In lieu of an abstract, here is a brief excerpt of the content:

My conversations with peasant-bureaucrats invariably revolved around numbers . They either cited them from memory or presented me with reports containing tables, graphs, charts, and so on as evidence that governing documents , although vaguely worded, do produce quantifiable changes in the countryside. In fact, when I asked whether there was any aspect of their work that could not be conveyed in numerical form, nearly all of them said no. This answer often prompted me to press further and inquire whether official statistics helped create the very realities they claim to document. Mr. Dũng, who had served for many years as an information officer within the Ministry of Agriculture in Bắc Kạn Province, was among the many people who rejected my suggestion. I first met Dũng in late 2001 as part of a team hired to assess the impact a Swiss-funded development project had on food security and environmental sustainability for upland groups living on the periphery of a national park. I spoke with Dũng at length on several occasions during the three-week mission. But whenever the team asked probing questions about the numbers government offices provided to them, he would paraphrase what I later learned was Lenin’s view on the matter: “Statistics are among the most powerful means of acquiring social knowledge.” Dũng’s uncharacteristically formal response sidestepped the question and did nothing to resolve doubts the team members had about the numbers; it did, however, end any further explicit discussion of their factual accuracy. Dũng’s position was not unusual. His colleagues similarly maintained that official statistics are “scientific,” “objective,” and therefore “Marxist-Leninist” in nature. As one sign of this, Lenin remains an important reference point for 89 3 Administrative Templates Standardizing Vagueness high-ranking technical experts within the General Statistics Office—a half dozen of whom cited the very same line that Dũng did in their speeches commemorating the agency’s sixtieth anniversary in 2006.1 Interestingly, the second half of Lenin’s statement, drawn from his well-known but incomplete manuscript—The Capitalist System of Modern Agriculture—is never included.2 In it Lenin denounced the misuse of statistical information, particularly the tendency to combine data across scales (e.g., small-, medium-, and large-scale farms). Doing so, he explained, provides information on aggregate changes in total production but elides critical differences in the relations of production that differentiate them. The failure to take these relations of inequality into account, Lenin argued, enabled agricultural census data to be “converted . . . into a monstrosity, into statistics for the sake of statistics, into a game.”3 Lenin’s critique, though directed at “bourgeois economists,” equally applies here. All states seek to create information systems that will facilitate administrative control. Socialist ones are no exception. Indeed, since centralized decision making rather than price signals related to supply and demand determined the allocation of resources in Vietnam, the production of accurate and timely statistical data that technocrats could consult was essential. However, the legibility devices the relevant ministries created to regularize data collection and present the findings using fixed formats consistently made living conditions on agricultural cooperatives less rather than more legible to policy makers in Hanoi. The devices did so because official representations of reality increasingly obscured actual conditions in the countryside following collectivization . As a result technocrats found it quite difficult to manage the centrally planned economy, then taking shape, in a “scientific” manner. Several processes promoted “papereality.” First, “administrative templates” (biểu mẫu văn bản hành chính), the legibility device featured in this chapter, required low-level cadres to fill in standardized forms rather than writing long narrative-style reports as they previously had done.4 Second, the increased use of templates, which began in the late 1950s shortly before collectivization and then accelerated after it, obligated them to sequence and layer the information in predetermined ways. Sequencing refers to the procedures lowlevel cadres now needed to follow to gather information, transform it into data (typically in statistical form), and then submit the data to their administrative superiors on a more frequent basis. Layering refers to the graphic elements (headings, margins, blank spaces, boxes, tables, columns, etc.) that structured how these data had to be entered and visually presented.5 Ironically , both requirements contributed to the growing “vagueness” (mập mờ) of the information high-level officials had at their disposal because the formal 90 The Centralization of Documentation properties of the...

Share