HS2024 - Lecture 7 - Numbers, Facts, and Good Government PDF

Artificial Intelligence AI&HV & Human Values Spring 2024, ETH Zürich / Prof. Margarita Boenig-Liptsin Fall 2024, ETH Zürich / Prof. Margarita Boenig-Liptsin Lesson 7: Numbers, Facts and Good Government - Part I  Where We Are AIxWorlds Unit Ti...

Artificial Intelligence AI&HV & Human Values Spring 2024, ETH Zürich / Prof. Margarita Boenig-Liptsin Fall 2024, ETH Zürich / Prof. Margarita Boenig-Liptsin Lesson 7: Numbers, Facts and Good Government - Part I  Where We Are AIxWorlds Unit Time, October 30, 2024 Big Data, AI and conditions of knowledge at-large Riding on the metaphor of data as "new oil" “Big Data” Slide from as sociotechnical fact (high volume,variety, Lectu velocity) re 3 as imaginary and “world schema” (“dataism”) Context or situation of data: world as laboratory How is data valued in different contexts? Political economy of data Balance shifts from academic to corporate context The Economist, February 2010 "Dataist state" (Fourcade and Gordon, "Learning Like a State: Statecraft in the Digital Age, 2020) Destabilizing the relations among numbers, facts, and ideals of good government The Guardian, August 17, 2020 Sora demo, March 2024  Questions for today's lecture How did the science of statistics arise and what relationship does it have to nation states? How has the state’s relationship to its citizens and territory been shaped by the sciences and technologies of identification and calculation? How do nation-related concepts like sovereignty, territory, and population matter in contemporary debates about AI? The origins of statistics Why begin with a history of statistics?  Putting the "state" in statistics "Statistics" comes from the German "Statistik" – science of state "Population" as an object formed by state knowledge and power, an object of governance Administrative data collection on populations makes possible modern mathematical statistics How the modern state, "populations," and statistics co-produced one another Why did states begin collecting population data? Modern states, since 17th c., governing through knowledge Administrative aim: governance of a territory and population - Increasing wealth of the territory - Raising revenues (taxes) - Raising armies - Growing population and improving its health (emergence of "biopower") Frontispiece for Thomas Hobbes' Leviatha (1651)  17th and 18th centuries 1. Statistik: Science of the state in German-speaking central Europe - Perspective of the state/principality - Aim at increasing state power - Management of territory, natural resources (forest management), trade - Qualitative, holistic  17th and 18th centuries 2. Political Arithmetick (England/Great Britain) - William Petty and John Graunt - Quantitative, atomistic (1662) (1690) A world of tables Deaths in 17th c. London Where did this data come from? Why was it collected? What was done with it? Why?  17th and 18th centuries 2. Political Arithmetick (England/Great Britain) - William Petty and John Graunt - Quantitative, atomistic - Numerical data initially collected locally and repurposed for use by government (1662) (1690)  State technologies of vision "Certain forms of knowledge and control require a narrowing of vision. The great advantage of such tunnel vision is that it brings into sharp focus certain limited aspects of an otherwise far more complex and unwieldy reality" – James Scott, Seeing Like a State, p. 11 Image: Scott, "One aisle of a managed poplar forest in Tuscany," p. 17.  Statistics and Eugenics Francis Galton, "Rate of Regression in Hereditary Stature," 1886 To read more about the history of statistics and eugenics, see Aubrey Clayton, Nautilus, 2020  Artificial Intelligence & Human Values Fall 2024, ETH Zürich / Prof. Margarita Boenig-Liptsin Lesson 7: Numbers, Facts and Good Government - Part II  Numbers in Democracy (drawing on Nikolas Rose, "Governing by Numbers: Figuring out Democracy," 1991; Alonso and Starr, Eds., The Politics of Numbers, 1987) Two ideals of democracy (in US history) - Object of government is the pursuit of a public good that cannot be divided, where politics is an exercise of virtue - View that society is made up of multiple interests that need to be governed with information about the composition of the public (Rose, p.685; Petersen in PN, p.268)  The census is mandated by the US Constitution; first conducted in 1787 ○ Figuring political representation: "3/5ths compromise" Public numeracy: public ability to calculate is deemed republican not because it is generalized to all, but because it is "a pedagogy of reasoning itself" (Rose p. 683; Cohn Calculating People, 1982). Scientific American, 1890 US Census  What is counted is what is politically problematized "To count a problem is to define it and make it amenable to government" (Rose, p. 686) ○ US 1810 Census: questions on manufacturers ○ US 1840 and 1850 Census: statistics about social problems (illiteracy, ill health, insanity, pauperism, crime..) ○ US 1870 Census: collection of wage statistics and railroad and insurance industries Numbers are linked to problematization, to evaluation of government, and are the authority's claim of legitimacy Intertwining of political and numerical authority: "From this point on, [...] political disputes will be waged in the language of number" (Rose p. 685).  Privatization of the relation between numbers and politics in neoliberalism c. 1980s - 2010s, neoliberal era of governance Defined by suspicion of the capacity of governments to calculate and regulate national interest Delegation to personal, social and economic actors to produce numbers for reasoning and calculating their freedom. Requires: ○ a numericized environment ○ expertise of number (economists, accountants, statisticians, demographers) ○ techniques (censuses, surveys, national income tabulations, international benchmarks, accounting practices..) – see Rose, p. 690-691 "Dataist state" (Fourcade and Gordon, "Learning Like a State: Statecraft in the Digital Age, 2020) Destabilizing the relations among numbers, facts, and ideals of good government The Guardian, August 17, 2020 Sora demo, March 2024  Key Takeaways (I) Statistics, the state, populations, and the “social” co-produced one another over the preceding three centuries Statistics and “population thinking” arose from the state’s interest in governing its territory and increasing its power. Collecting and analyzing data to infer knowledge about “populations” states created an “administrative grid,” allowing them to “see” their populations by creating simplifying abstractions that reduce the complexity of the social and natural world in order to govern it. Democracy as a practice depends upon the composed relation of number and numeracy to support a calculated and calculating government Key Takeaways (II) Democracy requires citizens to know how to calculate, experts who calculate and provide their techniques to politicians and entrepreneurs, public habit of numbers Numbers are composed and stabilized, but this is not an intrinsic capacity of numbers–"we should not expect to find any essential unity to the relations of numbers and politics. Rather, it is a question of the 'what' and 'where' and the 'how' of their alignment with other governmental technologies" (Rose, p. 691). The relationship among numbers, facts, and ideals of good government evolve through history; in 20th c., scholars distinguish three periods: high-modernist state (c. 1950-1970), neoliberal state (c. 1980-2010s), and what some call "dataist" state (2010s-present)  Coming Up AIxWorlds Unit Reading for November 7: - Greenwald, Glenn. 2013. "NSA whistleblower Edward Snowden" (Video). - Eubanks, Virginia. 2018. “The Allegheny Algorithm.” In Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor.