Our Libraries: Can We Measure Their Holdings and Acquisitions?.

We are a " knowledge society," a society that devotes a large and ever-increasing part of its gross national product to the production and distribution of knowledge. The "knowledge industry" has been growing at a faster rate than most other sectors of the economy, and the number of people working in "knowledge occupations" is between two fifths and one half of our potential labor force. Statements of this sort were advanced fifteen years ago, and statistical research on "knowledge production" has been going forth ever since.1 Yet, on some rather elementary questions regarding "knowledge embodied in print" we know so little that we must admit deep embarrassment. Knowledge contained in books and journals has in fact been the earliest object of measurement in this area. The size and growth of our library collections have been taken to be the most reliable and most easily obtainable indicators of our engagement in knowledge production. For decades apodictic statements have been passed around to the effect that "knowledge" stored on the shelves of our libraries has been doubling every ten years, or every seven years, or some such number. A few of us have been skeptical about the meaning of such assertions; we have asked, for example, whether one hundred books really represented twice as much knowledge as fifty books, and whether fifteen journals really conveyed thrice as much information as five journals. But we have not questioned the physical meaning of the measurement. We have not questioned the reported "facts" about the rate at which the numbers of books and journals on the shelves in our libraries have been increasing. We believed the stories about the doubling every few years, because we had not known that the librarians themselves were so very unsure about the collections under their control. Now I know a little more about the extent of my ignorance and I want to share it with others. By sharing the realization of my ignorance I may relieve my conscience as an investigator of the dissemination of information.