the censuses are secondary but so too are the GRO certificates and all transcriptions, and it could be argued that the photographic copies are secondary as well. In practice we could hope for three details as persuasive evidence. There are also occasions when one has to trust elimination procedures. If you have married off or killed off all the other people of a certain name in a village then the one left standing is likely to be your ancestor.

Peter

[KiteRunner]

If you have married off or killed off all the other people of a certain name in a village then the one left standing is likely to be your ancestor.

You would have to be very sure of who all the ones who married and died were, and that the married ones didn't marry a second time to become your ancestor. Also, it could be that there was another one of the same name who moved to the village or didn't appear in the baptism register (e.g. non-conformist.) I think the more different documents you have, the more confident you can be that you have the right person, and certainly letters, family Bibles and wills are invaluable as sources of evidence whereas certificates can be misleading. Even gravestones can be wrong, though!

[Henry]

I am glad my post has generated some discussion. I want to make it clear that I am not against rigour in research. Quite the opposite. My description of the evidence in court was meant to emphasise that we must do everything we can to collect stuff (primary and secondary) that is both valid & reliable. However I agree with the spirit of what Peter says and that is why I emphasised the importance in research of imaginative argument - of which his "last man standing" is an example. This does introduce the logical concept expressed in the words MUST BE. There are dangers here, but logic can be very fruitful if it is handled with care and the researcher has the abiding intention of reverting to the scientific concept expressed in the word IS.
Then we review our decisions along the line of Kiterunner and ask ourselves "Can I go on from here or must I go back?" It was this that really prompted my original piece. It was an attempt to free one from the shackles of three primary pieces, when three may not be available. (Or if they are, they may not be as valid as you had hoped, it seems). We go forward knowing we have to keep going back. Check, review, revise, move on/back. My purpose behind identifying the notions of doubt and confidence was to give the power of decision making back to the family researcher. BUT it does not take away the responsibity of being rigorous at all times.

[Mary from Italy]

Even if census evidence isn't considered primary evidence, it certainly has very strong weight, especially, for example, if the person in the census is living at the same address that appears on a birth, marriage or death cert, or in a will.

[KiteRunner]

Yes, Mary, but some people new to family tree research think that if they find a family on the census that consists of "Head", "Wife", and sons and daughters, they can take that to mean that the head and wife were married to each other and all the children are children of both parents, which as we know, is often not the case, so census entries do need to be treated with care. For instance, my great-grandmother is listed as unmarried in her father's house on one of the censuses when she had actually got married 2 years previously, and her husband is listed as Head in a different house with a Wife and children! But censuses are certainly a very important source for family tree research, I wouldn't deny that.

[Merry Monty Montgomery]

but some people new to family tree research think that if they find a family on the census that consists of "Head", "Wife", and sons and daughters, they can take that to mean that the head and wife were married to each other and all the children are children of both parents

Yes, I made that mistake at the start!

This isn't helped by Ancestry encouraging the addition of "other surnames" to the indexing, many of which I'm sure are incorrect.

[Olde Crone Holden]

I think because ANY written source can be faulty, is the reason why SoG suggest three pieces of primary evidence.

Obviously there are times when this is not possible and you have to use whatever is available.

I do not have three pieces of primary evidence for my 2 x GGF, yet I know he existed because I wouldn't be here today if he didn't. What I do have is a lot of circumstantial evidence. I have to be very careful not to get carried away and turn "very likely" into "absolutely definitely".

I agree with Kate - last man standing doesn't make him yours! I have recently seen someone work up a very credible scenario which they have now set in stone as fact. It is no more than a credible scenario and there are other, equally credible scenarios which would fit. But there is not a shred of proof, just circumstantial evidence. This would not be good enough for me.

OC

[Henry]

OC,
And it would not be good enough for me either, but it would be a "possibility" and gives me grounds to move that way and explore further.
You mentioned earlier that FH is not a game of certainties, but one of "beyond reasonable doubt" and with that that I agree entirely. Circumstances must determine, however, what is reasonable. As Peter implied in his posting, the criteria for what is reasonable when researching records in this country which relate to the 19th and 20th centuries cannot fail to be more demanding than when working on foreign archives for say the 17th century. And that is why I worry so much about employing notions such as proof and truth because they, along with certainty, are logical concepts. The only logic we are entitled to apply is in prompting the search for new evidence - to support the shift from "must be" to "is". Or, to be more elegant, from the logical concept of possibility to the scientific tool of probability.
My beef is simply this, the decision to move or not to move my research in any direction should be my own and should be based on my feelings of doubt and confidence in doing so. Henry.

[Darksecretz]

If I might Hi-jack this thread a moment,...

can I ask a question?

what would be consistuted as 3 peices of "primary" evidence, I should know this, but my memory for some things is somewhat lacking..

I'm sure that there would be other members out there as well that arent sure what is primary/secondary etc...

[Phoenix]

You can place the most reliance on documents that an ancestor wrote/witnessed/used. That would be considered primary evidence as it is straight from the horse's mouth.

If the information is written about an ancestor, but they have no opportunity to correct it, it is less reliable. That would be secondary evidence.

A post 1753 marriage requires the signatures or marks of bride, groom, officiating minister and two witnesses (in c of e). A pre 1753 marriage by banns just has an entry in the register which might have been copied down perhaps 11 months after the event. Both may contain errors, but you might regard the first as primary, the second as secondary evidence.

[Katarzyna]

If we negate the BMD certs and censuses as primary/secondary sources because of the possibility of mistranscriptions then for most people there is very little primary evidence out there apart from wills and other signed documentary/ court evidence. Many will not have any wills, manorial records or court evidence to look to.

For the majority of folks I would say their ancestry is pretty much easy to follow using BMD and censuses and parish records. It is only a very few if any odd characters in our family history that signal an uneasy doubt.

[Peppie]

You can place the most reliance on documents that an ancestor wrote/witnessed/used. That would be considered primary evidence as it is straight from the horse's mouth.

If the information is written about an ancestor, but they have no opportunity to correct it, it is less reliable. That would be secondary evidence.

A post 1753 marriage requires the signatures or marks of bride, groom, officiating minister and two witnesses (in c of e). A pre 1753 marriage by banns just has an entry in the register which might have been copied down perhaps 11 months after the event. Both may contain errors, but you might regard the first as primary, the second as secondary evidence.

Ohhh so the 1911 Census becomes the first Primary source as it will actually be the householders returns they filled in......

[Uncle John]

My beef is simply this, the decision to move or not to move my research in any direction should be my own and should be based on my feelings of doubt and confidence in doing so. Henry.

Exactly. What you put in your tree is what you are comfortable with. I am a lot more rigorous with my tree now than when I started, but I do have some quite big twigs that were grafted from other people's GEDCOMS for which I don't have anywhere near 100% proof. I've kept them in because the computer is the easiest place to store the information, but I know what my own uncertainties are. If I get a hit from GR or elsewhere, I make it clear how confident I am about the information.

[merleyone]

This thread has been most illuminating and explored the concept of proof/truth with regard to Family research fairly comprehensively. I do not think I can add much to this debate but would mention that reverting to the court analogy, the proof required for a convction in a criminal case is proof beyond reasonable doubt, whereas proof in a civil case is proof on the balance of probabilities, a lesser standard.
It does not matter whether a crime is under investigation or family history is being researched, the proof sought in either scenario is that of proof beyond reasonabe doubt but the very nature of researching family history means that we will enounter proof on the balance of probabilities long before we attain proof beyond reasonable doubt, if ever. I concur with Henry that at that stage we should not cease our search but continue striving for the higher standard although the reality may be that we stick with the balance of probablities for a long time, maybe forever, simply because the definitive certainty is not there to be found.
For any responsible researcher, accepting proof on the balance of probabilities in these circumstances should not be a green light for slapdash research but the credibility of such proof must depend entirely on the quality of the evidence and the dispassionate way in which it has been analysed. Investigators seek merely to establish the facts and anyone who allows their own opinion to cloud their judgement will probably unwittingly seek evidence to confirm that opinion rather than analsying all evidence and accepting the fact of whatever it reveals, even when inconvenient.

I have rambled on a bit here, so my apologies for that,

Merleyone

[Henry]

Uncle John,
I am in the process of putting together a post on the Twice Baptised thread. I have two such. The first I researched myself from LDS films of the registers and that, together with census data and a death certificate, convinces me pretty well that things were as I think they were. The second is from a source which is impeccable: chapter and verse, all original spellings and never known to be wrong. But because I did not actually research the data myself I am hesitant.
However, the first of these is a far more central and signicant figure than the latter, so my concern and my caution swing back the other way. But, as I see it, all that is part and parcel of the dynamics of tree growing/building. You know where the unreliable twigs are and you have some idea on how much weight they can bare. Henry

[Little Nell]

Quite often when I go back over my notes, I find I am deleting some connections and yet this often enables me to add others.

THe most frustrating bits are those where the evidence (primary and secondary) hasn't survived or is illegible and then you just have to use guesswork and accept there will be blanks.

[Henry]

Merleyone,
Thank you for pointing up the continuity between the balance of probabilities, reasonable doubt and beyond. If we view them as continuous and not as discrete elements then there is at least the potential for smooth progression towards a level of confidence that satisfies us. As bits of corroborating evidence (primary and secondary) are gathered so we become aware that the balance of probabilities is becoming stronger. (Or conversely weaker if we suffer a set back!!) This sounds a bit fanciful I know, but that is the way it works with me. As you say, one just has to be very careful not to be seduced by an imagined green light.
Henry

[Henry]

Little Nell,
Yes - the gaps. they certainly disrupt the smooth ride I rather optimistically described in my previous reply. I have a good number of those. One I describe in another current thread, "Two Josephs or One?" At most the time span is nine years but the stories are separated by two continents. The balance of probabilities is very, very strong that there is only one Joseph, but it has not been shown beyond my reasonable doubt and I must keep looking. (Don't I sound priggish - don't answer that). Henry.