For anyone who had occasion to read—or to reread—Okakura Kakuzō's The Book of Tea in 2005, the centennial anniversary of Albert Einstein's annus mirabilis, it was difficult not to be struck by Okakura's contention, occurring a third of the way through the text, that “Zennism, like Taoism, is the worship of Relativity” (Okakura Kakuzō, The Book of Tea [New York: Fox Duffield, 1906], p. 65). Okakura was, of course, writing in English in 1906, shortly after the still-obscure Einstein published his first account of the special theory of relativity in Annalen der Physik. Okakura almost certainly had no inkling of the revolution that was under way in the scientific community, but his book is an odd testament to how an observing subject's location determines his view of the world. For Okakura, “relativity” meant that “nothing is real except that which concerns the working of our own minds”; for Einstein, the implications were rather different: that there was a reality dependent on the observer, but no less real for that.

This disjuncture between Okakura's metaphysical conception of relativity and Einstein's description of an observer-dependent reality lies at the heart of Gregory Golley's study of Japanese literary modernism, When Our Eyes No Longer See. Part of Golley's aim is to chart how a new conception of reality, based on the discoveries of early twentieth-century physics, shaped the way that Japanese writers approached the external world during the moment of high modernism. Reality came to be redefined as exceeding human perceptive capacities, and writers in turn embraced “the language of non-mimetic referentiality” (p. 7) to describe a reality they could no longer see. For Golley, this untold story, “a singular and anomalous moment of perceptual and referential expansion” (p. 3), leads to a reexamination of how we understand broadly used (though often ill-defined) literary historical concepts such as realism and modernism. Golley argues that an overly narrow conception of what constitutes realism (based on classical notions of mimeticism) has led to a false dichotomy between the modes of realism and modernism in Japanese literary history and that “a species of realism that could be called dialectical” (p. 4) in fact served as a unifying stance for many of the texts of high modernism.

Golley's book is eloquent and erudite, offering subtle critiques of our understanding of the literary history of Japan in the 1920s and 1930s through both a fine-grained historical account of the discourses of the “new scientific realism” (p. 25) in prewar Japan and through a series of rereadings of some of the major figures of the interwar period: Tanizaki Jun'ichirō (chapter 1), Yokomitsu Riichi (chapter 2), and Miyazawa Kenji (chapters 3–6). What unifies these readings—and what makes them often so provocative—is Golley's argument that these works of high literary modernism need to be read as fictions of “realism” even as his study urges a redefinition of that category. “To be sure,” Golley writes, “the works that will be examined in this book all stand emphatically at odds with the thematic and stylistic conventions of literary realism; but none of these works opposes realism as an ontological or epistemological disposition” (p. 17). There is a strangely defamiliarizing effect to Golley's argument, in which what we thought we understood to be works antithetical to the dominant realism of naturalist fiction become exemplars of a new kind of realism based on a new apprehension of what constitutes reality.

Central to Golley's study is the question of the relationship between language and reality—and of the adequacy of language to apprehend and describe reality—and it is a point on which Golley reflects at length in his introduction. In a rereading of Natsume Sōseki's “Gendai Nihon no kaika” (1911), Golley highlights Sōseki's own skeptical take on the ability of any non-ostensive definition adequately to describe reality, arguing that “Sōseki's essay exemplifies the moribund posture of a positivistic skepticism driven by a quest for empirical truth that was reaching the limits of its own logic” (p. 14). Yet for the modern reader, encountering Sōseki's lecture almost a century later, it is hard not to see in Sōseki's own discomfort over the way in which abstract definitions “freeze” or “rigidify” their object something of the approach to language that would come to underlie Wittgenstein's linguistic thought (and, in turn, the historicist impulse in the humanities in recent decades), in which the “uses” of words in their everyday context becomes the central problem. “If I want to draw sharp boundaries in the area of (such) blurred language use,” Wittgenstein writes, “then the sharply demarcated areas will relate to real language use like the contours of a pen-and-ink drawing to the gradual transitions of colour patches in the reality that has been sketched” (Wittgenstein, The Big Typescript: TS 213, ed. and trans. C. Grant Luckhardt and Maximilian A. E. Aue [Malden, Mass.: Blackwell, 2005], 56).

Golley's own approach to the problem of definitions seems to push in almost the opposite direction from Sōseki's and from the historicist approach to language. Even as Sōseki's “Gendai Nihon no kaika” opens with a lengthy digression on the problems of defining words, Golley's own study opens with a bold definition: “There is a way of knowing the world that reaches beyond the limits of the human body. I will call this mode of perception realism” (p. 1). In its search for a concrete definition of “realism”—and for an ontological basis for realism—When Our Eyes No Longer See embodies a critical impulse to go beyond an apprehension of language as a conventional, and thus historical, system to offer a corrective definition of “realism” that transcends the historical uses of the word (even in the period under discussion). This break with what is often called the linguistic turn in the historical sciences suggests an interesting homology at the heart of Golley's study in which Golley himself embodies the kind of ontological stance he finds central to the new scientific realism of the twentieth century: a reality that exists beyond the linguistic structures through which it is comprehended.

This radical stance toward meaning pays dividends by forcing literary historians once again to confront the very conventional ways in which definitions of terms such as “realism” have become sedimented and the often uncritical and overly schematic deployment of these terms in charting literary history. And yet even as Golley rejects Sōseki's own apprehensions over the possibility of achieving any fixed definition that entirely dispenses with the question of “conventionality” (yakusokujō), the search for a fixed definition of realism (and fixed in both senses of stable and repaired) raises again the vexed question of what role transhistorical concepts can play in any historical discipline, literary history included. To what extent, that is, can or should literary history be concerned with analyzing and understanding the discursive deployment of a term such as “realism” historically (and at the multiple moments of its enunciation), and to what extent should literary criticism seek nonhistorical categories that can be used to comprehend literary texts from historical moments that used such categories in very different ways? In the context of ongoing debates about the linguistic turn and its aftermath, When Our Eyes No Longer See can be read as much as a reflection on literary critical method, as much as a sustained mediation on the Japanese modernist moment.