When in doubt, follow the citations. That probably sounds a little pat, but it's always amazing how often you'll have a bunch of people citing the same source, sometimes at odds with each other, and you can even get a chain of citations where this person is citing this person who's actually quoting this person and... well, ideally you get back to a primary source or some jerk that doesn't cite anything.

Don't be afraid to think a little outside the box when it comes to sources. In pulp studies, if you want to find out something about, let us say, Robert E. Howard, you would look at the biographies of Robert E. Howard and his published letters. Which is cool. But the letters are sort of a periodic sampling from his life; they don't tell the whole story. So sometimes you have to go for different sources - like the published letters of H. P. Lovecraft, Clark Ashton Smith, and August Derleth (some of his correspondents) or Otis Adelbert Kline and Otto Binder (his agent). You wouldn't normally think they would have much insight to offer, but these are people that are all corresponding with and talking about Robert E. Howard. The same basic principle applies in any other history specialty: if you want to figure out about rum production in the Virgin Islands in the 1600s, your extant sources might be fairly slim - but if you expand your search to things like sugar and molasses production in the same period, and import lists in North American colonies that traded with the Virgin Islands, you might be able to find some more factoids that add up.

Finding a source is only half the battle; then you have to physically locate it and obtain a copy. If you're attached to an academic institution with a big library and access to interlibrary loan, this might be easy. Other folks might have travel - or see where a particular article was printed or reprinted, what's affordable. A number of older academic and quasi-academic articles are available online if you can narrow your search parameters sufficiently to locate them.

#HOW TO DO YOURSELF AN ARCHIVES RESEARCH

By: An American Archivist™ (and so probably less useful outside of America)

I’ve given out bits and bobs of this information over the subreddit in my ...several years here, but I don’t think I’ve ever written it up linearly! So here is a guide to how to hunt for archives. It assumes you, dear reader, already are pretty fluent in academic-level library research, and fluent enough in working with archival records, but possibly you’ve only worked with them in a “predigested” format, like a digital collection, or a subject-focused published collection, like a source book. So there is nothing about primary source theory and whatnot, this is just about that jump further down the history rabbit hole.

But what is an archives anyway? Is it a library? Is it not a library? How do you find one? How do you get in it?

Caveat: “An archives” here is being defined rather unpoetically as “a physical place housing and providing access to multiple records, arranged and described in some sort of fashion, under intellectual control via finding aids etc, all of which is under the control of a professional.” There are other definitions of “an archive” in use in academia, and beyond. Some people call a shoebox of receipts “an archive,” and nobody can stop them. And not all archives can (or even should) meet these qualities, but are still totally worth a historian’s time. (And yes, it has an s at the end in singular, why, because it does idk. It’s a shibboleth, you can shib your leth into the in-group with with a little sss noise, if thou so wishes, or not, if thou wishes not. Myself I shib, because I used to work at an archives with THOU SHALT NOT SAY ARCHIVES WITHOUT THE S framed on the wall.)

So before starting your research, I think it’s good to ask yourself a few big questions.

##What do I want to find? (No Books!)

If you want old books, you don’t want an archives. Old books are collected, and should be, but that’s under the field of Rare Books/Special Collections libraries, and they are organized differently, intellectually, and require different discovery methods. (Some places will essentially run both a rare books library and a traditional archives in one room, and that’s cool too.) Archives focus on unpublished things, and eschew published records as best we can. Basically, If multiple copies of this thing ever existed, not archival, look in a library first. So an archives should have:

1. Internal business papers: meeting minutes, agendas, limited financial records, ledgers
2. External business papers: “Whitelit” and “Graylit” these would be things that were sort of published (there were multiple copies) but wouldn’t have been widely distributed, such as: annual reports, project reports, newsletters, etc.
3. Ephemera (sometimes, not everyone collects this!): loosely defined as “printed stuff born to be thrown away,” advertising material of all sorts, birthday cards, postcards, general crap
4. Personal papers: diaries, letters
5. Academic papers: old professors’ stuff: notes, field journals, research material (non published), spicy intellectual correspondence fights (my favorite)
6. Unpublished media: non-commercial photographs, films, videos, audio recordings
7. Newspapers: oddly liminal category here: major papers try a library first, itty bitty local papers, try an archives first.
8. Religious records: baptisms, marriages
9. Civic records: business-type records mixed with marriage, birth, adoption and death records

Don’t want to mess with any of that? That’s cool, you want a library. Do you want any of that, and do you suspect it exists somewhere? Keep reading!

##How are archives organized? (Welcome To the Land Beyond MARC Records)

The biggest switch I think people need to make is to switch from thinking of material being organized by SUBJECT (which despite being a pretty new concept in libraries, Dewey, bless his racist soul, has trained everyone to think it of first, because that’s how we put books on the shelf) to being organized by CREATOR. Archival holdings are often tagged by subject but it’s very, very limited. When I write a finding aid (an inventory for a collection), I will usually give it 1-4 Library of Congress subject headings, with the mind that they will be harvested into some unified catalog, maybe Worldcat or ArchiveGrid. So that can be 1 to 300 cubic feet of intellectual material, sharing 4 measly, often very basic, subjects. Not great if you are searching exclusively by subject.

Books are organized in catalogs, with each book (or journal series, or DVD, or whatever) having it’s own little discrete record. These are called MARC (MAchine Readable Catalog) records, and have existed virtually unchanged since the card catalog was invented in the 19th century. One discrete intellectual unit, one record. Archives, however, do not have time for this nonsense, we have too much stuff and all of it is one of a kind, and so we have have multiple items sharing one record. This record is called a Finding Aid, and as you can see it contains in some ways a lot less information, but also a lot more, than if you tied all that paper together in a book and made a MARC record for it. But you’ll notice the records are focused around one person, the person who created them, and not their content.

Let’s consider the finding aid for the Margaret Sanger papers vs. a published selective version. You’ll notice (under the “index” tab) that the finding aid is pretty lavishly indexed with what creators are also included in the papers (like people she correspondended with.) None of that is listed on the book’s catalog record. Archivists think like that though, so they listed tons of creators, and then for subjects, it’s just like, “birth control, sex stuff, idk.” There’s a lot of subject tags I suspect any historian of Sanger would consider important enough to put on her papers, but they’re not here. Your job, not mine!

So, to think like an archivist, you need to think in fonds. Which means you need to think first: who made the records I want? This can be quite hard, and you need to learn to think down from subject. Let’s walk through working down from a subject to finding an archives for it.

Step 1. I want to research the history of the Internet.

Step 2. Who are major figures in the history of the Internet? What were major organizations involved in developing the Internet? (Insert me doing a brainstorm/deep Wikipedia session) I want to look for records created by:

• Advanced Research Projects Agency
• National Science Foundation
• Information Sciences Institute
• People on this page (focusing on the dead ones)
• Obviously a lot more leads you can dream up on this very broad topic, but this is a start

Step 3. How can I find these archives?

Google can be the enemy here, if you just start typing in “archives” willy nilly, because the word has been borrowed by everyone and their dog. Consider “jon postel archives” mashed into Google’s “I’m Feeling Lucky” will get you this page. Which is not Jon Postel’s archives, needless to say. Normally you’ll get better results with “papers” for people and “records” for groups, but if you try “jon postel papers” you get this, also no good.

Start thinking hard about where dead people’s stuff ends up. ARPA and NSF, as arms of the federal government, are going to be under the National Archives, DARPA is possibly going to be annoying to get at with security and FOIA requests etc, but NSF is pretty darn open. ISI archives are going to be housed at USC unless something very weird happened there, and lo! here are the records of somebody important! Now why did that not turn up in the googles? Because they run their finding aids on ArchivesSpace, and that software is not formulated for search engine optimization. Also the collection appears unprocessed, and has a “restricted” note with no information on what that restriction entails (Expires in a set period of time? Do you need archivist approval? Do you need a note from the Pope? No idea). This is a crappy job on some important papers actually, they should be embarrassed. But we have confirmed the Postel papers exist somewhere! And the key way I found them was by guessing where they’d have ended up, his employer.

You’ll notice a larger problem here: you basically already have to know the history of something to figure out how to research it in an archives. That’s why historians get the big bucks! Haha…

Medieval manuscripts. If anyone knows about one big beautiful database I've been missing all this time, please, lemme know. Nobody's ever really given me a straight answer on how to hunt them down. Here's a few things I've learned finding my own way.

For literary sources, edition bibliographies are the first stop. It would probably be a lot more productive to make editions of individual manuscripts, not the collated narratives we have, but that's not how we do things, so in any given edition you'll probably have a list of a few manuscripts the editor looked at. Then it's just a matter of looking those up in their respective archive catalogs. The problem is when it's a text that there are a LOT of manuscripts of, like Historia regum britanniae or Brut. Obviously nobody's gonna pay for 200+ editions of each HRB MS, so an editor will say something like "I used X, Y, and Z manuscripts for this edition of the text because they're the best and here's why they're the best." And the more I read the more I have a problem with the idea of "best texts" as just being closer to the author's intention, but that's another story.

Digitized manuscripts aren't too hard to find. Bam: http://digitizedmedievalmanuscripts.org/app/ for example, because digital humanities are wonderful and I'd like to marry them. I honestly don't know why more people aren't flocking to digitize everything medieval they can get their hands on (jk, I know the answer is funding). Libraries are usually super pumped to show off their digital collections, too, so digital catalogs are usually pretty nice and user-friendly.

But when you get down to things like charters, letter correspondences, and other stuff that generally falls into historians' neck of the woods, the problem I run into most is that I just don't know what I'm looking for, exactly. Bibliographies are a good starting point there, sometimes. Say I want to check out, I dunno, merchant's letters within a certain time period. Start by finding some secondary work on the subject, look up some good primary sources in the bibliography, and go from there. In the best cases, that'll take you to a library archive entry where it's not hard to find similar collections of letters you're looking for in proximity. But at a certain point, these things are hard to find just because they're still 'undiscovered'. William Marx 'discovered' a Brut chronicle nobody knew about 15 years ago, just sitting in the National Library of Wales. A lot of libraries will have an archive catalog entry that says something like "collected letters 1200 - 1300" and nothing more, and they're just gathering proverbial dust down there. So there's still a lot of archives-plumbing to be done, which is exciting, it's just hard if you can't physically get out there and plumb.

I've had great luck with both digital archives, and genealogical work. While the first are growing in importance for historians, I've seen the second tending to be ignored or even criticized (with some good reasoning).

With many archives making digitization efforts over the past decade, they often begin with making Finding Aids available online. Even simple keyword searches in Google can turn up some great resources that haven't been used before.

Even better, in my opinion, are family genealogies. By tracking relationships you can gain insight into why and how people ended up where they were. For example, by looking into family histories of Salem shipbuilders, I found that the last generation of builders were often related by marriage to those already established in town. More still, it gave me more names and locations to widen my search.

That being said, you have to be wary of family histories. Just like every other secondary source you should be validating the source against anything you can find. For example, the House of Seven Gables complex promotes their Retire Beckett House as belonging to the Retire Becket who built Cleopatra's Barge at the beginning of the 19th century. This is based off of a deed at the Phillips Library for the land the house was originally on (the house was lost by Retire in 1803 during his bankruptcy) which is marked in the Finding Aid as being from the late 18th century, though a reading of the dead shows it is from the 1720s..

Problem is, Retire had a grandfather of the same name, who is the purchaser on the deed. The house was divided in thirds after his death, and the portion that remained on that original lot went to another family member - John Beckett.

Without verifying the family and local history no one would know. In fact, Seven Gables never has after the initial submission for Historic Landmark status from the early 20th century. And because this change doesn't fit their established narrative, nothing will come of it.