- original sources as "material in its first oral or recorded form."
- derivative sources as "material produced by copying an original or manipulating its content."
Reasonable researchers may slightly disagree about whether certain specific sources are original or derivative. And that's ok. It really is. The genealogy world will not end.
What's important is that the researcher thinks about the source that they are actually using, how it was created, and how it came to their possession. Did they see the actual deed the owner signed giving their ancestor ownership to a certain piece of property? Did they see the record copy of that deed in the county recorder's office in the courthouse? Did they see a microfilmed copy at the Family History Library of the record book? Did they use a published transcription of the book made by a local genealogical society? Or did they see an abstract of title for the property in question which was compiled from extractions made from the record copies of the deeds? Whew.
Making that clear is what really matters. Letting others know exactly what you used really matters.
I tend to take a conservative approach--original is the first time something is written down and derivative is everything else. Usually original items are a will or deed an ancestor actually signed, the baptismal certificate signed by the pastor, etc. I just find that use of the phrase easier and more consistent. Anything that doesn't meet this "first time criteria" is derivative. Once I start making exceptions to that "first oral or recorded form," I start getting confused. I don't like being confused.
Describing something as original or derivative is not the same as evaluating its perceived reliability. Some original sources aren't worth a pewter of warm spit and some original sources are highly credible. If your analysis of a source only hinges on classifying it as original or derivative, then you simply need more hinges.
Don't get me wrong. It's crucial for researchers to understand the difference between original and derivative sources. But it's important to acknowledge that reasonable researchers may see a variety of shades of gray between those seemingly black and white definitions.
And that's ok. If a researcher clearly explains exactly what they used (eg. The Family History Library's microfilmed copy of the county deed record book) then others can use that to base evaluation decisions.
And if you don't know what source you used, it matters little whether you classify it as original as derivative.
That's classified as dreaming!
