Week 4 Study Guide: How Was the World Datafied?

Big idea

1. States needed to govern people and territory
2. So they had to count and classify people
3. This created "populations" as an idea
4. Statistics (science of the state) was invented to describe those populations
5. Those same statistical tools are invented and got hijacked by eugenics
6. And after WWII, this same logic went global (UN, IMF, development, fertility control)
7. What data we leave out and whose interests it encodes just reshapes the reality it claims to just describe (performativity).

Population

Core idea: States can't govern what they can't see. To govern, they simplify people into countable categories. This creates "the population" as a concept that didn't exist before.

• Sovereignty through knowledge: knowing your subjects = a form of power. Foucault's biopower: managing human life itself as a resource for political power.
• "Statistics" literally means "science of the state" (German Statistik). In English it was called "political arithmetic(k)" (William Petty, John Graunt, late 1600s).
• John Graunt's "Table of Casualties": counted deaths in 17th-c. London. Turned unique individual deaths into an abstract, aggregate "population." First move from individuals → statistical patterns.
• The "administrative grid": states impose simplified categories (like Pennsylvania's surveyed grid, or census categories) onto messy reality.
• Performativity: describing/classifying/measuring the world also changes reality to fit the categories.
• U.S. Census (1790): first count sorted people into: free white males (16+ and <16), free white females, other free persons, slaves. Shows how categories reflect power/priorities of whoever's counting (race, gender, freedom status all baked in from day one).
• Quetelet's "Average Man" / social physics (1830s): Belgian statistician applied the normal (bell) curve to human traits (height, chest size, crime rates). Created the idea of "normality": an ideal type, with deviation = "abnormal"/"pathological."

Global

Core idea: After WWII, the same "view from above" that states used on their own populations got scaled up to a global view through the UN, IMF, World Bank, etc. "The global" is invented during mid-20th-century as a word/concept.

• National → global administrative grid: standardized categories, processes, sampling/estimation methods, indices (CPI, GDP) now scaled up.
• Who else uses these tools besides states? Insurance companies, political pollsters, marketers.
• What made "the global" emerge (post-WWII):
  1. Cold War (First/Second/Third World framing)
  2. Decolonization (new independent nations, "Global South," non-aligned movement)
  3. Economic globalization (flows of goods/capital/labor, multinational corporations, "development")
  4. New international institutions like UN, World Bank, IMF, mostly created by/for the Global North, producing data "in service of international management and development."
• Case study: Gapminder.org: a visualization for global population/fertility/health data. It questions whose narrative does "global data" tell, and whose does it leave out?
• Population control narratives:
  • Fear of a "population explosion" → framed poor women (esp. women of color in the Global South) as the "problem," fertility management as "the solution."
  • This ignored the real drivers (capital flows, wages, jobs, migration, environment/pollution).
  • Continues colonial power relations by turning some women into "experimental subjects" and economic resources to be managed by elites.
  • Shift over time from 1950s-60s "population control" language to 1990s-2000s now reframed as "women's empowerment" and family planning as individual choice.
• Representing people as data involves a core tension:
  • 👍: It simplifies complex information so systems and institutions can function.
  • 👎: It can feel unfair, incomplete, and agency-denying to the people being modeled.

James Scott, Seeing Like a State

Core idea: Scott uses the history of scientific forestry in 18th-century Prussia/Saxony as a metaphor for how states (and other powerful institutions) manipulate the reality.

• States/officials couldn't see a forest. They could only see "commercial wood": a narrow fiscal abstraction (revenue-yield-per-acre). Everything else (fodder, medicine, habitat, local uses) fell outside the frame as it was "bracketed."
• Foresters eventually didn't just simplify their description of the forest. They physically re-engineered real forests into neat single-species, same-age rows (monoculture) to match their tables and match the "Normalbaum" (standardized/model tree). Simplification became a tool of control, not just measurement.
• Short-term, this worked great (easy to count, manage, harvest). Long-term, it caused ecological collapse (Waldsterben, "forest death": soil depletion, pest outbreaks, storm vulnerability) because so much of the real complexity had been ignored/bracketed out.
• Scott then draws the exact same logic to taxation and measurement: pre-modern states couldn't tax effectively because local measures were wildly inconsistent.
• The metric system (post-French Revolution) is presented as the parallel case to scientific forestry: a standardizing, simplifying technology imposed from the state, which:
  • Helped commerce/trade (uniform units)
  • Helped the state tax/administer more precisely
  • Was tied to Enlightenment ideals of "the same law, same measures, same rules" for all citizens (bundled with an emancipatory promise of equal citizenship)
  • Took a very long time to actually replace local practice (people kept using old units informally for decades)

Reading Question 1:

What makes German forestry similar to the metric system? What's the point of the comparison?

Both are cases of the state imposing a single, simplified, uniform standard on a messy, locally varied reality, in order to make that reality legible, countable, and controllable from the state. Both did so by literally reshaping reality to fit the standard (real forests became monoculture grids; local measurement customs were forcibly replaced by the meter).

Reading Question 2:

What does Scott mean by "legibility"? Legible to whom, for what?

Legibility = making something readable/measurable/manageable from a distance, by a central authority, using simplified categories (tables, maps, grids, standard units) instead of engaging with the full, messy, local complexity of the thing itself. Something is illegible when it exists only in dense, local, particular knowledge that outsiders (state officials) can't parse or act on. Legibility becomes performative once they start reshaping reality to actually match the grid.

Aubrey Clayton, How Eugenics Shaped Statistics

Core idea: The three "founders" of modern statistics (Galton, Pearson, Fisher) happened to also hold eugenicist views. In fact, eugenics was the actual motivating purpose behind many of their key statistical inventions.

• Galton: invented regression/correlation, popularized the bell curve for human traits
• Pearson: invented the chi-squared test and the basic 3-step template of significance testing. It was developed specifically to argue that distinct races/skull types existed and differed in worth.
• Fisher: wrote the textbook (Statistical Methods for Research Workers, 1925) that made significance testing the industry standard across all of science. He also pushed for sterilization of the "feeble-minded" and defended Nazi geneticists after WWII.
• Central argument: significance testing only asks "does a difference/effect exist at all?", not "how much," "why," or "does it matter." That "sizeless" yes/no framing was exactly what eugenicists needed (they just needed to prove two "races" were statistically distinct, but they didn't care about magnitude or cause).

Reading Question 3:

How should we think about statistics, given this history? What does this help us do (since we're not going to stop using stats)?

We should stop treating statistics as a purely neutral, objective, "view from nowhere" tool, and instead recognize it as a human practice shaped by the interests and worldviews of the people who built it. We should also be suspicious of any claim to pure objectivity, focus on magnitude, mechanism, and context, and push for a more inclusive, self-aware, "decolonized" statistics.

Alastair Brooks, Why Are Certain Countries Poor?

Core idea: Global inequality isn't a race where some countries are "behind". It is rather a single connected system where wealth in one place often creates poverty in another (relational comparison).

• HDI (Human Development Index): Measures health, education, and income. It is better than GDP, but still misleading (e.g., Ireland ranks #2 largely due to corporate tax-haven profits, not actual living standards).
• The "Ladder" Myth: The popular idea that poor countries are just "behind" and will naturally catch up by following the Western playbook ignores history. Wealthy nations built their wealth through colonialism and resource extraction from the global south.
• Geography Isn't Destiny: Bad geography (hurricanes, heat, swamps) doesn't force a country to stay poor. Regions like Florida, Hong Kong, and the Netherlands thrived because political and economic systems matter far more than location.
• "Backward" Culture as an Excuse: Labeling colonized societies as "primitive" masked the reality that their exploited labor built the colonizers' modern economies.
• Case Study (Sierra Leone vs. UK Healthcare): Sierra Leone's struggling healthcare isn't "stuck in the past." It is directly tied to the UK's present success where doctors trained in Sierra Leone move to work for the UK's NHS, simultaneously enriching British healthcare while crippling Sierra Leone's.

Quick Reference

Concept	Definition
Performativity	Measuring/classifying doesn't just describe reality, but it also reshapes reality to fit the categories
Legibility	Making messy reality readable/manageable from a central, distant point of power
Biopower	Managing human life/bodies as a resource for state power
The "view from above"	Statistics/data lets elites/states see populations in aggregate, from a distance, instead of as individuals
"Objectivity" as rhetoric	Claiming neutrality can be a strategy to make biased conclusions seem unquestionable
Norming/pathologizing	Defining an "average"/"normal" type inherently defines deviations as "abnormal" or inferior
Relational vs. comparative thinking	Comparing entities as if independent (ladder/timeline) vs. recognizing they're causally connected (one's wealth = another's poverty)