Robert Meyer’s published lecture to the 191st plenary assembly follows the Austrian census from administrative preparation through mechanical tabulation to scholarly publication. Its central argument is that counting machines do not eliminate human work: they reorganize it into coding, sorting, supervision, calculation, and decisions about what deserves publication. Drawing on the 1900 census while describing preparations for the 1910 enumeration, Meyer makes the usually invisible production of statistical knowledge the basis of an appeal for adequate funding.
The lecture first distinguishes the political administration’s responsibility for enumeration from the Statistical Central Commission’s advisory and processing functions. House numbering, locality registers, ministerial regulations, municipal participation, and instructions to respondents precede counting. Changes in the inquiry—including questions on disability, property, occupational change, and expanded housing coverage—connect technical preparation to social-policy needs. Local summaries provide rapid preliminary results, while centralized processing permits more elaborate analysis. Publication thus proceeds at different speeds and levels of detail rather than delivering one simultaneous census result.
Meyer then compares lists, individual cards, manual sorting, and alternative counting techniques before explaining punched-card machinery. Electrical contacts register the perforated attributes, while sorting compartments help arrange cards for subsequent operations. Yet the machine’s conspicuous speed rests on a costly preparatory transformation:
Dagegen ist die Anfertigung von mehr als 28 Millionen Lochkarten eine äußerst umfangreiche, schwierige und kostspielige Arbeit, die auch nicht der Fehlerquellen entbehrt.
English translation: By contrast, the preparation of more than 28 million punched cards is an extremely extensive, difficult, and costly task, which is not free from sources of error either.
Before punching, workers must annotate the lists so that occupations, birthplaces, and other answers correspond unambiguously to numerical codes. Up to 300 auxiliary workers will be required for more than a year. Mechanization therefore depends on prior interpretive and organizational labor; standardized input is an achievement, not an automatic property of census returns.
The physical design of the apparatus also determines the scope of possible knowledge. Cards contain 240 fields, and information omitted from them is lost to subsequent processing. Requests for additional distinctions encounter an inflexible machine format. The 77 counters similarly restrict simultaneous combinations: ten language categories crossed with eleven age groups already exceed their capacity. Presorting, repeated counts, intermediate tables, and subtraction compensate for those limits, but introduce further work. Meyer insists that these technical arrangements must be governed by the intended public result:
Alle diese Vorarbeiten und Einrichtungen, die ich geschildert habe, müssen geleitet sein von der Zweckvorstellung dessen, was schließlich zur Darstellung gebracht werden soll.
English translation: All these preparatory tasks and arrangements that I have described must be guided by a conception of what is ultimately to be presented.
This reverses any assumption that tables simply emerge from collected facts. The publication plan must shape the workflow from the outset. Selection is unavoidable because exhaustive combinations would overwhelm both the budget and the reader; greater substantive complexity generally requires less geographical detail. Separate processing of housing, buildings, or selected population groups extends the inquiry, but demands additional cards and carefully timed operations. The final publications also require textual interpretation and scientific assessment, completing the movement from individual returns to intelligible social knowledge.
The closing argument treats expenditure as a problem of public priorities rather than mere technical efficiency. Each additional sorting or counting operation carries substantial costs, as do typesetting and printing. Meyer invokes marginal utility to explain why even useful statistical proposals must compete with other state needs. Yet he equally rejects economy that leaves collected information unused or prevents nearly completed work from reaching publication:
Ganz besonders aber sollte vermieden werden, bereits eingeleitete oder ausgeführte Arbeiten liegen und stehen zu lassen, die dafür aufgewendeten Kosten daher völlig zu vergeuden, weil die Fertigstellung noch einen Zuschuß von ein paar tausend Kronen erfordert.
English translation: Above all, however, one should avoid leaving work already begun or carried out unfinished, thereby completely wasting the expenditure incurred on it, because completion requires an additional allocation of a few thousand crowns.
The lecture culminates in a request for sufficient resources to complete the census responsibly and preserve Austrian statistics’ international standing. Its lasting relevance lies in showing how administrative authority, human classification, machine capacity, and fiscal choice jointly determine what a census can make known. Economy requires limiting ambitions, but also protecting the analytical work that gives the much larger expenditure on enumeration its value.
This work was divided into 6 sections when it entered the library's research corpus—an apparatus for search and citation, not necessarily the author's own table of contents. Each title opens its summary.
Put a question to this work; the Librarian answers from its 6 sections and cites the passage.
Ask the Librarian