Scoring rubric — social-angle analysis of the PK textbook corpus

Purpose. Make cross-board, cross-class and cross-subject judgements comparable. Every score must be anchored to cited textual evidence (file + page marker). A score without evidence is not a score.

Unit of scoring is the BOOK, not the board. Board-level figures are aggregates of book scores. This matters because the boards’ books are not of one generation (see Vintage Control).

The six lenses

Scoring axes

Each axis runs 1 to 5. 1 is the secular / pluralist / liberal pole; 5 is the religious / exclusivist / conservative pole. The poles are descriptive, not evaluative — a high score is not a verdict, it is a measurement of a textual property.

A1 · Religious saturation (L1)

Applies to books other than Islamiyat/Mutalea Quran, which are religious by definition and are scored separately on A1b.

1. No religious content beyond a conventional bismillah.
2. Isolated religious reference, incidental to subject matter.
3. Recurrent religious framing; at least one dedicated religious lesson/section.
4. Religious framing structures the subject (openings, worldview claims, providential causation).
5. Subject matter is subordinated to religious instruction.

Indicators: religious selections as share of units; whether books open with religious material; providential causation (“by the grace of Allah, made the impossible possible”); Quranic epigraphs in science; Khatm-e-Nabuwwat declarations; honorific density.

A1b · Doctrinal vs ethical register (Islamiyat and religious content anywhere)

1. Ethical/character framing, comparative or historical treatment.
2. Devotional and ritual instruction, in-tradition but non-polemical.
3. Polemical, boundary-policing, or supremacist framing.

A2 · Civic thickness (L2) — reverse-scored: 1 = thickest civics

1. Constitution, rights instruments, elections, civil society, dissent all taught.
2. Substantial civic content, some instruments taught.
3. Civic content present but thin or purely institutional.
4. Civic content residual; state described, citizen not empowered.
5. No civic content; ideology occupies the slot.

Indicators: named instruments (1973 Constitution, UDHR, CRC); protest/boycott/petition taught as legitimate; civilian control of the military stated; whether pupils are asked to evaluate the government; rule of law; minority political rights.

A3 · Inquiry vs received truth (L3)

1. Contested questions presented as contested; evidence weighed; multiple theories.
2. Mixed — inquiry verbs present but conclusions pre-loaded.
3. Conclusions handed down to be defended or memorised.

Indicators: “justify that X” style prompts; whether decline/causation theories are plural; whether historiographical disagreement is acknowledged; source-criticism teaching.

A4 · National narrative closure (L3, L4)

1. Multiple actors, internal causes admitted, failure and wrongdoing acknowledged.
2. Standard nationalist arc with some internal criticism.
3. Externalised blame, no internal fault, providential inevitability.

Indicators: 1971 framing; whether the surrender is named; treatment of military rulers; treatment of Zia/Islamisation; whether any Pakistani actor is criticised.

A5 · Militarism (L4) — 1 = absent, 5 = pervasive

1. Armed forces absent or purely institutional.
2. Wars narrated; forces praised in historical context.
3. Martyrdom taught as aspiration; forces in non-history subjects; jihad-framed warfare.

Indicators: Nishan-e-Haider; martyr biographies; martial content in English/Urdu readers; “spirit of Jihad” applied to conventional war; named military operations.

A6 · Othering intensity (L5) — 1 = pluralist, 5 = hostile

1. Out-groups treated neutrally or warmly; shared heritage acknowledged.
2. Neutral-descriptive.
3. Out-group as historical adversary only.
4. Collective negative characterisation of a group.
5. Essentialised hostility; group framed as inherently threatening.

Indicators: “Hindus” as a bloc noun; whether Hindu/Sikh/Christian Pakistanis appear as protagonists or only inside Two-Nation exposition; India as enemy vs neighbour; treatment of the West; minority-rights lessons; whether coexistence is modelled.

A7 · Gender conservatism (L6)

1. Women as agents across roles, including intellectual and professional.
2. Women present, mostly commemorative or domestic.
3. Women largely absent, or roles prescriptively bounded.

Indicators: female-to-male pronoun ratio; women as authors vs subjects; women profiled as scholar/scientist/ruler; whether norms about women’s conduct are stated descriptively without critique; dedicated gender units.

Composite indices

• Secularity index = A1 (and A1b where applicable). Lower = more secular.
• Conservatism index = mean of A1, A2, A4, A5, A6, A7. Lower = more liberal.
• Report both, plus per-axis scores. Never collapse to a single number without showing the parts.

Vintage control — mandatory

The corpus spans four curriculum generations in simultaneous use, and this confounds every board comparison:

Balochistan is the control the study did not expect to have. Because it runs two generations side by side, it supplies vintage-matched partners for both of the others: its classes 6–9 match Punjab’s SNC 2022 books, and its class 10 matches KP’s and Sindh’s NC 2006 books. Any claim that survives both matched comparisons is provincial; any claim that reverses across them is generational.

Every vintage above is read from the books’ own front matter. Where a board publishes a governing scheme on its website, that scheme is not treated as evidence of what its textbooks were written against — Balochistan hosts the federal “NCC Inclusive Scheme of Studies 2024”, and the phrase appears in none of its 28 volumes.

Rule: when reporting a board difference, state whether the compared books share a vintage. Where they do not, give the vintage-matched comparison as well (e.g. compare Punjab’s 2022 books against KP’s 2022 class-8 books, not against KP’s 2006 ones). Flag any finding that cannot be vintage-controlled as confounded.

Evidence and confidence rules

1. Quote only from English-medium sources. Urdu Nastaliq OCR is theme-grade: it supports structure, topic coverage and recognisable names, never wording.
2. Cite as file · PAGE n. Page markers are PDF indices unless the book had native text.
3. Record confidence per score: high (English text, direct evidence), medium (English OCR, or inference from structure), low (Urdu OCR, or absence-based).
4. Absence is the weakest claim. Report “not established” rather than “absent” for Urdu files, unless the term would survive degraded script (proper nouns usually do).
5. Where boards are running the same book (established for Islamiyat 6–7, four science subjects, Urdu 9), do not report a board difference — report the shared score once and say so.