DPHU1008 Chap.5 Classification Systems, Binaries and Social Constructs
Classification Systems, Binaries and Social Constructs
Lecture 4 looks at data infrastructures: the materials, systems and classifications that hold data together. Its key concepts are classification systems, binaries or binary oppositions, and social constructs. The required reading is the chapter of D'Ignazio and Klein's Data Feminism titled after Joni Seager's slogan 'what gets counted counts'.
It argues that every data set rests on classification, that systems become naturalised until they are contested or break, and that binaries such as male and female were built rather than found.
The chapter also names the paradox of exposure, in which those who could gain most from being counted face the most danger from it, and the privilege hazard, in which designers from dominant groups fail to anticipate those dangers.
Linda Tuhiwai Smith's Decolonizing Methodologies adds the historical view: imperialism classified who counted as human, and history itself was written around binaries that served power.
The chapter's cases give you material for any lens.
Facebook expanded its gender options to more than fifty in 2014 while still resolving users into a binary for advertisers; a newspaper replaced pink and blue with suffragette purple and green to unsettle a gendered hierarchy; a map of trans and nonbinary members of Congress was left blank so that absence became the finding. Each shows classification operating through design choices.
What this chapter covers
- 01
What gets counted counts: counting directs policy and resources
- 02
Classification is naturalised until contested or broken
- 03
Binaries are social constructs with histories
- 04
Inadequate categories versus an inadequate system
- 05
Paradox of exposure and privilege hazard
- 06
Smith: order, fragmentation and history as power
- 07
Counting with consent: when the ethical choice is not to collect
- 08
Design cases: colour, blank maps and expanded options
Worked example · free
Evaluating a dating app's profile categories
- +1Classification system: the profile fields sort every user into two categories that the matching algorithm depends on.
- +1Binary as social construct: the design treats a constructed gender binary as natural, so nonbinary users cannot represent themselves.
- +1System, not category: adding a third box would not fix matching built on two complementary groups, so the system itself needs redesign.
- +1Paradox of exposure: any new identity field must be optional and protected, since visibility can bring harassment.
Key terms
- Classification system
- A set of categories that sorts people or things and becomes part of an infrastructure.
- Binary opposition
- A pair of opposed categories, often ranked so that one side counts as normal.
- Social construct
- A category that exists because people made and maintain it, and can therefore remake it.
- Paradox of exposure
- The double bind in which those who gain most from being counted also face the most danger from it.
- Privilege hazard
- The blind spot of designers from dominant groups who are rarely the ones harmed when people are counted.
- Matrix of domination
- The intersecting system of race, gender and class that shapes whom data systems serve.
Classification Systems, Binaries and Social Constructs FAQ
Does Data Feminism say we should stop classifying people?
No. The chapter states that working infrastructures need classification. Its critique asks who designed a system, what values it encodes and who bears the costs of its gaps, and it treats counting as useful when consent and safety are protected.
What example explains the paradox of exposure?
The proposal to add a citizenship question to the 2020 United States census: undocumented immigrants would be less likely to respond, so their communities would lose representation and funding that follow the count.
How does Smith connect to classification?
Smith shows imperialism sorting people into hierarchies of humanity and splitting Indigenous knowledge across museums and disciplines. Her history gives a long background to the present-day form fields that Data Feminism analyses.
Why are binaries described as hierarchies?
D'Ignazio and Klein argue that paired categories such as reason and emotion, or male and female, usually rank one side as normal and the other as lesser. Rethinking binaries therefore means questioning the ranking as well as the number of categories.
What is the difference between an inadequate category and an inadequate system?
An inadequate category can be fixed by adding an option, such as a third gender box. An inadequate system builds a ranking or purpose into the whole structure, so the fix is to redesign the system or question why the data are collected at all.
Assessment move
Find three forms you have filled in this year and note every closed category they used. For each, decide whether the problem is a missing option or the whole system. Summarise D'Ignazio and Klein's key terms on one side of a page and Smith's on the other, then write a paragraph linking one of each to the same rendering.
Draw a quick table of every binary you meet in a week of advertising and note which side the design treats as the default. Practise one sentence that links a present-day form field to a historical example from Smith.
DPHU1007 Media, Society and Politics · DPHU1001 Journalistic Voices and Values · DPHU1003 News Fundamentals
