I think people are confused by the definitions of primary/secondary sources.

As an example, almost all textbooks are actually secondary sources. The primary source may be a journal article. And yes, often, those secondary sources are far superior than the primary sources. Insisting on reading all those journal articles as opposed to a good textbook is simply indulging in a fetish, IMO.

> WTF do you know about me or anyone else to judge their choices?

I don't, and did not claim to. And yes, one can and usually does judge at some level. I don't need to know much about someone to have an opinion that intentionally touching random strangers' private parts in public is problematic.

Try not to be so sensitive. My comment was not a criticism of anyone.

> Your ignorance of the need for direct knowledge of things shines through.

Now who is making assumptions without knowing me?

I spent years being paid to do research at academic institutions, so it's not a random person on the Internet saying it. Sometimes the primary source is fantastic. More often it's inferior to better written pieces.

The reason journals like Rev Modern Physics have a high impact factor is precisely due to the fact that distillations can be superior to primary sources. It's a journal where a highly influential person in the field writes a (lengthy) journal article that is basically a well written, accessible tutorial of the whole field.

Have you tried reading Maxwell's original works? Do you not think there's a good reason pretty much every EM textbook teaches Maxwell's equations in a very different form from what he proposed?

Thanks for your explanations. It would be lovely if primary sources and data were some kind of ground truth with a capital T. But no: they are often messy first pass experiments. If we knew what we were doing it would not be research ;-) A dose of Ian Hacking or Paul Feyerabend or Nancy Cartwright could be curative.

Many, many times in historical research the "primary source" (I'm using your definition here because in historical circles the primary source is not always the original artifact or document) is lost to time. It's then up to historians, anthropologists, and other researchers to figure out whether the artifact in question actually existed and whether there's agreement among other sources on what it said.

In historical research at least having a primary source like the "actual letter written by Jefferson" is often rare and only found in more modern historical sources preserved by stable institutions. A great example is the Library of Baghdad where we know many texts existed but we still don't know the breadth of all the available texts nor their contents.

Thanks and very interesting. Are you an historian?

> in historical circles the primary source is not always the original artifact or document

What does it mean in historical circles? I know people will often use a digital scan, a reliable transcription, etc., and not necessarily the original - but then isn't that based on the assumption that it's a reliable representation of the original?

I didn't mean to suggest primary sources (by my definition) are always available; I was just trying to give an example where 'primary source' connects the world of ideas to the physical reality. Yes, life, including historical research, isn't that neat and simple.

I totally forgot to hit "Reply" to this. Apologies!

> Are you an historian?

I published some work in undergrad but stopped afterward (much to the chagrin of my professors.) I still keep up with historian circles, folks who either are academic historians or participate in the general community.

> I know people will often use a digital scan, a reliable transcription, etc., and not necessarily the original - but then isn't that based on the assumption that it's a reliable representation of the original?

Veracity of a source is a very, very complicated topic in historical research. Sometimes documents are authoritative. For example, the American Declaration of Independence is a well-preserved document that has been commented on millions of times at this point and has many, many copies. We can be very sure that the Declaration of Independence exists and says what it says. This isn't the case for a lot of historical documents.

Until use of the printing press became widespread in a culture, distributing copies of documents was laborious. Usually a scribe would have to read the original and copy it. Writing media from this time (paper, parchment, tablets, etc) were also usually expensive. Each document was usually considered an individual work of art and so even if a scribe was copying a document, we had no idea whether the scribe editorialized the work, forgot to copy some of the work, or otherwise modified its original intent. That means that while we may find a document with a title in our modern era, we don't know if that document was the authoritative original or even if the contemporary readers actually consumed the version of the document we have in our possession or a different edition.

Then there's the question of veracity of the claims of the document itself. A lot of surveys and statistics from historical writings are false. There are usually numerous cultural and political reasons for this (maybe the culture didn't count women as people, maybe the culture only counted landholders, maybe the author didn't see a certain tribe in the area as people to be counted, etc.) and so even when we're reasonably confident of when a document was authored and by whom it's hard to be certain of the veracity of its claims.