Research Heresy

Research Heresy: As a high school student it was my ambition to become a librarian. As a result, my first paying job, as well as years of volunteer jobs and classes, were spent in libraries, working, studying their systems, and hiding out. This put me at a singular advantage all through college and grad school, and I was generally regarded by students and faculty alike as a sort of guru of library research. I could, and did, do major graduate research papers on obscure subjects in 3-7 days, start to finish, and get highest marks. So it was no great surprise to me when the faculty asked me to do a seminar on research for the assembled faculty and students in the department’s weekly lecture series. The faculty sat down expecting I’d give a serious harangue to my fellow students encouraging them to stick their noses to the grindstone of the library, and the students braced to snooze through yet another “scholarly paper”. I seemed to shock everybody, pleasantly or unpleasantly as the case may be, by actually explaining how I did my research. You see, because I understood the system, I understood how to “cheat” the system. So I explained that one could get information out of the library in bulk, in less time, without so much “nose to the grindstone.” I was, I’m afraid, even flip about it. And I admitted that the recent research paper I wrote that was considered by the faculty to be “of publishable quality” on French Revolutionary Festivals, was in fact the product of one weekend’s cramming. Well, the students didn’t fall asleep, and my advisers were looking not at all happy with me, but, I thought: “I have a mission here—I must lead the righteous to the path of better grades, no matter what the cost!” So I did. And here is my lecture (with some new tricks I’ve learned since):

THE LAZY MECHANIC: First, to preface my system for doing library research, let me explain the story of the Lazy Mechanic and the Industrious Mechanic. According to my Dad, who comes from a line of machinists, engineers and other early technocrats, it’s better for a factory to hire a lazy mechanic to service it’s machines, than an industrious one. He says that an industrious mechanic will carefully oil, and service and repair a machine so that it works perfectly (with lots of his help and labor) and never needs replacing. This will leave you with an outdated and labor inefficient machine, that is in too good a repair to justify scrapping. A lazy mechanic, on the other hand, doesn’t want to spend his day oiling and repairing and servicing. So he will use his mechanical knowledge to alter the machine to make it more efficient, (expending extra time at the beginning), so he needn’t work on it for the future. He will also ignore doing any kind of servicing that isn’t actually necessary to the operation of the machine. In short, he will expend his energy on making it a better machine, rather than maintaining an inefficient status quo. This is one of life’s important lessons. If something takes a lot of time, trouble, and effort to do, when it is something that should be simple, chances are that the system is set up wrong. In this case you need to apply your effort to finding a better system that works faster, rather than simply dumping all your effort into forcing an inefficient method to work for you. This will take extra time at the beginning, and save you time in the long run.

THE PROBLEM WITH THE SYSTEM: Doing research in the visual arts in a normal library is one of those inefficient systems that need fixing. There is a simple reason for this. Libraries are predominantly set up for books. Books are predominantly set up for the words in the books. Library systems therefore are based on the assumption that you are looking for information, in the books, in the words. Card catalogs and computers in libraries file everything, books, pamphlets, maps, charts, posters, everything, based on the words in/on them. Ditto online search engines. The problem is, you are looking for the pictures. This is a very big problem. The bigger the library, the bigger the problem. The solutions are relatively simple, if time consuming. The “sensible” library approved solution is to deeply think about the subject you are looking for, and to think of related subject headings, based on the words in the books, that might lead you to the pictures. This is a good system to begin with, (and the only one you can use with search engines) but it is still a case of working within the inefficient system. It works, sort of, but it will swallow lots of time each time you use it. It also presupposes that you know a lot about the subject you are researching before you even begin.

WHY IT WORKS BADLY: For example, you are looking for costume ideas for a play by Chekhov written in 1905. What are the best subject headings to find costume pictures? If you answer Russian costume, Chekhov, and Russian history in 1905, you will get books showing colorful peasant dress, play scripts without pictures and lots on the Revolution of 1905, all of which will help you very little. What you needed to look up was Photography in Russia before 1917, the painters Valentin Serov and Mikhail Vrubel, the history of the Moscow Art Theatre and Konstantin Stanislavsky, and the costume collection of the Hermitage Museum. How do you find this out? There is no way, working within the regular library system that you can find this out except by hitting subjects at random (Russian painting, history, sculpture, engravings, photos, costume, etc., etc.) and hoping you hit one that will stick. It is different for each country and period and play. There is no escape. Or is there?

SWALLOW THE WHOLE THING: For the long run, there is. It just costs you a lot of time at the beginning. What you will need to do is to abandon the filing system and embrace the whole library for several days. In a large university library (your best bet) you just go to the top floor and work down. You begin at the beginning of the stacks, and walk them. That is to say you look at the titles on each and every shelf, no exceptions, and when a title or a binding looks interesting, you pull it off the shelf, and look for pictures. Four out of five books you pull out will be duds. When you find a good one, write down its call number and it’s subject heading. Save these notes for later. My first published Article, Library Costume Resources; A Supplement was a list of subject headings and call numbers I compiled in just four hours of walking through the stacks of the CSU Fresno library:

KNOW WHAT YOUR LIBRARY HAS: Most major research libraries have many sections, all of which contain information on costume, and all of which you should “walk”. The main section is what is called the stacks, these are stacks of books, available for checking out and wandering through, usually filed according to the Library of Congress system. In addition to this main section are many others. First, usually extra large books are filed in an oversize book section or sections as part of the main stacks, this area contains a large proportion of costume information. The reference section has books too valuable for check out, and books that are references for other books and information. These include indexes like The Readers Guide To Periodical Literature that can help you find articles in magazines on your subject. Within the reference section is usually the reference desk collection, normally only used by the reference librarians. These are exclusively indexes useful for finding other works, including indexes of periodicals, videos, and organizations that may be useful. Also in the reference area are computer databases that list articles and books available in your library, your region, and even sometimes on the database itself. Most large libraries also have video, film and audio collections, a map collection, new and old periodicals (magazines and newspapers), rare book collections (which include not only books, but other fragile old items like photos, stereographs, fashion plates, paper dolls, posters, theatre programs, manuscripts, renderings, letters, drawings, and scrapbooks), and a regional collection (with local “rare books”).

A REALLY BIG LIBRARY: A very large American library also is likely to be a Federal Government Depository Library, which means it will have government publications filed by an annoying system understandable only to a special librarian stationed in that section to help you. For posters of the history of American military dress, info on W.W.II civilian home made clothing, and Smithsonian publications, this is the source. The 100 biggest libraries in the US were also given, on B&W microfilm, a complete copy of all Sears Catalogs from the beginning in the late 19th Century to the 1970’s (Oh, rapture!) Big universities often have to break up their libraries into several buildings across campus, and so often have a law library, science or medical library, and a business or technical library. These places need to be explored as well. For instance, at UGA those Sears Catalog microfilms live in the UGA Science Library. Books on period cutting, sewing, and millinery, will be located in a technical library if one is kept separate from the main library. Business libraries have information on unions (photos of strikers and police), servants, prostitutes, farm workers, garment workers, and the textile industry. Big libraries also have the Union List of Periodicals, and other massive library indexes useful for finding very obscure books that you may order through interlibrary loan. Many also have microfilm and microfiche collections with works too bulky to get in print. For example, the Victoria and Albert Museum has sold color microfiche copies of a large selection of the theatre costume and couture renderings in their collection to libraries, many more than could reasonably be copied in a large book.

THE MAGIC AP’S: Within the periodical section is one of those sections you would never likely find in a conventional catalog search, the “general interest” section known in the Library of Congress system as AP. AP’s include all such general interest magazines as Time, Life, The Ladies Home Journal, Godey’s Lady’s Book, etc. If the main content of the magazine is not fashion, but “general interest”, no matter how much fashion information is in it, it is filed in this section. Naturally since most of “women’s magazines” fall into this category, it means it contains huge amounts of costume information under one, non-costume, heading. Since big libraries often have collections of magazines in this section going back to the 19th, or even 18th centuries, it is an extremely useful one for getting contemporary fashion information, photos of important figures, and even patterns for clothing, just by looking up magazines in the period you need. If your library doesn’t seem to have old periodicals on the shelf, check and ask. Usually there are older volumes filed either in a separate old periodicals section, or the old volumes are on microfilm or microfiche in metal cabinets nearby.

LOOK FOR RED BOOKS: One of the bizarre things to keep in mind is that books with lots of pictures often are in standard sizes. Art books with color reproductions of paintings are often kept in the oversize book shelves, and so this part of the library should always be checked extra carefully. Other books come in sizes that you will eventually recognize from experience to be likely picture books. Another clue is gold letters on red binding. For reasons known only to the publishing industry, very many books with this color combination also contain pictures. Pull anything that you suspect contains interesting visuals, and write down the numbers that are useful.

TAKE NOTES FOR LATER: While doing this systematic search for all important visual costume information in your library, you will be acquiring information even more randomly than by hunting subject headings, however you will be gathering much more visual information in total than you could do in dozens of subject searches. You also cut out the word based dead ends much more quickly, by bypassing them altogether. You may also use this system backwards with the subject heading system. That is, once you have your list of numbers and headings which are useful, you can apply them to other parts of the library like periodicals, reference, and government publications to find even more information. These sections too must be walked and mined for numbers and headings on their own. Once you do this you will be able to easily put your hands on three fourths of whatever you need for costume research for a show within an hour or less. There will still be that one fourth that obstinately will require a conventional catalog search, but by doing this walk through the stacks just once, you will have the majority of your costume research gathering done before you even start. You can then concentrate your time on using your research instead of finding it.

THE OBSTINATE 1/4 THAT NEEDS A NORMAL SEARCH: You can shorten the search several ways. One is that you choose a wide variety of related headings, but only take enough notes to get to the right stack. That is, when looking for Inuit dress for the children’s show The Ice Wolf, you look up Inuit dress, art, textiles, history, photographs, tools, religion, Native American costume, art, textiles, history, photographs, tools, religion, and circumpolar dress, art, etc. in the catalog, making the broadest possible search, but you do not write down or print out every number. Instead, you look at the numbers in each subject section, and write down only one each of similar numbers. This is enough to bring you to the proper shelf, where you can hunt using the visual system. If a particular title in the catalog seems promising, right down that number as the sample number for it’s section. This way you can often find a hundred books while only writing down a dozen numbers. Another important option in conducting a normal search is to ask a reference librarian for help. Reference librarians have MA’s in library science, which means they spent two whole years studying library systems, research, and reference books. In a conventional search, they are the unquestioned experts, and can often shorten your time spent considerably. Taking a course or two in library science yourself, will also make doing conventional searches faster.

FILE IT IN YOUR HEAD: Over a very long time, of course, there is another solution. You learn enormous amounts about costume history, and can retrieve them from memory. To do this, you need to make a point of checking out and reading/looking at all the good books you find, one by one, till you can accurately do period shows, without a large scale research search. To do this kind of systematic mining of costume information for your lifetime, I recommend Janet Arnold’s A Handbook of Costume, which contains “the word” on how to do costume research from the leading costume historian in the world. Janet said, among other things, that costume research does not stop with picture books, but should also take you to museums, archives, and old houses to see actual garments, paintings, sculpture, accessories, legal documents like patent records, and so forth. Ms Arnold practiced what she preached. The costume world knew Ms. Arnold was the most determined and dedicated costume researcher when her Patterns of Fashion 3, came out, and we saw she had made patterns from 16-17th Century clothing that was in an advanced state of decay. The decay came from the fact that these garments were grave clothes and had spent a long time with a decaying body in each one. (Yuck!) Thinking about the dedication it must have took for Ms. Arnold to do this should convince anyone that library research, even with all it’s difficulties, is, comparatively, a piece of cake.

A LOW DIRTY TRICK: While I do not recommend this method for anyone who is not truly desperate, it does work. If you are truly “stuck” on a particular costume (usually a uniform in my experience), and are in some remote place where you cannot possibly drive to the next city to look for further information, and you have exhausted all the local sources, you can go to the video store. (WHAT!!!) Yes, sometimes you have no alternative. For example, in Fairbanks, Alaska I had looked for days in the largest library in the State for a picture of a prisoner in a 1953 Russian gulag winter uniform. The best I got was the fuzzy B&W photo of Solzhenitsyn, from the waist up, in The Gulag Archipelago. I needed something clear, in color, and full length. So, I rented A Day in the Life of Ivan Desinovitch, and took notes. The reason that doing this is not advisable, except in an emergency, is that if the film costumer had made a mistake in research, I would have promptly perpetuated it. However, one may generally avoid this by being careful about which movies one rents, and which pieces one steals from them. One should never, for instance, steal costumes worn on leads, because usually they are altered to suit the fashions of the period the film was made in. It is both unnecessary, and dangerous to take bits of fashionable, non-uniform dress from movies, since you can get this information elsewhere, and these parts too are sometimes altered. Movies done before 1960 are often highly fanciful in terms of costume, as are deliberately stylized movies like Bram Stoker’s Dracula, The Boy Friend, and anything directed by Ken Russell. In general, the best films to steal from are large budget, post 1960 English made films, and post 1970 European films. There are many highly historically accurate American made films, particularly westerns, but there are an equal number of fanciful, and plain sloppy ones, so it isn’t safe as a bet. For more on the alteration of costume history in film costume read The LA County Museum of Art’s Hollywood and History, and Seeing Through Clothes by Anne Hollander. The safest films in terms of uniform hunting are those with Berman’s and Nathan’s, Angel’s or Cosprop as the costume makers, and John Mollo as consultant or designer. So this means if you just got stuck to do The Cherry Orchard, and can’t find what a Turn of the Century Russian station master, student, and post man look like, you can rent Nicholas and Alexandra, and Dr. Zhivago, and make do, because these two films contain those three uniforms between them.

AN INDUSTRIAL REVOLUTION IN RESEARCH: Using this system will speed up your research considerably. However, there is something else that will speed it up more. The Xerox machine can, if you let it, double your time on top of it. The first secret is getting a library copy card, so you don’t mess about with dimes. Then, when you have found your books, do not check them out, but rather, in the library, quickly scan for what you want and copy it. Return to the stacks and get more books. Copy them. Repeat. You can walk out of the library in under three hours, with all the pertinent information from 100 books, in a small folder. This saves considerable hauling, and leaves the books back at the library, where others who are working on the same show can get at them. If you staple up these folders into booklets each time you do this, you will end up with a compact research library of specialized period information. As things are, back home, you will have only one booklet to look through for your source material, as compared with fumbling around through dozens scattered all over your room.

WRITING RESEARCH PAPERS AND ARTICLES: Because most costume jobs and training in the U.S. are connected with academia, costumers also must, as students and professors, write research papers with footnotes and quotes. You may speed up your research from word based sources by using the above Xerox system with a few alterations. First, you must, when gathering up your books, make a list of the author, title, date, and publisher of each book. Then assign each of these titles an abbreviation, also written down. Then as you copy relevant pages out of the books, write the abbreviation and page number of the book on the back of each Xerox. When at home, read the pages, putting highlighter on those sections you think you wish to use as quotes and footnotes. Divide the pages of quotes by subject, then order each section, and each page in a section, into the order in which you wish to discuss your topics. If necessary, cut and paste pieces of pages into proper order, but always remember to keep note of the abbreviation and page for each part you use. When it is all in order, you may start writing. You will find your writing will easily follow the path of your source material with a flow you have rarely felt before. As you use some part of the Xeroxes in reference or quotation, you may easily footnote it by comparing the abbreviation on the Xerox with the list you made at the beginning. All this saves days of inefficient fumbling about with stacks of books, and also makes it far harder to lose or forget small references. Staple together this sort of research too, and you’ll have a source book to check back on if you need it later.

Three People have written in with improvements to my ideas! Here they are:

I just stumbled onto your _Research Heresy_ article, and what a great find! I work as an image researcher (mostly historical stuff) and have discovered many of the same tricks that you mention. I swear I can smell a heavily illustrated book from a distance at this point!

I also have a Master’s in Library Science, and would like to share a little secret to help narrow down catalog searches. The official Library of Congress Subject Heading for “lots and lots of images” is PICTORIAL. Sometimes I’ll do a catalog search on a broad topic keyword (such as Civil Rights) plus the keyword “pictorial.” One of the advantages of this method is that I can request that a specific book be returned to the library if it’s checked out, whereas I would have missed it entirely by browsing the stacks.

Keep up the good work, and happy hunting!


You seem to have developed the same research style I did – pillage and burn through the right sectors, scavenging anything that looks useful. Two refinements to the technique. Make a photocopy of the title page to use for the bibliography information. And, keep a notebook to list of what you have looked at and whether it seems useful for the current or future projects. I have a long list of books to ignore, a list of books that would be useful on other projects, and those that are good for the current one. (I’m a technical writer not a costumer, but research is research.) —

I also read you article on research and enjoyed it. I am also a trawl-net bibliophile. I have a tip to add: rather than using abbreviations to track photocopies of materials to be footnoted I write the call no. on a small and Xerox the ISBN page at the beginning of the book with the post-it attached, this page has author/title/publisher/date/LC# on it so I don’t have to copy them out… then I slap the post-it in the margin of each page I copy… the yellow is invisible to most B&W copiers and the call number links me to the copy of the bibliographic info and will lead me directly to the book on the stack shelf if I have to go back and pull it again… I developed this procedure as well as copying from back to front (in order to avoid collating after the fact) in the course of my information-panning all libraries large and small that ever cross my path.” —Dan Wasserman Coord. Adult Education, St. Louis Art Museum e-mail:

SAVING MATERIAL FOR POSTERITY: Before everyone crucifies me for encouraging the deforestation of the planet through copious Xeroxing, please let me tell you about another endangered resource: research material itself. It is an unfortunate fact that nearly all paper made and used for books and magazines from about 1820-1990 was acid based paper. This means nearly every book, newspaper and magazine produced during that time is on paper that ultimately self destructs, like a time bomb. The libraries of the world are in a crisis over what is to be done to save these resources. To save the information on this paper requires highly expensive de-acidification or moderately expensive copying, or comparatively inexpensive microfilming. The simple economics of library funding means that libraries will mostly use microfilm, (which is troublesome for costumers), as well as have to pick and choose which books and magazines are “important” enough to go to the expense of copying. I need hardly tell you where items like Turn of the Century tailoring and millinery manuals will rate in this lifeboat situation. Therefore, for the good of future costumers, including one’s self, we each need to notice when books like this are on the shelf of our local library falling apart. A book that is coming apart is a book whose days are numbered, soon to be thrown out. In the case of some pre 1910 catalogs, and drafting and sewing manuals, you may be looking at the last one of it’s kind. The libraries can’t afford to save all of them, so, you need to grab these books, before they are gone, plain paper copy them onto an acid free paper like Copysource, and bind up the copies, so they can replace the originals. At present I’m trying to copy dozens of them into paperless online format so that many people can have them at once. If each of us does this with a few books, we will save these endangered resources, as well as quickly and cheaply expand our own private libraries of rare research works. Expanding your personal library also saves lots of time doing research.

THE NON-LAZY, LAZY MECHANIC: If all this time efficiency still seems like “cheating” to you, relax. Just because you can do your research in a rush, you needn’t do so. In fact, if you are a research junkie like me, it just means you can get three or four times as much information in the same time as you used to do before. In terms of the costume and academic food chains, this is often the difference between having enough research material to feed your work on, or having all your work time eaten away by a research project.