Punjab · Sindh · Khyber Pakhtunkhwa · Balochistan · Federal

What Pakistan's school textbooks actually say

167 textbooks in use today across Pakistan's four provincial boards and the Federal Board, catalogued page by page, alongside 43 withdrawn editions reaching back to 1949. The scored corpus was read against a rubric fixed before scoring began, and every figure here traces to a book, an edition and a printed page.

167

current textbooks, acquired and verified

25,740

printed pages analysed

12

subjects across classes 6–10

10

curriculum vintages in simultaneous use

A further 43 historical Punjab editions (1949–2005) are catalogued separately. They are held to show what changed, and are never counted in the figures above or scored against the rubric.

The finding

The expected story — four provinces, four ideologies — is mostly not what the books show. What divides them most is the decade a book was written in. Four curriculum generations are in print at once, and because Sindh and Khyber Pakhtunkhwa share one, they have converged: in Islamiyat at classes 6 and 7 they run what is effectively the same book, and in four science subjects their chapter lists match down to section headings.

Where edition age is controlled for, a real provincial difference survives — and it runs opposite to what "old books awaiting revision" would predict. Punjab prints the newest textbooks in the country and produces the most closed historical narrative among them.

Five boards, seven measures

Each panel is one board's average profile across the seven axes, on the same 1–5 scale. Low is the secular, pluralist, open pole in every case. A line that climbs to the right is a curriculum that stays secular on religion while scoring conservative on othering and gender — the commonest shape here. Gaps are axes with no scored book.

  1. A1 Religious saturation
  2. A2 Civic thickness
  3. A3 Inquiry vs received truth
  4. A4 National-narrative closure
  5. A5 Militarism
  6. A6 Othering intensity
  7. A7 Gender conservatism

Lowest religious saturation: Federal at 1.70. Least conservative overall: Federal at 2.49. Every mean is over English-medium books only, and coverage differs by axis — the axes pages print each denominator, and several of these gaps are generational rather than provincial. The synthesis separates the two.

Two scales that disagree

Balochistan is scored across all seven axes and included in every figure below. It runs two curriculum generations at once — classes 6–9 on National Curriculum 2022-23, class 10 on National Curriculum 2006 — which makes it the only board supplying vintage-matched partners for both of the others, and the study's most useful control for separating era from province.

Scored across seven axes, Khyber Pakhtunkhwa's curriculum is the most religiously saturated of the five — yet it carries the most open historical narrative, naming its own military rulers' overreach. Punjab measures as less religious and more closed, more militarised, and more hostile toward Hindus and India.

Being religious and being conservative are different properties here, and they vary independently. Any single ranking that merges them will mislead.

Adding the Federal Board — which teaches Islamabad and the federal-system schools in the provincial capitals and cantonments — changed the shape of that comparison. It is the most secular and the least conservative of the five on both indices at once, and it is the board the provinces are nominally working from. Every provincial version of the same curriculum is more religiously saturated, thinner on civics and more gender-conservative than the federal text it derives from. Its base is the smallest in the study, though, and that caveat travels with the claim.

“Justify that the Treaty of Sevres is based on fair justice.”
Balochistan Textbook Board, History 8, page 32 — an exercise setting pupils to defend a proposition the same book treats as a grievance. Read the article →

Start here

Four pieces written for a general reader, each drawn from the evidence and each traceable back to it.

The textbook that tells the truth and never asks what you think

A book gives sixteen-year-olds a frank account of military rule — then asks them to memorise the dates.

The woman who explained nuclear fission

We published a finding. A reader told us it was wrong. He was right, and the truth is stranger.

Justify that the treaty was fair

Textbooks full of the vocabulary of critical thinking. Look at what the verbs are attached to.

Same curriculum, different books

Two provinces, one federal document. Only one teaches the constitution as a living guarantee.

All articles →

Not sure where to begin? Four reading paths — parent, journalist, policymaker, researcher →

Findings →

Discrete, page-cited findings you can quote and check one at a time — each with its source, its confidence, and the evidence against it.

The database →

All 167 books, filterable by board, class, subject, medium and curriculum vintage. Each opens to its full citation and scores.

The studies →

Nineteen reports: six social lenses, six subject tracks, the synthesis, the rubric, the edition audit and the long-form articles.

Method →

Corpus construction, evidence tiers, the rubric band definitions, vintage control, and how to interpret a score.

By board →

Scored profiles for each provincial board, with evidence-tier composition.

By subject →

Coverage of each subject across the boards, including where they structure a subject differently.

Search →

Full text across every book record, study and method page.

What every board shares

Some findings show little or no provincial variation, and they may matter most. Across roughly 530,000 words of English-medium matriculation science, a woman scientist is named four times — all of them by two of the five boards, none below class 9. The "Hindu teachers" explanation for the 1971 separation survives across 2004, 2023 and 2024–25 editions. Partition violence is absent from every book. And at matriculation, open-opinion questions are set freely on climate, technology and literature, and never once on a contested national question.

Two of those sentences are corrections. This site previously claimed no woman scientist is named anywhere, and that inquiry collapses entirely at matriculation. Both were overstated; both were withdrawn, and the reasons are in the corrections log and in an article about how the first one was caught.