On Not Using Libraries
Here is one example of my guilty knowledge: I try not to read anything until I have three times been pressed to do so by reputable sources-There are countless articles that people have pressed upon me as vital or brilliant or decisive that I have not read. I have not read them because my network does not turn up two other people who, in person or in print, share that judgment. Think of how effective this device is in reducing one's "must read" list by perhaps 80 to 90 percent.
Is it risky? Not at all. When I started out in the field of organizational analysis, there were only a handful of books and no journals. (Since I am a very slow reader, certified as flunking an Evelyn Wood speed reading course taken when I was a new assistant professor in 1960, this is perhaps why I went into this area and did tolerably well for some time!) I worried constantly that someone would preempt me, making whatever article or book I was doing redundant. I now know that the world is far too disorderly for that. The rarity is any convergence among scholars, on theory or fact. Our theories and facts are so imprecise and diffuse that littIe replication is possible even when we try.
A quite different reason that one can use the "three urges" rule for discriminating reading is that while no replication occurs, there is also no great divergence. Scholarly products are like holograms; each bit contains within it the essential properties of all other bits. Read one and you have read most of them! Better still, read the first and last chapter, or every third page of an article, and if you are reasonably familiar with the field, you know the essential argument and data. We are simply not clear enough to produce new ideas or new data in each of the thousands of articles and books that come out on, say, organizational culture, or organizational design, or sources of group conflict. Take a moderately narrow classification category using Dewey Decimal or Library of Congress or whatever, and use a random number generator to select items within it, and I suspect a 10 percent sample will miss few ideas or data, and will provide a branching daisy chain of citations to reach all the others. Considering the mountain of material, there is no way that all but a few shards can be unique, and thus essential. And three others will find those and urge them upon you.
It does suggest that the frantic efforts to save space through microfilming and CDs has less to do with increased knowledge than with hologramic proliferation. If so, librarians may be benign in recording our conceits and accreditations, but not in furthering "scholarship."
I have referred to skimming and skipping. It is essential with most of the material I have in hand. But let me mention a contrary principle I find useful: when I really learn, really benefit from some print material, it is because I have read it very carefully. But one can only find time for a few of these a year, say 500 pages. One odd example: In 1979 or so, I found that I had to examine the U.S. Senate's hearings on the conflict between capital and labor, held in the late 1860s I believe. There had been just too many references to it. The nearest copy was in the New York Public Library, and for some inconvenient days I journeyed in from Stony Brook on Long Island to read it in that forbidding institution. I quickly found that in one sense I didn't have to. All the familiar passages that I had seen in scholars' articles were clearly underlined in the worn library copy. But I became intrigued with what was underlined, and what was not. The latter was often fatuous, repetitive, trivial, or so it seemed. Then I realized that what was leaping out at the famous scholars who had quoted from it, and leaped out to me because of the underlining, did so only because it made sense in our time. It was our reading back into those troubled days that made those passages stand out. For those testifying and interrogating, those marked passages could have been cornmonplace. I tried reading the passages that were not underlined. I read them very carefully, trying to listen to their time, trying to imagine how anybody could say these trite or fatuous things sincerely (or even insincerely) and why they would repeat them. I began to get a different sense of the period. I began to sense the wide scope of a stubborn insistence that large organizations were violating people even as they covered them in goods. I could find no quotes to underline to prove this, but I hope to return to that some day and make a convincing textual analysis. The usual quotes complained about the organizations and the conditions, and were insightful enough, but they jumped to conclusions that fit ours, not, I believe, those of the speakers of the time. They did not live in a society of organizations as yet, and were able to see its novelty in a way that we today, great grandchildren of that society, take it for granted.
Today one can buy this volume free of underlinings from Arno Press, and read It without having to slip out into the rain or snow for a smoke, and without having to scribble out quotes in longhand in an overheated room which no dictaphone or typewriter could violate, but I rather liked seeing what the others thumbed over and put ticks beside. But my main point is that there are unique documents, and there should be slow careful reading of a few books. Not everything is a hologram, even if 99 percent is.
If there is so much redundancy, then the following estimates will not be so alarming. I estimate that only about 10 to 20 percent of my reading is useful in the sense of adding something. The rest I have heard before (even though I may not be aware of it at the tim~l, at least, appear to forget that I know things even though I really do). I may have to read something to make sure there is nothing new there that I might have missed. Perhaps half of my reading is of this sort. The rest is just dull labor, padding, repetition and inconclusive data and arguments. One works very hard for that 10 percent that will constitute an increment in knowledge. Could we just clear out the libraries of the other 90 percent? Of course not, since my 10 percent will not be everyone's, and we have to go through much of the 90 percent to be sure it is redundant, a hologram that perhaps is not quite as felicitous, or pithy, or accessible as our favorite hologram. And finally, we are all very slow to learn things and require great repetition and redundancy. If so, an idea whose time has come will have to insinuate itself into countless articles and books for us
to even notice it. "I will see it when I believe it" the deconstructionist statement goes, and believing it takes much repetition. So perhaps you are guarding a necessarily large mountain after all.
But l still do not browse there, or even send my minions there all that often, because most of what I read is neither published nor voluntarily selected by myself. About 95% of what I read is institutionally, or structurally, determined for me. I make only trivial choices within this stream, and the stream has nothing to do with libraries. I do not conduct a search" for 95% of my reading materials, instead, I am the object of someone else's search! There are people out there who are constantly sending me my daily words. These words come in the form of undergraduate student papers, graduate student seminar papers and dissertations (since I typically chair about six dissertations and sit on another eight committees, that can be a heavy burden), papers and book manuscripts from former doctoral students (with time you collect a lot of intimate ties, and depend upon these ties to keep you abreast of the field) and colleague friends and acquaintances, manuscripts from scholars I have not met, manuscripts from journals for refereeing, book proposals and book manuscripts from editors, and only rarely, a paper or book manuscript that I myself have heard about and solicit. The appearance of student term papers and seminar papers fluctuates with the season; dissertation work goes on steadily; and the material from colleagues etc., that comes mostly by mail, arrives at the rate of about five pieces a week, and most of that I read only the title or abstract (if there is no abstract, it is not likely to even be scanned).
Students aside, the rest of this population searching Out me and others probably has a 10% chance of getting comments from someone they send material to. When I mail out a draft of one of my projects~AlDS, crisis of capitalism, society of organizations, flexible production systems
-it will generate comments from about 10% of the recipients. The exception is in unusual cases where I make a special appeal and know my recipients are working in the same area. Here the response rate might be 800/~but that responsiveness cannot be abused. You do this maybe once every other year at the most, and you acknowledge their contribution personally in detail, and in a footnote in the next draft. Thus, much of our work is on unpublished materials, and we are sought out, or seek out others to send material to, mercifully bypassing the library.
This does not mean that the world of scholarship resembles a vast lottery where you have a 10% chance of scoring. If I mailed at random to all listed sociologists I might only get a half of one percent responding. My world is highly clustered. I read the people I know and they read me, and since there is so much redundancy and so little novelty, we do not suffer from this insularity and clustering. Indeed, it makes what search procedures I use more efficient. For example, when I started out to work on the AIDS epidemic, very few sociologists were working in the area, and I personally knew none. But I knew that people' at Berkeley were a good bet, so I called a friend there, who put me in touch with some others that I had never met. When I want to find out if there is anything on the role of city governments in supporting flexible production systems, I will call one or two or three people in political science, economics or the business school at MIT, and they will refer me to, say, someone in Italy.
Go to Part 2
By Charles Perrow
Department of Sociology
Yale University
I am delighted to be associated with humanists in Tony Orum's kind introduction, even though I would distinguish between a social scientist and a humanist. But what I am going to say today about the kinds of work I do and how it relates to libraries will probably confuse the distinction between the social sciences and humanities even further. One might think of a sociologist as rather well ordered, investigating identified and demarcated areas, using statistical data perhaps, but at least using high technology devices to search the literature and continually refine one's area. In contrast, the humanist might be considered to read widely, ranging over a large amorphous area, keep crabbed notes on three by five cards, and wander in and out of the materials and the libraries. The humanist mulls and turns over the pile of three by five cards endlessly; the sociologist should know what she needs to learn about, and with all the wonderful retrieval devices we have, zero in on it and capture it in a table or a tight theoretical statement.
But what I am going to say in a variety of ways is that my search procedures are disorderly and accidental, structured by others and often fortuitously, they are pressed upon me by the exigencies of time, students, and publishers and journal editors rather than by any "rational" process that libraries can help with, and that the main thing I require of libraries is that they build ever stronger brick walls to keep that mountain of literature from engulfing me.
I require libraries to hide most of the literature so that I will not become delirious from the want of time and wit to pursue it all. There is just too much material. The problem is not access, it is the reverse, containment. And when I need to poke a small hole here or there to tap a tiny bit of that coded and classified print material, I will do so in the safest possible manner. I will send someone else on that risky mission to your soul, someone indifferent to all that is left behind in the alleyways when they leave. I use a "work-study" employee, or a secretary, to quickly run in and preferably xerox an article or chapter, or if necessary, get a whole book, and run out again and bring it to me. Or I would send an innocent graduate student who has the time to linger if they want, someone who still has the happy delusion that they have a lifetime of continuous reading in which to catch up. Were I now to browse the stacks, as I could do in the luxunous day~ ~ st~~ent status or that of a very new ass?stant prolessor, I would drown, o(panic, and certainly lose my way.
One major caveat to all this: I am not referring to the specialized historical collection, e.g., tugitive radical newspapers trom 1960 to 1~75, or the pay ~oks of inside co~ractors at the ~ SeW(ng Machine plant in I 880. Libraf)es are absolutely essential in these cases.
Let me first indicate the range of my work at present. It will suggest how ~a~ant interests and a rather long career make it impossible for me to ever ~ant to work as libraries expect me to work, for example, using their bibliographic search tools, computerized or not, or suggesting in person titles or stack areas or whatever. I will suggest that should I do so, l would have to jettison se"e(al areas of s~ong interest and settle for one or two. don't thin~ I aft~ unique in this variety of interests, though it certainly is a characteristic of tenured professors more than untenured ones.
1)1 have to keep up with sociology in general, but do so b'r ristening to wheaQues rather than reading. From casual contacts in the department and from sociology conferences, a~d a lime skimming of the book review jo~nal in my field, I get a sense o~ what is going on, what is new, what is hot. As a result I may skim through two articles a month from one ol our main journals, and once a year read a part of a book dealing with sociologica~ matters outside of my specialties. lelut most of what I get is trom talliing casually with colleagues from these other fields, and they learn about my fields horn me in much the same way. Ot course, some just teati a great deal and are much more inlormed than I am, but I get enough to get by on. ~o libraries. (ln large part I am still trading on everything I learned as a student and as an assistant professor, when I luaw much moie about my Ismalter) field than I do now. But even then I ~ion'1 recall using tib~aries that much~)
Mv area of specialization is organizational theory, or complex organization~, arnl since I teach in this, have to keep up. But most of what I read is unpublish~ ~ pressed upon me by others, rather than sought OL£t by rre by browsinQ the journals or the stacks. I read st~~ent papers; pa~rs sent by oolleaQues whom may know or may not; papers sent by journals br tele~eeir,~ ~d manuscripts from friends and from publishers. I get to know about the work of others through the work of still 0th-ers, and then may send someone to the library for material to consider for dlass lectures or class assignments. ~f,e library just lies there, a mounlain of books with a protective wall of bricks, and book runners ma~e their forays, in and out with no browsing.
3) Tlie next area I try tO remain current in is that of accidents in high tech industries, the resull of a book published five years ago called Normal Acc~eaIs: LY~'n~ ~th k#gh R~~A Technologies. Though l am in a sense ~e with the pro'blem, having said my say, the industries are not done with us. Shortly after the book came out a series 0! major catastrophes appeared to illustrate my theon.es~-8hopal, Chemobyl, Challenger, tr'a,frlez, and on and on. ~ecause of my interest, speaking requests, "expert" testimony, and conferences on the subject, I just have to keep somewhat abreast. IBut I do it through the current periodicals, and only rarely through a book on tire topic. The xerox machine and the book runner again, and even so only on the basis of a very informal search process. People tel~ me about an article in the Alianlic Mon th(y or Spectrum, or a student does a term paper and reviews the latest material. I used libraries in researching my book, and in the case of marine accidents and engineering materials on nuclear and chemical plants and aircraft, I actually talked with librarians, but very little. If you find the key lournal in an area, even a popular one like Aviation and Space WeekIr, I found it will lead you to more material than you can possibly use.
4) A major area of interest to me because it encompasses my tield of organizational analysis is the "crisis of U.S. capitalism." I teach a large lecture course in this, though most Yale students think it is curious that capitalism should be thought to be in a crisis. Here the periodicals and the professional journals in economics, political science and sociology suflice, and the current books that get prominent reviews )~ The New York Tiines. Occasionally l will have to go deep into an unfamiliar area, rather than treat it at the popular book level, and I will mention one case that I will relurn to. This is the new development of what are being called flexible production systems" in ~urope and Japan. They' are one major source of our industrial decline, since we seem to be incapable of develo~ng these ourselves.
But this is very new material. Most of the material on these Systems I use is still unpublished, and some that I can't use is in Italian, (3erman, or Japanese. Here libranes are only marginally useful I suspect, though I admit I have not gone to mine to find out. The material fits in no neat discipline or category, is very new, and sf111 not structured ioy any powerful theoretical statements. (The closest is a rather popular book of a lew years ago, The Second iDdustrial ~ivide b~ M'chael Piore and Charles SabeI.~ This is a case where there is not a huge pile of words that l must be protected from, but it is worth noting that as small as the literature in ~nglish is, I have the ~me to read only about 30 percent ot it. Compared to my usual sampling of 5 to 10 percent in areas that concern me, t~at is a lot. When I hear, as l did this morning, that someone reads everything on the shelf about a topic, l gaze in envy and wonderment.
sy Al~& is another area that I try to keep current on. I am perhaps twothirds through a manuscript, with a graduate student colleague, on the organizational aspects of the Al~S crisis. I will return to this since this too is a very new area, but with almost no scholarly literature on the
organizational response that I can draw upon. Indeed, that response is changing daily.
6) The last area I am working in is a projected volume, A Society of Organizations, an interpretation of the industrial revolution in the U.S. from 1820 to the present in terms of large, complex organizations. It argues that organizations, once just one part of "society," have absorbed, or vacuumed up, much of what was once society, and the result has been increasingly unmanageable system dynamics as huge organizations interact and collide without the buffer of small, autonomous, local, culturally distinctive groups-the society that has been absorbed. Here the search process should be the most conventional-historical materials by and large, and finishing up with the kind of materials I use in the area of the crisis of U.S. capitalism. But what I note here is no different from the other areas: I do not do a formal search, say, for labor's reaction to deskilling, or the Progressive movement's attempt to tidy up the stables of capitalism. The materials flow in through informal contacts ("Chick, I ran across this book about foundations in the first quarter of the century. It seems to contradict what I think your interpretation would be; better check it out."), student papers on the occasions when they share my enthusiasm for the topic, skimming the book review journals, and the ever branching path of citations. Already I have cardboard boxes of reprints and notes and some shelves of books that constitute more than I can digest. Already I have a mini-library of largely unread materials that I need to be protected from! I often think of turning all the books around so that their seductive titles no longer sing. Only because I have written other books do I know that somehow or other it is possible to use enough of that material to get by and not reveal the mute testimony of immense ignorance that the rest of it provides.
But Anthony Orum's generous introduction reminds me of some saving graces or safety nets that protect us from the calamitous knowledge that we have not read everything and that we best avoid libraries. These devices fall somewhere in the area of what we now call "bounded rationality," a testament to our ability to make out, imagine order, and muddle through, even though our world is not nearly as rational, linear, evolutionary or functional as our theories pretend. By imagining, or "constructing" order, we can act as if it is there, and thus act more effectively than if we were shorn of our innocence.
Reprinted from:
Charles Perrow, “On Not Using Libraries,” in Humanists at Work, papers presented at a symposium held at the University of Chicago on April 27-28, 1989, p. 29-42 (Chicago: University Library, University of Illinois at Chicago, 1989)
