History of Specialty Fermented Soyfoods

A Special Report on The History of Traditional Fermented Soyfoods

A Chapter from the Unpublished Manuscript, History of Soybeans and Soyfoods: 1100 B.C. to the 1980s

by William Shurtleff and Akiko Aoyagi

Copyright 2007 Soyinfo Center, Lafayette, California

This chapter contains brief discussions of the histories of a number of little known and less important fermented soyfoods.

Soy Wine . The earliest extant reference to soy wine appears in a Chinese work by Wang Hsi-chih titled "The Record of Soy Nugget [shih] Wine" and writte in about AD 346-379. He noted "When I was young I drank shih chiu [soy nugget wine]. It was very good." The next existing reference appears in the Ch'i-min yao-shu , the world's first encyclopedia of agriculture, written by Chia Ssu-hsieh in about AD 535. This work quotes from an even earlier nonextant work, the Shih ching ( The Classic of Food ; date and authorship unknown, but it may predate Wang Hsi-chih), giving the Shih Ching's recipe for making "one thousand year bitter soy wine." This was apparently made by soaking soy nuggets in a grain-based alcoholic beverage or medicinal tincture.

The liquid left after boiling soybeans may also be used to make soy wine. Shinoda (1977) states that such a wine was developed during the Sung dynasty (960-1279); the liquid from cooking black or dark soybeans was mixed with distilled liquor and sugar; it may have been fermented. The Sung dynasty work T'u-ching yen-i pen-ts'ao (Sung dynasty materia medica: 1115 A.D.) by Kou Tsung-shih gives a soybean wine preparation to be taken after childbirth. Selected black soybeans are washed and soaked in wine. The beans are then fried (until all smoke is gone) and dipped in wine. When the wine has turned a reddish-purple color, the beans are removed and the wine is consumed according to one's needs. China's famous materia medica, the Pen-ts'ao kang-mu (1578-97) by Li Shi-chen referred to a soy wine called tou-lin chiu ("bean soak wine") which it described as a sake-like fermented alcoholic beverage made from black soybeans. A recipe was given in the section on soybeans and it was stated that the Ts'ung-shu chi-ch'eng ch'u-p'ien (1473) said that it cures post-partum white sickness.

During his trip to China, William Morse referred to "soy wine." In 1931 he stated that the liquid left after boiling soybeans was drawn off after 1 hour, mixed with citric acid and sugar, then fermented (Dorsett and Morse 1928-31). In Vietnam a distilled soy alcohol called rou dau nanh was traditionally made in both the north and central parts of the country prior to the recent wars. The fermentation may have included both soybeans and rice (Hoang 1981, personal communication). In 1976 Vivien Yee wrote her PhD thesis at Cornell University on new products from soybeans, including two types of soy wines; one was made from soymilk and one from the whey left over after making tofu. Some 25% (w/v, weight-to-volume) of sucrose was added to the soymilk or tofu whey and the liquid was fermented anaerobically with a wine yeast, until the alcohol concentration reached 12% v/v or higher. With soymilk, the proteins precipitated out and had to be removed. Otherwise clarification and aging were the same as for any wine; the finished products were similar to sake.

Cantonese Wine Starter (kiu-tsee) . The first and only known reference to this product appeared in 1878 in an article by M.P. Dabry de Thiersant in the Bulletin of the Society for Acclimatization (Paillieux 1880). The French transcription kiu-tsee probably referred to a Chinese term such as chiu-chi or chiu-chieh ; chiu means "wine" or "alcoholic beverage" in Mandarin. The starter, used in Canton to make a grain-based wine and spirits, was made from 75 lb of rice, 27 lb of soybeans, 14 lb of the pulverized leaves of the Chinese glycosmis ( Glycosmis citrifolia , called Chan-kiue in Cantonese; a shrub or small tree that grows mostly in Kwantung province), and 4 ounces of kiu-tsee from a previous fermentation. To prepare: boil the rice and beans, spread the rice to cool on a large table, then sprinkle the well-cooked soybeans, dried and powdered leaves, and pulverized starter over the surface. Mix in a shallow vat, mash underfoot to a paste, then shape into 1-lb bricks in a mold 6 inches long and 1 1/2 inches deep. Press grains of dry rice into the surfaces, arrange vertically on a board, and allow to mold for 4-5 days until covered with a whitish mycelium. Place on nets and allow to dry in the shade for 4-8 days, then in the sun on screens for 2-3 days. Store in a desiccator for use as wine inoculum.

Soy Fermentation Pellicle (tou-huang) . The first known description of this product was published by Stuart in 1911 in his Chinese materia medica. He described it as "bean ferment" (the characters mean "bean yellow"), or the fermentation pellicle (mycoderma) which formed on the top of fermenting beans, and probably contained various yeasts and molds. To make the product, "Take a peck of black beans and thoroughly steam them. Spread upon matting and cover with artemesia stalks, as in the process of preparing soy (sauce). When the pellicle is formed on top, take it off, dry in the sun and powder, when it is ready for use. The taste is sweet and cooling." It was recommended for the treatment of many ailments, including rheumatism and eczema. Shih (1918) repeated Stuart's description.

Meitauza (Mucor-fermented okara) . The earliest known description of this product was given by Y.K. Shih in 1937. The product was traditionally made in Wuchang, preferably in winter, as follows. Form okara into round cakes 4 1/2 to 5 inches in diameter, 1 inch thick at the center and 3/8 to 1/2 inch thick at the edges. Place the cakes in a vessel of the same size and allow to ferment in a room with moderate aeration until, after 10-15 days they are covered with a white mycelium of Mucor meitauza Shih. Dry in the sun for several hours, then sell. Meitauza is served either fried or cooked with vegetables, and is widely considered to be tasty and nutritious. In his article, Shih (working with Dr. Jun Hanzawa in Hokkaido, Japan) isolated the Mucor mold, studied it and described it as a new species. Hesseltine (1965), noted that the mold was probably a synonym of Actinomucor elegans , already noted as one of the principal fungi involved in the preparation of fermented tofu. This product must be a close relative of okara tempeh. See also Chapter 29, Okara.

Okara Onchom . This traditional Indonesian fermented food (now spelled oncom in Indonesia and formerly called ontjom ) is a close relative of okara tempeh, except that the fermentation is done with a Neurospora (rather than a Rhizopus ) mold, which envelopes the cakes with a brilliant orange mycelium. Made and consumed for centuries, but only in West Java and especially in Bogor, the food is known there as onchom merah , onchom bereum , or onchom tahu . A lengthy description and bibliography is given in The Book of Tempeh by Shurtleff and Aoyagi (1979). Early?? accounts are given by Went (1901 Ref??), Ochse (1931), van Veen (1968), and Ho (1976 Ref??). If early reference, move up this category. Nishiwaki 1924?? See Greenberg 1984.

Soy Onchom . Surprisingly a tempeh-like product made with soybeans but inoculated with the onchom mold Neurospora was never traditionally made in Indonesia. The soybeans were used only to make tempeh, and the onchom was always made from peanut presscake or okara. In 1965 Steinkraus, Lee and Buck developed and described an acceptable soy onchom, which resembled tempeh except that the flavor was more nutlike. Liem et al. (1977) noted that traditional onchom is a good source of vitamin B-12.

Soyidli, Dosa (or Dosai), and Dhokla . These traditional Indian fermented foods typically made with ground rice and legumes, began to be studied seriously by Western microbiologists such as Steinkraus and van Veen during the late 1960s. It was soon realized that the nutritional value of the foods could be greatly enhanced by using soybeans instead of the traditional low-protein legumes. Idli is a steamed bread or buns typically prepared from a fermented batter of rice and blackgram dal ( Vigna mungo ). Idli is unique and important in that it is a sourdough bread made without the use of wheat. Dosa or dosai (called thosai in Sri Lanka) is made from a batter of the same fermented ingredients, but they are fried to form a pancake. Dhokla is a steamed bread like idli but the batter contains rice and Bengal gram (garbanzo beans; chick-peas). According to Ramakrishnan (1979) idli and dosa have been used as basic foods in South India since at least AD 1100. Dhokla is from west India, especially Gujarat. The dominant organisms in idli are Leuconostoc mesenteroides and a number of Lactobacillus species. The first work on using soybeans in any of these Indian fermented foods was done by Steinkraus, van Veen, and Thiebeau in 1967. They substituted soybeans for all of the black gram in idli using proportions of 1:2 or 1:3 (soybeans:rice) to yield satisfactory products with higher protein content than their traditional counterparts. This research was continued by Ramakrishnan and co-workers in 1976 and expanded in a 1979 report to include dosa and dhokla; the research was supported by PL 480 funds. In 1977 Akolkar in Baroda, India wrote his PhD dissertation on soyidli fermentations. It is not known if these soy-containing fermented foods are yet being used in India. Similar traditional batter fermented foods (iddli??, thosai and hoppers) are also widely consumed in Sri Lanka, and could incorporate soy.

Soy-Ogi . Ogi is a fermented sour corn pap or porridge widely used as a breakfast food and weaning food among more than 10 million Yorubas in the western states of Nigeria. Akinrele (1966) and Oke (1967) showed that there were considerable losses of nutrients (especially protein and calcium) during ogi processing. Akinrele (1966) found that the biological quality of the protein in ogi was so poor that it did not support growth in rats, but when it was fortified with 30% heat-treated whole soy flour, the protein efficiency ratio (PER) increased threefold, making the protein usability almost equal to that of casein. Oke (1967) also recommended fortification with soy flour. Prior to 1969, Cadbury's (Nigeria) Ltd. did an acceptance study of soy-ogi based on 600 consumer tests and found that except for its container, it was preferred to the other products compared: a commercial dried cow's milk (Lactogen), a filled milk preparation (SMA), and plain ogi. In 1970 the Nigerian Federal Ministry of Industries published a detailed report on soy-ogi, written by Akinrele and co-workers (Ref??). They cited the palatability study noted above and stated that soy-ogi (developed at the Federal Institute of Industrial Research, Oshodi) could be produced at one-third the cost of the branded infants foods commercially available in Nigeria. In 1971 Akinrele and Edwards reported on the high protein quantity (20.3%) and quality of soy-ogi, and its relatively low cost. It is not known if soy-ogi has ever been commercialized in Nigeria and if not, why not??

Sere or Seredele . This semifermented Balinese soyfood is usually classified as nonfermented. Therefore it is discussed in Chapter 23, Whole

Dry Soybeans.