Shen Gua (or Shen Kuo, 1031–1095) was a civil official in Song China who is known in China today for his multifarious achievements in knowledge domains, such as astronomy and medicine, that fall into the modern category of science. Instead of highlighting those achievements, which have been the subject of a vast body of scholarship on Song history and the history of science, Shen Gua's Empiricism investigates features of Shen's thinking that drove his life-long interest in technological affairs. These features are encapsulated by Zuo Ya in the term “empiricism.” Based on a close textual analysis of Shen's corpus and other important Song thinkers’ writings, Zuo contextualizes Shen's empiricism within his bureaucratic life and the broader theme of eleventh-century Song intellectual history. In doing so, she not only reveals “what empiricism looks like in the Chinese context” (14) but also advances our understanding of the diversity of Chinese thinking.

This book's narrative follows a twofold structure. The first narrative, comprising chapters 1, 3, 5, 6, and 8, delineates Shen's life experience. The second narrative, comprising chapters 2, 4, 7, 9, and 10, analyzes his thinking and compares it with the dominant way of “learning” (xue) in Shen's day. Each chapter in the first narrative corresponds to a particular chapter in the second, with each set of corresponding chapters together constituting a “life and thought” pair. The aim of these complementary elements is to highlight “the key conceptual issue(s) arising from the particular set of experiences” that made up Shen's life and career (19).

Chapter 1 describes Shen's family, childhood, and the first decade (1051–1063) of his statecraft career as a low-ranking local official. Chapter 2 provides the general features of the mainstream Chinese way of learning, a way designed to help educated men pursue the “most profoundness of the reality” (dao), and charts Shen's engagement with observing “things” (wu, referring to both objects and affairs) in his early life. Chapter 3 illustrates Shen's rising bureaucratic career from 1063, when he finally obtained a jinshi degree, to 1075, when he was appointed vice head of state finance. During this period, he held a series of positions in the Northern Song capital, Kaifeng (in Henan province), designing and conducting projects in calendrical studies and hydraulics, and conducting a border survey of Song China. These projects, as Zuo points out in chapter 4, demonstrate Shen's keen interest in observing and recording things.

Chapter 5 describes the political alliance between Shen and the minister and leader of New Policies reform, Wang Anshi, detailing Shen's work in important reform areas between 1071 and 1075. Chapter 6 traces how political conflict over the New Policies era from 1075 to 1085 put Shen's statecraft on a roller coaster. Among the top and lower echelons of officialdom that Shen occupied during this period, for example, he once was the fiscal grand councilor, supervising the state economy but by 1985, following a military disaster in 1082, he had lost his political ranking entirely. Considering that the political conflict between Shen and Wang bore the marks of their intellectual differences regarding the nature of learning, in chapter 7 Zuo discusses Wang's vision of learning. Zuo sees Wang as representative of the mainstream way of Chinese learning in the eleventh century, a way she characterizes with the term “system(s).” The chapter concludes with Shen's departure from Wang's New Policies.

The intellectual differences between “systems” and Shen's way of thinking are further demonstrated in chapters 8, 9, and 10. Chapter 8 surveys the last ten years of Shen's life, during which he wrote Brush Talks from Dream Brook (Mengxi bitan). Analyzing Brush Talks, chapter 9 discusses the features of Shen's nonsystem under the rubric of empiricism. Chapter 10 compares Shen's empiricism with the ways of learning espoused by two other important intellectual leaders, the Cheng brothers (Cheng Hao and Cheng Yi). The Cheng brothers’ way of learning is identified by Zuo as the second stage of the system-building mainstream, a step forward from Wang's first stage. The concluding chapter discusses how three groups of historical actors appreciated Shen Gua and Brush Talks as examples of the general reception of the man and his book after Shen's day. The three groups Zuo identifies are system builders, represented by the leading figure of Neo-Confucianism in the twelfth century, Zhu Xi; reliability seekers, represented by notebook (biji) authors in the Southern Song (1127–1279); and eighteenth-century evidential scholars.

Zuo's study addresses a number of issues from the intellectual history of Northern Song China (960–1127) within an overarching conceptual framework, a comparison between Shen's empiricism and the mainstream systems. It would exceed the scope of this review to recapitulate all the similarities and differences between Shen's empiricism and systems, as the comparison is spread over the five “thought” chapters. The following summary of the comparison focuses on characteristics of Shen's empiricism, introducing some of their crucial differences from systems.

First, Shen's ways of knowing relied heavily on sensory perception as a source of knowledge. Second, Shen's empiricism was not a global doctrine: in some areas (such as astronomical observations) Shen relied largely on sensory perception, whereas in others (such as the Course-qi scheme, yun qi), he used his sensory data to coordinate existing normative orders, which encompassed numerical principles and cosmological orders. In this sense, what separates Shen from other system-builders can be found in cases where Shen “held a more critical stance” when operating with the numerical principles (199) and attributed less significance to the heart-mind (xin) in spontaneously generating knowledge (200, 221). Third, Shen never verbalized his empiricism as a normative proposition. Zuo stresses over the course of the book that we should not understand Shen's empiricism as “a philosophical position defined through normative articulation”; we should instead regard his empiricism as “an empirical stance,” a stance to be discerned via Shen's deeds (17–18).

In contrast, for system builders the ideal way of learning is to model “orders of relations.” System builders like Wang and the Cheng brothers maintained that one could approach the most profound order of reality by advancing through each of the stages by orders of relations. For example, Wang suggested a learner “settle his self,” “enhance his virtue,” and then “apply himself to all under Heaven.” Not always linking deep orders with particular things, Shen seems to feel satisfied in Brush Talks to focus on things as subjects of individual existence, providing no universal unity to connect the miscellaneous variety of the points of knowledge he recorded. As a nonsystem thinker, Shen was more interested in being an evaluator of knowledge. System builders looked for the “epistemic authority” that “prompted one's beliefs with testimonies regarding orders of relations” (186).

In crystalizing these characteristics of Shen's thinking, Zuo's study makes important points about the multiple meanings that knowing based on “seeing and hearing” (jian wen 見聞) contains. Scholars have observed that knowing based on seeing and hearing became remarkably distinctive in literary and historic genres during the Song era, especially in notebooks. Notebooks, a genre of writing that originated before the Song but whose numbers proliferated dramatically during the Song, bear two distinguishing features. First, a notebook might cover a wide range of miscellaneous points of knowledge. Second, a notebook presented knowledge in an item-to-item format without obvious threads to link the recorded facts. Zuo shows that knowing based on seeing and hearing as characterized in notebooks could place particular emphasis on sensory perception, or on “the epistemic autonomy of an individual knowledge seeker who actively exercised critical skills” (17). With the phrase “critical skills,” Zuo refers to an array of evidence-seeking methods, which include meticulous textual studies, finding informants, witnessing, and practice. The application of those critical skills was designed to prove the “reliability” (xin) of the author and his work.

It seems that Zuo's study terms Shen's ways of knowing “empiricism” primarily in virtue of his reliance on sensory perception as a guiding source of knowledge. A similar reliance can also be found, however, in many other Song writers, such as Su Shi and Zhang Lei. What separates Shen from those authors are their distinctive stands on a wide range of issues concerning learning, such as the degree to which they linked their sensory data to orders of relations. To be sure, the similarities and differences are acknowledged and clearly shown in Zuo's discussion, yet “empiricism” in the book nevertheless seems to serve as shorthand for any one of a person's many epistemic dimensions, and in that sense is interchangeable with other phrases such as “empirical attitudes” or “empirical approaches” to learning.

Shen Gua's “empiricism” thus constitutes only one of numerous versions of educated men's “empiricism” in the Song. It would be interesting to know more about how Shen's version of empiricism “came into being under broad historical and conceptual circumstances.” The ten chapters of the life–thought pairs that comprise Zuo's book do not fully explain such a formation process; instead, they provide something more like an anatomy of how Shen's writings, technological projects, and political actions revealed a constant empirical attitude over his forty-year bureaucratic life. Beyond Shen's disagreement with systems as a way of learning, a more detailed discussion of other possible inspirations for Shen's empirical stance, such as the books he had read, would have been useful.

Another point worth further developing concerns the two-phase development of notebooks, the periodization that Zuo proposes when comparing Brush Talks and other Song notebooks. During the first phase, which ranged from the late tenth century through the late eleventh century, many notebook writers often framed their reliability-seeking as a means rather than an end. For example, “a dominant rhetoric for evaluating the credibility of a biji was to compare it to official histories,” and many notebook authors “pursued accuracy as a meritorious practice well established in the historiographical tradition” (193). From the twelfth century onward, which constitutes the second stage, notebook authors began to value credibility and critical skills for their own sake. Those authors further applied knowing based on seeing and hearing to subjects that previously were “mainly associated with modeling.”

The two-stage shift sheds new light on the history of biji writing and the pursuit of reliability in Song China by demonstrating a new and useful analytic approach to differentiating contexts in which the issue of reliability appeared in notebooks. However, some late Tang writings that shared the two abovementioned characteristics of notebooks and were printed in the Song, such as Records of the Land of Northward Facing Doors (Beihu lu 北戶錄, late ninth century) and its preface, already fit Zuo's description of the second stage: “when authors and readers discussed the credibility of biji, they no longer exclusively referred to historical composition as the methodological horizon; instead, they started to address a broader reliability based on a range of critical skills” (229). If we could take those late Tang writings into account, we might not only produce a more nuanced history of biji writing and the pursuit of reliability-seeking in middle-period China, but we might also reveal a solid precedent for the Song reliability seekers, including Shen.

In sum, Zuo's study offers lucid analytic language and concepts with which to examine Chinese versions of empirical attitudes and yields a new angle on the diversity of Song thinking through the lens of sensory data and, by extension, empirical evidence. The book contributes impressively to our understanding of intellectual history and the history of science in Northern Song China and will be of interest to scholars working in these two fields as well as on the development of empiricism in the premodern world.