What is Proof?
Proof is difficult to define, but one definition is "evidence sufficient to establish a thing as true, or to produce belief in its truth". What is presented here is meant to give some insight to the family historian and genealogist into generally accepted practices for establishing genealogical proof.
Scientists and historians generally agree that there is no absolute proof. Instead, they seek the highest level of probability that something is true, with the understanding that evidence may later surface that will call into question what is now accepted as fact. For the genealogist and family historian, the available evidence is not as precise as a scientific law or theorem but, properly collected and verified, historical evidence can be trustworthy.
Finding the Missing Pieces
Basic Principles of Family History Research
The genealogist or historian attempts to unlock the past by searching records and interviewing witnesses. Since not all evidence is accurate, and some is purposely misleading, they must carefully evaluate the evidence they find.
The Genealogical Proof Standard (GPS)
Begin your research with these miminal standards of proof from The BCG genealogical standards manual:
- A reasonably exhaustive search has been conducted
- Each statement of fact has a complete and accurate source citation.
- The evidence is reliable, and has been skillfully correlated and interpreted.
- Any contradictory evidence has been resolved.
- The conclusion has been soundly reasoned.
The image shows that the proof process never ends. New information or a fresh insight can trigger a reevaluation process.
How do I Verify and/or Prove My Data?
Family history data is authenticated by collecting evidence about it, and then testing that evidence to see if it is true. The level of probability that something is true will be determined by the number of "witnesses" (evidence) that attest to the data in question, and to the level of confidence placed in those witnesses.
What Types of Evidence Are There?
There Are Two Generally Accepted Categories Of Evidence:
Scientific evidence. The scientific method requires that a physical observation undergo empirical testing. To be considered valid, the results of these tests must be repeatable. For instance, water repeatedly boils at 100oC at STP, and the electron, though never directly observed, has its existence deduced from consistently repeatable phenomenae. (DNA evidence is a form of scientific evidence that can be used in genealogical research.) The physical aspects of nature work well with this method, but determining who actually gave the Gettysburg Address (or whether it was given at all) does not, as there can never be another physical occurrence of that speech, by that speaker, at that time. That requires "Historical Evidence".
Legal or Historical evidence. This method is used in court cases and by historians to show that a particular event happened in the past. A person or event from the past cannot be revisited or observed, they must be deduced from accounts of witnesses, which can include people and documents, DNA, etc. The past life of an individual, or the past events surrounding that individual, require this type of verification. This method is based upon the available evidences or witnesses and upon their credibility (see below).
Sources, Information & Evidence
The weight given your proofs depends on an evaluation of these three things:
- The Source – Original or Derivative
Where did it originate? Is it an original work or is it a copy, transcription or hearsay evidence (derivative)? (For example, an oral history can be original if the orator observed the person or event, else it is derivative if it is being relayed by a non-witness.)
- The Information – Primary or Secondary
Was it created when the event occurred and by a trusted observer or was it created after the event or by someone with second-hand knowledge? For example, a death certificate contains both primary and secondary evidence. The person's name, date and cause of death are primary. The names of their parents and their birth date are often secondary if the people providing the information were not present at the birth or if they did not personally know the parents. (Be careful – even primary evidence can contain errors or be falsified.)
- The Evidence – Direct or Indirect
Does it directly address the person or event being proven or is it indirect or circumstantial evidence?
What Types of Historical Evidence Are Accepted?
These are in no particular order and each must be carefully evaluated and weighed (i.e. assigned a level of confidence).
- Birth & death records
- Circumcision, christening and baptism records
- Family Bibles (family events are often recorded there)
- Diaries & logs
- Tombstones & cemetery records
- Wills, probate records, deeds and other courthouse records (e.g. civil or criminal records)
- Directories, membership lists
- Church records
- Masonic and other organization records
- Passenger manifests
- Immigration records
- Military enlistment, service and pension records
- Social Security Death Index (SSDI) and Social Security application records
- Census and tax records
- News stories and obituaries
- Personal knowledge
- Correspondence – letters, postcards, etc.
- Interviews — use caution and thoroughly research
- Family Lore — use caution and thoroughly research
Some evidences are typically given more weight or credibility than others. For example, legal documents are usually created by a professional, unbiased observer and are generally considered more accurate than a newspaper article or a book. A story passed down by word of mouth can't help but be embellished or changed over time and should be carefully evaluated before accepting it.
Are All Evidences Accepted As Proof?
It depends on the level of confidence that is placed upon the evidence. In a courtroom, for example, a witness is believed until they are shown to be untrustworthy through cross-examination. In the same manner, the researcher should cross-examine everything, remembering that any record can be in error, as they are vulnerable to mistakes, lies and misinterpretation. Everyone has experienced hearing a convincing story, only to find that it was a fabrication, and you won't search through census records long before finding an error. The researcher must determine the relative level of authentication they believe each piece of evidence presents.
Is There Any Absolute Proof?
There will always be room for doubt. Anything accepted as a fact must be supported by evidence which must be tested and verified and it should always be open for reevaluation.
Is DNA Evidence An Authentication Or Proof Of Ancestral Ties?
DNA is used by comparing test results with another person or, ideally, a large group of people who also have DNA test results. These results can show that two people do not share a recent common ancestor, which can save time and money by preventing searching in the wrong direction.
DNA test results can also imply that one person is descended from the same line as another. Larger data sets increase the confidence level – i.e. the amount of DNA tested and the number of individuals in the group increase the possibility of a relationship between people who match.
In addition to providing evidence that two people share a common ancestor, it can also indicate an approximate time frame within which the common ancestor lived. Having evidence of relationship to a particular family line can narrow the search and, with other supporting evidence, can provide a level of proof not possible with traditional methods alone.
As with other evidences, DNA can be misinterpreted. It is incumbent upon the researcher to understand the principles involved.
Assign a Weighting or Confidence Level
This is an important part of research as not all evidence is equal, some is more trustworthy than others. The assignment of relative merit to a collection of evidence is a subjective process that requires training and experience.
- Reliability of the source – determine a level of trust for this source witness (and reevaluate as more evidence is found).
- How relevant is the evidence to the subject?
- The level of trust is diminished if all the evidence comes from a single source?
- The level of trust is elevated if multiple unrelated, independent sources verify the evidence.
- Contradictions or controversy reduce the confidence level.
- Missing data is also evidence and can reduce confidence in other evidence.
- Is it first-hand testimony or second-hand (hearsay)?
- Anything not witnessed directly is second-hand or hearsay evidence.
- Registrations such as military draft registration and enlistment records are original records based on first-hand testimony.
- Social Security applications are original records based on first-hand testimony.
- Death certificates contain first-hand details about the death, but personal details like birth information and the names of parents are second-hand if the informant did not directly observe them.
- Death notices, obituaries and tombstones are often second-hand unless contributed by a direct witness.
- Census records and directories are often second-hand, depending upon who gave the information.
- Books, newspaper articles, internet pages are almost always second-hand evidence unless authored by a witness to the event.
- Family oral histories are second-hand – unless the source witnessed the event.
- Written family records (e.g. Bible records, correspondence, logs and diaries) are second-hand unless written by a witness to the event.
- Governmental and church records are typically more accurate than census records or family documents e.g. Bible records, etc.
- Information from a publication can be completely unsubstantiated – look for source references and examine and evaluate them.
- What was the motivation for the record? Is there a financial interest or a desire to manipulate opinions or to be related to a famous family, etc.?
- When was the record produced – was it recent to or distant from the time of the event? Errors increase as the time span increases.
- If controversy surrounds an event, was the record produced prior to or after the controversy began?
- Who created the record? Are they reliable? Does the author exist or is the author an impostor?
- Is the source unbiased or does it have an agenda?
- How was the record preserved – could it have been compromised?
- Is the record incomplete, are parts of it missing?
- Collateral or background sources are useful but can also be misleading. An example of collateral information are inferences derived from the names of family friends or a place of residence or occupation.
- Reevaluate your evidence after more evidence is discovered
Adjusting The Confidence Level
Reduce Confidence Level
- When there is only a small amount of evidence.
- If there is conflicting evidence.
- If the evidence is illogical or contains reality-defying facts, names, places & stories.
- If the time line of events is illogical.
- If controversy surrounds the evidence.
- If evidence is missing or incomplete.
- If the evidence is not directly relevant to the subject.
- If the evidence comes from a single source.
- If it is second-hand (hearsay).
- For oral histories, unless first-hand from a witness to the event.
- For books, newspaper articles, internet pages, etc. with no verifiable source references / citations.
- For records created long after the event.
- For "delayed" documents e.g. delayed birth or death certificates.
- If the motivation behind the record is biased or personal.
- If the originator of the record is untrustworthy or controversial.
- If the record was not carefully preserved – aka "chain of evidence".
- If the evidence is an inference based upon other evidence.
- If the data contains inaccuracies.
- For elaborately crafted stories explaining holes in the evidence.
- Be especially careful with sources that are known to have distorted or falsified data!
Raise Confidence Level
- As more unrelated, independent sources corroborate the evidence.
- For first-hand or eye-witness testimony.
- For original or primary sources – e.g. governmental records and baptism and christening records.
- If the record was produced when the event occurred or shortly thereafter.
- If the recorder of the evidence is unbiased and has no personal interest in it.
- If the chain of evidence is sound – i.e. the document could not have been tampered with over time.
- If the record is thorough and complete.
- With DNA, the confidence level is enhanced as more markers are tested and when the test results of two people match closely.
Citing Your Sources
Before sharing or publishing your research, you should provide references to your sources so that other researchers can review them. The information contained in the citation should allow the researcher to locate it and to know which piece of evidence it supports. Genealogy citations typically follow the Chicago Manual of Style. Typical elements of the citation are:
- Name of the person or event found in the source.
- Name or title of the source.
- Identifying indexes e.g. volume, series, page, file, call number, etc.
- Publisher or person interviewed and date of publication or interview, etc.
- Name of author, editor, compiler and/or repository.
- Electronic address, if used, (e.g. web site URL).
- Comments about the source (optional).
Falsification of Data – Fraud
Be careful what you accept as factual. In addition to honest error, there are many examples of fraudulent data in publications, genealogical and historical records, etc.
Primary sources are the least likely to be altered or forged, but they can contain inaccurate or fraudulent information nonetheless. Secondary sources e.g. publications, letters, oral histories, etc. are the most common targets of fraud and manipulation. (Some have even had tombstones made with their false proofs on them.)
Common motives for fraud include such things as people wanting to inherit fortunes; wanting to be descended from a famous lineage; wanting a noble title, fame or bragging rights; wanting to join lineage-based organizations; wanting to prove religious beliefs; even committing tax fraud to steal money; etc.
One of the most notorious perpetrators of genealogical fraud was Gustave Anjou, who lived from 1863-1942. He concocted and published genealogies for the wealthy, and they are well-done and appear authoritative, weaving false data in with true. In 1942, the New York Times noted Anjou had "developed a profitable business in the sale of mail-order ancestors". Many of Anjou's fraudulent works reside in libraries today to be used by unsuspecting researchers. Information on this man can be found at Fraudulent Genealogies.
A recent example of someone wanting to be related to a famous (or, in this case, infamous) family is a man from Oklahoma named Bill Phillips who introduced himself as Bill Dalton and who fabricated documentation, filed for a delayed death certificate using false information and modified tombstones to show that his grandmother was a sister to the members of the "Dalton Gang". Even seasoned researchers can be taken in by people like this. Fortunately, careful examination has exposed this instance of fraud.
(See "The Bea Elizabeth Dalton Hoax")
Gustave Anjou & Bill Phillips aren't the first people to fabricate genealogical data, and they won't be the last. For a list of impostors, see the Wikipedia article
Be careful in your research and examine all evidence.
- The Genealogical Proof Standard
- Brief Review of Research Standards
- How to Prove Your Family Tree Connections
- Five things you have to know about every record
- Fraudulent Genealogies
- Cite Your Sources
- Citations: A Guide to Creating Proper Source Citations
- Principles of Source Citation
- How to Cite Sources
- The Chicago Manual of Style Online - Citation Quick Guide
- List of Impostors - Wikipedia