[ No headline ]

WHEN I was a teacher I would often give my classes a written assignment, and there was always a student (usually one who sat near the back of the room) who asked how long it should be. The question had a depressing effect on me; my enthusiasm was for the challenge of the assignment I had devised, and the last thing I had considered was its minimum length. That, of course, was the point of the question: What is the least I must do to avoid being penalized for the brevity of my response? The question was alw ays unanswerable -- the matter being one of quality rather than quantity -- and students found out too late if they had written at insufficient length. Similarly unanswerable is the question of longevity for books. We did not have much of a problem with books until just over a century ago, when high-acid paper made from wood pulp and using an alum-rosin sizing largely replaced paper made from rags. This explains the odd circumstance that books published before 1850 are often in better condition than later volumes. I have a 17th-century edition of ``Seneca,'' by the Dutch scholar Gronovius, which is in much better physical condition than some books I re ceived as school prizes during World War II.

The American National Standard for Permanence of Paper for Printed Library Materials (Z39.48-1984) was approved last year. But for many years most university presses and many ``trade'' publishers in this country have adopted acid-free paper which meets this standard. Books of this quality should carry the infinity symbol (a ``lazy'' eight) on the verso of the title page to facilitate identification. This would help librarians in the future;it would be an even greater help if foreign publishers also adop ted the standard, but most foreign books continue to be printed on high-acid stock. Acid-free paper is not so readily or inexpensively available abroad, and so it would be unreasonable to expect foreign publishers to be willing to change their present practices.

In a world accustomed to obsolescence, the realization that books printed on high-acid paper will not outlast their readers may not raise an eyebrow. After all, we know that many movies made in the '20s are lost to us, and how can we judge the quality of any tenor voice before Enrico Caruso? We also know that much of the literature of bygone ages has been lost and that scholarship devoted to the surviving fragments is one of our most popular indoor blood sports. So why all the commotion?

Experts tell me that there are 76 million books in our research libraries so fragile that any handling would cause further damage; that this represents about 25 percent of the total holdings; and that in another 20 years a further 38 million will deteriorate into this category.

We can now reword the question with which I began: What is the least we can save to avoid being penalized for the inadequacy of our response? Naturally, we want books to last longer, but the question is which books do we want to keep forever? What is at issue here is our understanding of the world of our parents and grandparents. If we take no action to save it, much of the record of the period 1870 to 1950 will disappear.

As a federal funding agency particularly concerned with the nation's history, the National Endowment for the Humanities has been addressing this problem since 1979; the agency has supported the training of archivists, librarians, and administrators in preservation issues and techniques, the development of cooperative centers, and planning for national approaches to preservation.

The agency's efforts in this arena are aided by the active cooperation of such national institutions as the Library of Congress, the Association of Research Libraries, the Council on Library Resources, and the Research Libraries Group. In the private sector, the Andrew W. Mellon Foundation has been more than generous.

The value of the past is self-evident at the National Endowment, since history is a discipline included in the humanities in the wording of the enabling legislation. For the society at large, however, the questioning continues. How much is the past worth? Can we afford the cost of preserving even a fraction of it?

How long should a book last?

Harold Cannon is director of Office of Preservation at the National Endowment for the Humanities.

We want to hear, did we miss an angle we should have covered? Should we come back to this topic? Or just give us a rating for this story. We want to hear from you.