Discussion  - 
Question for the community:  The reference "Massachusetts Vital Records, 1841-1910 - Daniel P Thompson - Death asserts that Daniel P Thompson was age 54 at death (Oct 17, 1852)."

Is this Direct or inDirect Evidence for the Birth fact?

Part of me wants to say it is indirect, since I have to calculate a date (range) from the information.
Cheryl Rothwell's profile photoChris Staats's profile photoWalt Quering's profile photoPamela D Lloyd's profile photo
I agree that this response was counterintuitive.  After all, with one exception, the US Federal Census never (directly) states a year of birth.  Shirley's answer is, however, correct.  MGP does say that the census response is direct evidence.  Later on the same page (15) Dr. Jones said that, "While direct evidence is what a source says, calculation based on information from one source does not make the evidence indirect."  Hence, subtracting the the given age from the date of the census, does not change the information to indirect."  I'm still adjusting my comfort level with that assertion.
This is one issue that I disagree with Dr. Jones (probably at my peril).  It is direct evidence of an age at death (or age at marriage, or in the census).  If I have to calculate a birth date, I think that's indirect - it didn't tell me an exact date.  It gets dicey if the death record said he was age 54 years, 7 months, 5 days at death.  I still have to calculate the exact birth date.
I'm having a real problem with this as well.  In one paragraph, Dr. Jones says "calculation based on information from one source" ... but in another, he says "combining information from different sources does not make evidence indirect" ... and another place he says, "indirect evidence is a set of two or more information items ... only when they are combined."  

So ... now I have three different definitions in the span of 3 or 4 paragraphs.  I'm really confused.  I may have to do a blog post just to get my head straight.
+Ed Thompson and others.

We aren't taking about Census, we are talking about a Death Record.

Based on what Ed provided, I think the answer is Indirect. We DON'T have the exact birth Date. We have a range. But, I think I agree with Randy, that if the Death record had provided the number of months and days, at the time of death, I might call it direct.

But, what is the Question that the Assertion is answering?

Ed's question was: "Is this Direct or inDirect Evidence for the Birth fact?"

The assertion only gave the Age and Date of Death. That would mean that he could have been born in 1798 or 1799, I think.

It's two separate assertions (direct) for the death fact.  But for the birth fact....  Tricky.  My first thought was 'indirect'.  But I don't think so.  The source does state clearly that he was 54 as of that date.  Even though you have to calculate a range to move further, I think that is more on the analysis end.  I guess I would treat the assertion as direct for the birth fact. 
While we are not discussing a census record, the basic facts are the same.  The birth year must be obtained by subtracting the age from the death year (or age at time of census).  I too am not comfortable with this being direct evidence.  My hesitation is based on experience, in that, if I had obtained the information from the informant in either example, I could not testify in a court of law that the informant had told me the birth year.  He/she did not.  Could I say what the birth year was? Yes, but that information was arrived at by a calculation from information provided by the informant, and not explicitly stated to me.

Direct evidence however, "..is an information item which answers a research question all by itself."  If the information concerned with age and date of event is contained within the source (such as a line on a census, or entries in a death record), then that information, although consisting of two or more facts, and requiring a simple calculation, does answer the question, "In what year was the subject born?" all by itself. 

Like I said, I'm still adjusting to that myself, but it is not an unreasonable assertion.
I think the confusion lies with direct versus primary. A birth date on a death certificate is not primary information, it is secondary information because the informant likely was not an eyewitness to the event. (MGP, Jones, p. 11, 56). However, it is direct evidence of age because that one piece of information does answer the question "in what year was this person born?" all by itself. Yes, you do need to calculate it but only from other information on the death certificate, not by combining other containers of information. See MGP, page 15 about 2/3 of the way down on the page.
Except that you can ONLY calculate the birth year when the age information is combined with the death information. By definition (p. 14), that is indirect evidence.
Jenny I think that is only when the death information is in a different source document
You're right ... "calculation based on information from one source does not make the evidence indirect."

I really wish Dr. Jones would have clarified "calculation based on multiple information items from one source ..."
I don't think its THAT critical, lol
There is a difference between calculate and infer and I think that is why we seem to be at odds with Dr. Jones' thinking. With 'calculate', the doc has concrete info we can work with, say for calculating his birth year from info in his death record.

Contrast this with the "purely in our minds" inferences we draw from a document such as "since he was dead by this date he cannot possibly be the father of my ancestor proved to have been born seven years later."

+Randy Seaver and +Christine Sharbrough's comments are compelling.
The late Richard Pence did a study of calculated birth dates, found there are three methods and two are likely to be inaccurate. Direct or indirectly incorrect.
Whether evidence is considered indirect or direct is entirely dependent on the research question you are trying to answer. If you are asking "What YEAR was David Thompson born?", it is direct evidence. If your research question is "What was David Thompson's birth DATE?", it is indirect evidence, since the age in years cannot, by itself, answer the question. This exact scenario (although the example uses a tombstone instead of a death record)  is used as an example in the BCG Standards Manual in Standard 29, discussing direct and indirect evidence.
I pondered this last night, and am far more comfortable with Dr. Jones' assertion.  I phrased it as a math problem: If A - B = C, and I am given both A and B, then I have the answer C.  Actually, Dr. Jones says that, "Combining information from different sources does not make evidence indirect." (pg. 15).  So, if I get A from one source, and B from another source, I still have the answer C.  Works for me.
I do not yet have MGP, so I don't know the context of what you've quoted. However, the very definition of indirect evidence is (from Tom Jones' Advanced Research Methods Course handout for "Devloping an Evidence Orientation"), "Information, or an inference from information, that cannot alone answer a genealogical question; instead, researchers must combine it with other evidence." In your equation, if neither A nor B gives you C, and you have to put them together, by definition both A and B are indirect. Ultimately, the classification of whether or not evidence is indirect or direct is simply that - a classification. It's far more important, to use the mathematical example, that A+B actually does equal C - that we've completed an exhaustive search, collected enough evidence, analyzed it correctly, and correlated it accurately. 
Actually, evidence is distinct from information.  "An information item becomes evidence when we consider its implications for what an unobservable event, identity, or relationship might have been" (MGP, 13).  In the example above, item A is a piece of information from whatever source.  Item B is another piece of information from whatever source.  The evidence is the derivation of C from the calculation involving those two pieces of information.  That derivation, since it answers the question, "What is C?" directly, is direct evidence.
We're saying the same thing as far as terminology goes (and I'm not sure why Tom used the term "information" in his definition of indirect evidence in the syllabus). The problem lies in the idea that evidence, not information is used to arrive at conclusions. Information is simply something on a page, and doesn't "become" evidence until we extract that relevant information and apply it to a answer our question (a restatement of what you quoted above). A and B have to be pieces of evidence to support a conclusion, not pieces of information - and that evidence is either direct or indirect. 
Correct.  The distinction is not unique to Dr. Jones or genealogy.  In the intelligence field, one of the definitions of "Intelligence" is the "result" of information which has been subjected to a process.  "Information" is what goes into the process, "Intelligence," or, in our case, "Evidence" is what comes out.  I suspect that our readings in MGP will reveal a similar distinction between "Evidence" and "Proof."
Add a comment...