Taste may be the most important thing in the laws of cuisine, but taste is not restricted to just eating with the mouth.
—Shsekiken Sken, Ryri mmoku chmish (Collected Writings on Cuisine and an Outline on Seasonings, 1730)
The mid-seventeenth century saw the rise of a new form of culinary writing, the printed culinary book (ryribon). Hchnin, the knife specialists serving high-ranking samurai and aristocrats, continued to write culinary texts (ryrisho) in the Edo period, but as in previous centuries these works were privately disseminated manuscripts not widely read and, before the twentieth century, seldom published.1 This makes the published culinary books, beginning with the 1643 work Tales of Cookery (Ryri monogatari), the first popular media for information about cooking and dining. Before I examine these works closely in chapters 6 and 7, this prefatory chapter offers some general comments on what culinary books reveal about the interplay between food and fantasy that constitutes the definition of cuisine in early modern Japan. Culinary books popularized ideal forms of cooking and dining premised on the idea that these could be enjoyed as forms of vicarious pleasure in the same way that popular literature and guidebooks offered readers descriptions of exotic locations they might never travel to in real life.
The rise of the publishing business has been called the most important cultural development in early modern Japan, because the culture developed inseparably from books and print.2 Printed books facilitated the spread of old knowledge previously confined to manuscripts, as well as new forms of systematized data. Published guidebooks and exhaustive catalogues of information in fields ranging from medicine to rankings of courtesans to street names contributed to the growth of what historian Mary Elizabeth Berry has termed the “library of public information,” a body of knowledge shared by a literate readership.3 Though the rate of literacy for this period is impossible to know, it was certainly higher in urban areas and among men, and according to Richard Rubinger, literacy “may well have reached most, if not all, of the taxpaying landowner class of peasants by the end of the seventeenth century,” who comprised approximately half of the rural population.4
The growth of printing in the Edo period relied less on technological developments than it did on demographic, political, and economic ones. Rather than depend on the newly imported technology of moveable type, which proved too costly, Japanese publishers in the seventeenth century used the well-established media of woodblocks for book printing.5 Yet publishing would not have thrived without the rise of an urban population able to support it financially, a population itself a function of the peace and social stability brought by Tokugawa rule. The policies of the Tokugawa bakufu gave impetus to the growth of cities in edicts that compelled samurai to be stationed in castle towns and cities, and daimyo all over Japan to spend alternate years with their troops in Edo. By the mid-eighteenth century, the populations of Kyoto and Osaka exceeded more than three hundred thousand, and Edo’s population reached nearly a million, making it the largest city in the world. Book printing and sales, usually undertaken by the same establishments, were indicative of the diversified economic activity occurring in these cities and spoke to the growth of urban consumer markets. As early as 1640 there were a hundred bookstores in Kyoto, and as many as 506 different publishers operated at some time in that city during the Tokugawa period.6 According to one estimate, Japan’s bookshops published an average of three thousand books a year.7 The cost of a book could be the equivalent of as little as a few meals for inexpensive storybooks or as much as more than a month of meals for more technical works.8 But customers who could not afford to purchase books could rent them. By one count there were 656 rental libraries in Edo by 1808, and 800 by 1830.9
Culinary books represented a small but important sector of the publishing industry. Some bookstores even specialized in these.10 The exact number of culinary books published in the Edo period is unknown, but scholar Harada Nobuo has surveyed 182 books that can be conclusively dated to sometime in the Tokugawa era. He found that most of these works were published in the late eighteenth and early nineteenth centuries, particularly between 1800 and 1830.11 The main types of culinary books were collections of recipes (chrih) and collections of menus (kondatesh), although some works included both.12
Some culinary books, especially those published in the first 150 years of the Edo period, popularized elite styles of banqueting. Secret Writings on Culinary Slicing (Ryri kirikata hidensh, 1659), a collection of h-chnin treatises introduced in chapter 3, provided lore about the “three formal rounds of drinking” (shikisankon) that preceded elite banquets, and it included dozens of diagrams for cutting fish and fowl in knife ceremonies (shikibch). Shij Takashima (n.d.) borrowed liberally from this work and expanded on its contents to publish the Complete Manual of Cuisine of Our School (Try setsuy ryri taizen, 1714), a text that also incorporated parts of End Genkan’s (n.d.) Guide to Meals for the Tea Ceremony (Cha no yu kondate shinan, 1696), among other works. Genkan’s book detailed the proper ways to entertain powerful warlords like the shogun. All these works contributed to the diffusion of elite customs like shikisankon, which became part of commoner weddings and other celebrations in the Edo period.13 A few commoners, particularly chefs (itamae) who worked in the growing number of restaurants found in the major cities, even chose to study the art of ceremonial cutting, although the exact number who did is unknown.14
At the same time that culinary books allowed commoners to learn about elite cuisine, the military government restricted them from adopting it. The Tokugawa shoguns continued policies of social control set by Toyotomi Hideyoshi in the last decades of the sixteenth century that sought to differentiate the ruling samurai, roughly 7 percent of the population, from the ruled classes. The Tokugawa bakufu subdivided commoners into occupational groupings of peasants, artisans, and merchants. These groups reflected geographic differences, as peasants, the majority of the population (approximately 85 percent), lived in rural areas, while artisans and merchants, described collectively as townsmen (chnin), lived with samurai in castle towns and cities. The Tokugawa warrior government took as its moral prerogative the goal of reinforcing the distinctions between these groups through sumptuary laws (ken’yakurei) to ensure an elevated place for the samurai. It also strove to ensure that peasants focused their energies on agricultural labor and tribute payment, and that townspeople lived in a manner appropriate to their lesser station. The earliest sumptuary laws began with the first shogun, Tokugawa Ieyasu, but a torrent of these appeared during the reign of the fifth shogun, Tokugawa Tsunayoshi (1646–1709). These coincided with the rapid economic growth and blossoming of urban culture in the late seventeenth century. Sumptuary laws sought to discipline this culture by ordering life in minute detail. For commoners, they specified proper clothing, the size and locations of homes, the scale of celebrations like weddings and funerals, and objects of daily life. “There were laws against expensive tobacco pouches, purses, incense containers, lacquer sake-cup stands, cake boxes, and ornamental decorations in the homes at festival times.”15
Food and dining became areas of intense sumptuary regulation as well. Commoners could not legally consume many of the game birds used often in elite banquets, including wild goose, wild duck, crane, and swan.16 The bakufu also prohibited commoners from serving special recipes restricted to samurai, such as a recipe for artfully peeled chestnuts soaked in salt water called mizuguri or mitsuguri. The bakufu limited the number of trays and dishes commoners could serve at banquets. According to a bakufu edict issued in 1668, commoners could have only two trays of food, with two soups and five side dishes total.17 When farmers began charging more for early produce (hashiri), and commoners vied with each other to buy it, believing that these foods promoted longevity, the bakufu regulated the seasonality of foodstuffs, fixing the dates when certain foods could be sold.18 For example, it stipulated that certain varieties of persimmons could be sold and purchased only after the ninth month of the year.19 It was the peasants, who actually produced these foods, who faced the severest limitations on their diets. In 1642 a bakufu edict restricted polished rice to tribute payments, compelling peasants to eat brown rice and other grains like barley, millet, and barnyard millet. Peasants could not make or purchase wheat noodles and tofu, and neither could they brew alcohol or travel to towns and cities to buy it.20 The bakufu reiterated these prohibitions again in 1649.21 Samurai did not escape from sumptuary legislation and had to live according to their rank, which ranged from lowly foot soldier to powerful daimyo. Certain foods were reserved for daimyo, such as dried sea cucumber (kushiko) and skewered abalone (kushi awabi).22 Top advisors to the shogun on the council of elders (rj) were allowed three trays of food with ten side dishes at certain celebratory banquets. Daimyo in charge of their own provinces (kuni mochi) could have only two soups and seven side dishes at the same events.23 This amount was scarcely more than a 1624 rule permitted to middle-ranking warrior vassals (hatamoto) of the Tokugawa. The bakufu allowed hatamoto two soups and five side dishes, and it limited the amount of sake they could consume while entertaining. A law a century later raised this limit to one extra side dish for special occasions, but stipulated that, for ordinary meals, hatamoto should dine on no more than one soup and four side dishes.24 This was little different from what most commoners ate.
The fact that the contents of culinary books did not conform to these rulings raises interesting questions about their intended audience and purpose. We might expect books such as Guide to Meals for the Tea Ceremony, which focuses on banquets for high feudal officials, to be aimed at a select readership, but the first book of menus—Collection of Cooking Menus (Ryri kondatesh), published in 1671—also includes menus for three-tray banquets, despite the fact that only senior members of the bakufu and aristocrats could legally partake in these. The cost of these meals, too, made them prohibitive to create. Besides the two dozen different varieties of fish and seafood and other delicacies required to make one menu detailed in End Genkan’s Guide to Meals for the Tea Ceremony, which is described in detail in the next chapter, the author stipulated a cooking staff of forty-one people, which would have been difficult for all but the wealthiest members of the military or aristocratic elite to assemble. Later publications may not provide as much detail as End Genkan’s books, but they have complicated menus of three or more trays of food, as exemplified by the works Collected Writings on Cuisine and an Outline on Seasonings (Ryri mmoku chmish), published in 1730, and Four Seasons of Menus (Shiki ryri kondate), published after 1750. Besides published collections of menus, books of recipes, too, contained directions for foods that commoners were prohibited from eating. Recipes for crane soup, for example, appear in Tales of Cookery, Edo Cookery Collection (Edo ryrish, 1674), and Guide to Meals for the Tea Ceremony.25 Despite the fact that dried sea cucumber (kushiko) was a dish restricted to daimyo, directions for preparing and serving it appear in Threading Together the Sages of Verse (Kasen no kumi’ito), published in 1748, Four Seasons of Menus, and ABC’s of Cooking (Ryri iroha bch, 1773), among other works.26 Collected Writings on Cuisine and an Outline on Seasonings provides a list of ways to consume all the prohibited game birds, including goose, duck, swan, and crane. According to this book, swan is best served raw with a miso dressing, while certain varieties of goose are tough and to be avoided.27
The publication runs of these culinary books far exceeded the small number of people who could legally consume or afford the foods and meals described in them. At any given time, there were only three to five senior councilors to the shogun who had the right to sit down to a banquet of three trays of food. Even if we were to add to their number the shogun and members of his household, as well as a number of aristocrats in Kyoto, the total would have been a fraction of the readership of culinary books, whose print runs are estimated at three hundred to five hundred to a thousand copies, with the most popular books reprinted multiple times.28 It seems doubtful that even one culinary book would be written for the small audience of people who could legally prepare and eat these foods, especially since this same group did not cook for itself and instead employed a staff of hchnin, who had their own reference works in the form of privately circulating culinary writings. In short, in the Edo period, culinary books made it possible for hundreds, if not thousands, of times more people to read about certain forms of dining and about dishes they could not legally consume or afford to assemble.
What we know about the authorship of culinary books also suggests a diversified audience beyond the military elite or a few culinary professionals.29 One of the foremost experts on culinary books, Kawakami Kzo, has researched the authorship of fifty-nine Edo-period culinary books whose authorship is known. According to Kawakami’s research, most authors writing in the seventeenth century came from the ranks of hchnin, but hchnin numbers declined and such authors became a minority in the eighteenth and nineteenth century. In their stead, restaurant chefs (itamae), tea masters, connoisseurs, medical doctors, social critics, and amateur cooks were authoring culinary books.30 Although we do not know much about these authors beyond their occupations as can be deduced from their writings, most appear to be either commoners or lower-level samurai—in other words, people who were prevented by sumptuary legislation from having direct exposure to the most elite forms of cuisine.31
Of course, commoners may have eaten prohibited foods in secret. The author of Threading Together the Sages of Verse admitted that it was “rare” (but not illegal) for commoners to consume meals consisting of three soups and ten side dishes, and so he avoided writing about them. But he also demonstrated to his readers how to bypass sumptuary rulings against having more than two trays. Rather than calling for a third tray and naming it that, his menus featured two trays and “an additional serving tray” (hikide). Like the two other trays, the hikide also contained a soup and side dishes, but technically it was not a third tray, since it was served separately from the other two trays and the soup was designated by a different term.32
However, if we compare the menus and foods depicted in culinary books with what we know of the diet in the Edo period, we might wonder whether commoners and most samurai could assemble a meal of even two trays except on rare celebratory occasions or a special trip to a restaurant. Simply stated, the diet for most people throughout the Edo period was dull. Writing of life for her ancestors who had been lower-level samurai in the Mito domain at the end of the Edo period, Yamakawa Kikue (1890–1980) states, “The diet of people in Mito was quite monotonous. They did not eat meat, of course, and only what fish can be gotten locally.”33 Culinary historian Katarzyna Cwiertka, while noting that people in urban areas after the seventeenth century enjoyed a more “differentiated gastronomy,” concurs that meals consumed in most homes, even among the wealthy, were “austere.” She writes, “Compared to what Japanese consume on a daily basis today, the number of side dishes served with rice and soup was much smaller and their variety very limited.”34 Culinary historian Ishige Naomichi agrees: “The meals of the commoners in the cities . . . were plain. Most Edoites had a breakfast of rice, miso soup and pickles, and their lunch and dinner were similar, with the addition of one side dish of simmered vegetables or tofu, or simmered or grilled fish.”35 This was just enough food for one tray, or one “box table” (hakozen) used by commoners that served both as an eating platform and as a storage container for utensils when not in use—hardly the three or even two trays of food described in most cookbooks.
The evidence we have examined thus far points to a disparity between the contents of Edo-period culinary books and the daily diet and experience of the authors and readers of these works. We should not be surprised by this. Priscilla Parkhurst Ferguson reminds us that “cookbooks legislate rather than document.”36 In other words, cookbooks and other types of culinary writings provide directions for preparing and consuming foods under ideal circumstances. They begin with the assumption that all the ingredients are available and affordable, and that these can be easily assembled and correctly served. Cookbook authors offer a potential use for food, but this does not ensure that these guidelines have been or ever will be followed and the dishes themselves created or the banquets served and consumed. A historian of Japanese cookbooks, Shoko Higashiyotsuyanagi, confirms this view concerning cookbooks in the early modern period: “For most of the Edo period, cookbooks were intended for a male audience and were more voyeuristic than practical. They were not intended for home cooking; rather, they described the preparation of formal sumptuous banquets and served the purposes of hobby reading. All authors and editors were male, and in cases where the books did provide practical information, the target audiences clearly were male professional cooks and dilettantes.”37
Consequently, one of the tasks for historians of Japanese cuisine is to try to judge how much the contents of cookbooks reflected the actual cooking practices of the work’s authors and readers. Culinary historian Harada Nobuo, who has extensively studied Edo-period culinary books, has documented what he sees as a historical shift in Edo-period culinary books that occurred in the mid-eighteenth century, when writings began to favor amusement (asobi) over discussions of the technicalities of food preparation.38 Harada indicates that, by the 1750s, cooking methods had reached a plateau, and authors shifted to writing more imaginative works that could simply be read for pleasure.39 “They regale readers with entertaining reading on culinary subjects, but provide little or nothing of significance on technical aspects of food preparation.” By the end of the Edo period, Harada concludes, Japanese cuisine had grown “stagnant.”40
Harada finds playfulness to be characteristic of culinary books published after the mid-eighteenth century, but when we recall that many of the works published nearly a century earlier, such as Collection of Cooking Menus (1671) and Edo Cookery Collection (1674), contained menus and recipes that nearly all readers of these works could not legally create or consume, it becomes harder to draw a line between works offering a serious technical discussion of cooking and ones meant for amusement. In that light, readers in the Edo period may have recognized Harada’s distinction between “amusing” and “serious” culinary books only if they tried to create the recipes in either one. But cookbooks, especially those focusing on foods served in formal settings, may not have been written with that sole purpose in mind. It was not until the Meiji period that most cookbooks began to focus on domestic cooking, describing what people actually prepared and ate.41
Writing and reading cookbooks in the Edo period did not depend on actually cooking and eating foods. All Edo-period culinary books could be read and enjoyed without having to create the dishes within them, in the same way that foods in the banquets of the elite were appreciated for their artistic qualities and symbolic meanings without being eaten. In these elite banquets, the inability to eat certain foods provoked reflection on their symbolic and artistic qualities. The inability to prepare or eat foods served the same function: one read about and imagined tasting crane soup or attending a banquet fit for a shogun, with the fantasy contingent upon the absence of the real thing. Indeed, the elite fortunate enough to actually attend these events and eat these foods would probably not have desired these vicarious experiences, which provides another indication of the wider audience for culinary books.
Whether or not cooking methods grew stagnant by the end of the Edo period, it remains true that the large number of published culinary writings allowed readers to become familiar with ways to think about food as cuisine—that is, as an intellectual and artistic practice. This accustomed Japanese to the use of foods to signify ideas, a cognitive foundation equally as important as any method of cooking to the later development of Japanese cuisine in the Meiji period.42
In that regard, the greatest contribution made by Edo-period culinary books was to accustom commoners to fantasizing about foods even if they could not eat them. Fortunately, this was not a difficult leap for readers, because the culture of print had introduced them to other experiences they might never have firsthand: the enjoyment of journeys through landscapes depicted in woodblock prints and travel literature; or visits to the pleasure quarters or meetings with famous actors backstage, as described in popular literature and in actors’ critiques. Imagining the taste of kushiko and visualizing the many dishes on a three-tray meal were easy to do once one had already made similar mental journeys. And the fact that these pleasures were illegal and relatively unaffordable made them all the more attractive.