The following discourse assumes the reader has some understanding of the basics of nagashizuki-style papermaking, either through reading one of the two references above or through hands-on training or experience. To define the parameters a bit, I would suggest that an excellent quality, entirely traditionally-made sheet must be made of fresh, locally grown and harvested kozo processed and made into paper in cool or cold working conditions, with cold water, in order to maintain fiber and formation aid freshness.(2) Such fiber, in turn, must be cooked in an alkaline solution made of wood or other plant material ashes and water, and beaten by hand. The formation aid used during sheet forming must be rendered from fresh (or possibly fresh frozen or once dried) plant material and cold water. The finished paper must be dried on boards outdoors in the sun. When such traditionally made sheets are expertly formed they are not only very consistent, but often lustrous and possessed of a curious soft yet crisp feel, when compared to more quickly made sheets. In fact, most nagashizuki papermakers in Asia and in the West are not able to make all of these steps routine. But many who feel the traditional methods are important are trying. At the University of Iowa Center for the Book (UICB) we use Japanese fiber that has been: processed in cool or cold working conditions; cooked in lye of wood ashes sometimes mixed with soda ash (sodium carbonate); beaten by hand; and formed into sheets using locally grown fresh or fresh frozen tororo-aoi roots as formation aid. We are working on growing our own fiber and plan on eventually offering a special grade of paper dried on boards outdoors in the sun. Even with great attention to detail we often end up with an irritating defect: fine strands of fiber in the sheets. While these may be attractive or desired in some papers intended for artistic applications, they are considered a defect in a sheet made to exhibit superior quality and mastery of the craft. Fine quality nagashizuki mending tissue used in book and paper conservation must have very even fiber separation. If fiber strands are present, it is very difficult to tear the small, shaped pieces required in repairs, reinforcement, or consolidation. Excellent bast fiber separation may also be crucial to the artist employing nagashizuki techniques and seeking a particular visual effect. So, where do these problem strands come from? How can they be eliminated within the context of traditional technique, thereby retaining the natural fiber luster and the softness and crispness? In general, strands in the finished paper result from one of the following: poor quality raw fiber, improper cooking, improper picking, improper beating, or insufficient mixing at the vat. Each of these categories is addressed separately below. A key aspect of kozo quality is fiber freshness. Kozo bark should be taken from one year old shoots, not older trees. The scraped and cleaned bark should be from that season's harvest, not older stocks. The bark should be clean, and free from reddish, diseased, or tough areas. Older, scruffier kozo fiber requires more cooking alkali, time, or both for tolerable fiber separation, other factors being equal. After importing our kozo bark from Japan for many years (and being unsure of when it was harvested) we are now working on growing our own from imported Japanese treestock. This will help us ensure both fiber quality and freshness. Insufficient alkali or cooking time are likely to result in more strands in the finished sheet. We are working hard to find ways of determining which lye solution produced from wood ashes will give a successful cook. Obtaining fresh, dry hardwood ashes seems important, although the ideal species is not yet clear. Simply monitoring pH and doing titrations with a known acid solution does not allow us to predict a good cook. Increasing the strength of the lye solution by adding more ashes does not usually promise a better cook.(3) Cooking with soda ash (sodium carbonate, 1.3% solution, 6.7% dry fiber in solution) often gives better fiber separation but does not seem to give the luster and warmer, more subtle color tones possible with wood ash lye solutions. Failure to remove any tough areas in the bark during picking after cooking can result in strands in the finished sheet. All of the fiber ready for beating should be tender and easy to pull apart. Soaking and scraping the bark prior to cooking is done routinely at the UICB, to minimize the work necessary during picking. For successful hand beating we find it wise to separate the thicker strands from the thinner during picking and to beat them for 20 minutes or so before adding the main mass of fiber to the beating table. In general, it is best to squeeze most of the excess water from the fiber before beginning beating and then to add water gradually during the final stages of beating, eventually ending up with a very wet mass. Our general practice is to beat for roughly one hour for every 500 grams of dry bark (weight before cooking). Insufficient mixing at the vat can be responsible for strands in the paper during sheetforming but if your vat is equipped with a "maze" (a comb-like agitator) and if you have worked on a given charge of fiber in the vat for five minutes you have probably mixed the stock enough. If strands appear in the vat during papermaking you can drain the vat, strain the fiber, squeeze out excess water and formation aid and do some additional hand beating. Recooking is also a possibility but is likely to leave the finished paper softer, whiter, and weaker--usually not desirable characteristics. What is to be done if you have followed all of these steps carefully and still find yourself with strands in the finished sheet? This, in a nutshell, is where we were about one year ago. There had been a few almost miraculous batches that had worked out splendidly, but for all practical purposes, we were unable to consistently produce the fiber separation we desired. What did we do? The answer was to build a naginata beater. Common in Japanese papermaking shops, the naginata is used following hand or stamper beating. The fiber is placed in the water filled tub, the machine is turned on and the curved arms hack at the water and fiber mixture, causing it to circulate in the tub. Treatment usually takes only 5-15 minutes, and serves to tease apart any remaining strands before the fiber goes into the vat. Naginatas are surprisingly simple and effective machines. Most Japanese naginata tubs are built of cement and are usually immobile once installed. We decided instead to collaborate with the J & J Corporation of Cedar Rapids, Iowa, a local ceramic/fiberglass spray specialty shop. Unlike normal fiberglass sprayed materials, a 1/2 inch thick piece of the proprietary J & J composite material was capable of stopping a bullet from a .357 handgun at point blank range. It seemed to me the material might make a rather durable naginata beater tub. Our first job was to design the machine and build wooden forms on which J & J would spray their material in layers. Then we gave Wenman Machine Works, of Iowa City, the job of cutting, welding, and shaping the stainless steel blades, collars, and shaft, and welding up the aluminum bearing supports. Final assembly was up to us and went well due to careful planning during design. The naginata we built works well. We usually cook 1500 grams of kozo white bark, beat it by hand for a total of two to three hours and then finish it in the naginata for about fifteen minutes. There is only one drawback. The paper has no strands, but it feels different. Instead of being crisp and soft like its entirely handbeaten counterpart, the naginata treated fiber gives a paper with a certain hardness and a duller luster. We are still evaluating the advantages and disadvantages of naginata treated pulp. In the meantime, our plan is to use the naginata only when our work on the other causes of poor fiber separation has yielded another batch of prepared fiber with unacceptable amounts of strands. The end goal is an entirely hand beaten fiber and a paper that is fully traditionally made. Such papers were routinely made during much of the history of the nagashizuki craft, and they can be made again today, with research, experience, and collaborative effort. These few suggestions are offered in the spirit of attaining such goals. Good luck to those inclined in similar papermaking directions.(4) Notes: 1) Donald Farnsworth's A Guide to Japanese Papermaking, (Magnolia Editions, Oakland, 1989), and my Japanese Papermaking: Traditions, Tools, and Techniques, (Weatherhill, Tokyo and New York, 1983). 2) Fibers other than kozo can be used, in Japan and elsewhere, but for the sake of the present discussion we will consider kozo the classic nagashizuki fiber. 3) Our most successful cooks to date have been done with 5000 grams of ash boiled in 30 liters of water for 20 minutes. The resulting solution is left to stand for several hours and then siphoned off to give about 20 liters of sediment-free cooking solution. This solution is used to cook about 1200 grams of kozo. 4) Those interested in building their own naginata beater should consult my book (see note 1 above), pages 104 and 127, and see also figures 98 and 99. Total cost of this prototype was approximately $3000, not including our labor in making the plugs and stand, and in final assembly. Those interested in purchasing a naginata should consult Lee S. McDonald who can quote a price based on the design of this prototype. For operational advice see my book, page 153. Readers are welcome to inspect the prototype and visit our facilities at UICB at any time.
