Lens L6 — Gender · Axis A7 (gender conservatism)

Scored: 74 books (75 files; Sindh’s Pakistan Studies Urdu and English files are one book, scored once per RUBRIC §5). Boards: PCTB Punjab, STBB Sindh, KPTBB KP. Classes 6–10.

Axis definition (RUBRIC A7). 1 = women as agents across roles, including intellectual and professional · 3 = women present, mostly commemorative or domestic · 5 = women largely absent, or roles prescriptively bounded.

Two distinct routes to a 5. The axis collapses two different textual properties into one pole. Throughout this document they are kept apart and tagged:

• 5-abs — absence. The book contains almost no female-marked language at all. Overwhelmingly the sciences and Punjab’s geography volumes.
• 5-pre — prescription. Women are present and their roles are stated as bounded. Sindh’s Pakistan Studies “four walls” passage is the type case.

This distinction matters because the two are produced by different mechanisms and would be fixed by different edits, and because the pooled board means are dominated by 5-abs science books whose gender content was never contested by anyone.

Citation convention. file · PAGE n, where PAGE n is the ===== PAGE N ===== marker in the extracted file, i.e. the PDF page index, not the printed folio (typically 1–6 ahead). Native-text files without markers are cited by printed folio or by nothing, and are marked as such.

Evidence rule (RUBRIC §1) applied strictly. Every verbatim quotation below is from an English-medium source and has been re-grepped against its file. The 24 Urdu-medium books (all Punjab and KP Islamiyat, all twelve Urdu readers, KP’s six History/Geography volumes, Sindh’s General Science 6 and Pakistan Studies Urdu) are scored at low confidence from structure, lesson titles and proper names only, drawing on track urdu, track islamiyat and track social studies. Nothing from an Urdu file is quoted anywhere in this document.

0. Headline

Both the scored axis and the independent lexical measure put the boards in the same order: Punjab ≈ Sindh < KP. Punjab and Sindh are close and differ in kind (Punjab concentrates its gender content in one chapter; Sindh distributes it and contradicts itself across vintages); KP is lower than both on every measure, in every class, in every subject that can be measured.

1. Method — the female:male gendered-pronoun ratio

1.1 Token sets

Two sets were computed for every English-medium file in the corpus:

A reconciliation finding. The published English-track ratios (Punjab 0.75/0.40/0.37/0.86; Sindh 0.41/0.40/0.45/0.52; KP 0.37/0.38/0.40/0.31) are reproducible only with Set B — i.e. the English track excluded hers and him. Set A recomputes those same four Punjab readers as 0.64/0.38/0.34/0.79. The published numbers are not wrong, but they are not the token set the brief specifies, and him alone moves Punjab 6 by 0.11. Both columns are therefore reported in §2 so that the new subject figures can be compared against the published reader figures without a silent apples-to-oranges step. All cross-subject comparisons in this document use Set A.

The English track also computed Punjab Class 9 from the OCR file, not the native one; the same book scores 0.86 (Set B) from OCR and 0.46 from the native text, because the native extraction is missing units 1, 4 and 5 — and unit 5 is Women Empowerment through Stories of Entrepreneurship. The OCR figure is the correct one for the whole book and is used here.

1.2 Counting rules

1. Case-insensitive, whole-word matching (\bwoman\b etc.). Word boundaries dispose of the substring traps by construction: \bman\b cannot match inside human, woman, Roman, German or Ottoman; \bmen\b cannot match inside specimen or implement; \bher\b cannot match inside other, there or whether; \bhis\b cannot match inside this. No stop-list is needed and none was used.
2. her is counted once per occurrence regardless of whether it is possessive or objective; the male side counts his and him (Set A) to keep the two sides morphologically symmetric. This is why Set A is the defensible set and Set B systematically flatters the female side.
3. Watermark lines removed. Lines containing perfect24u (KP mirror-site stamps) are dropped before counting. PCTB’s interleaved NOT FOR SALE / PCTB fragments do not contain any counted token and were left in place.
4. Element-symbol correction. In chemistry, He is the symbol for helium and collides exactly with the male pronoun. A capital-He token is counted as a pronoun only if it is followed by whitespace and a lower-case letter (He was, He wrote); all other capital-He tokens (H, He, Ne, C₆He, table cells) are treated as element symbols and removed from the male count. This removes 13 tokens from KP Chemistry 9, 10 from Sindh Chemistry 9 and 8 from Punjab Chemistry 9 — enough to move a chemistry ratio by ~0.03 and nothing else in the corpus by more than 0.02. The corrected column is labelled ratio (adj) and is the figure used everywhere.
5. Urdu-medium files are not counted at all. The measure is defined on English morphology.

1.3 Two traps the measure does not survive, and what to do about them

These are the most important caveats in this document, because in two books they invert the result.

(a) Nations and institutions are grammatically female. In Pakistan Studies prose, her is routinely a personification: “the problem of India is her partition”, “Muslim League felt that she can do for the safety of the Muslims”, “Pakistan, with her limited resources, housed the…” — all from pakistan-studies-9-en. KP Pakistan Studies 9 scores an apparent 0.217, but of its 25 female tokens, women occurs once, woman/girl zero times, and on inspection essentially every she/her is a personified state or party. Its true human-female content is one word in 24,912. The ratio overstates it by more than an order of magnitude. Sindh Pakistan Studies 10 is inflated the same way by a different mechanism: its his / her legal boilerplate in the UDHR and Ombudsman chapters (“when his/her rights are violated”, “he / she is entitled to realization of economic…”).

(b) OCR noise fabricates female tokens in low-density books. Punjab General Science 8 scores 0.211, but its eight female tokens are "On decay of dea ies 2) She organisms", "Genetics is the science of her . Bene is the basic unit", "Iron Her copper, sodium", "How do the following reactants eo her" — four garbage strings plus he/she generics. Its true value is ≈ 0.05.

Rule adopted here: a pronoun ratio below ~0.25 computed over fewer than ~50 female tokens is not interpretable and is reported with a dagger (†). Every such figure was manually inspected before scoring, and the A7 score follows the inspection, not the number.

1.4 What the measure is good for

It is reliable for the readers (English 6–9), where female tokens number in the hundreds, prose dominates, and personification does not occur. It is a genuine, independent corroboration of the scored axis there. Outside the readers it should be read as a screening statistic that tells you where to look, not as a result.

2. Scored table — every book on A7

ratio (adj) = Set A, element-corrected. Set B = the English-track-compatible figure, given for the readers so the published numbers reconcile. † = ratio not interpretable (see §1.3). 5-abs / 5-pre per §0.

2.1 PCTB Punjab — mean A7 4.15

2.2 STBB Sindh — mean A7 3.91

2.3 KPTBB KP — mean A7 4.35

2.4 BTBB Balochistan — mean A7 3.44

Vintage: classes 6–9 are National Curriculum 2022-23 pilot editions (matched to Punjab’s SNC 2022); class 10 is National Curriculum 2006 (matched to KP and Sindh). The “NCC Inclusive Scheme 2024” recorded in the catalogue appears in none of the books — see lens religion §2.4.

A7: 1 = women as agents across roles, 5 = women largely absent or prescriptively bounded. Feminine:masculine pronoun-and-noun ratio is given as a structural indicator, not as a score.

Balochistan A7 mean, English-medium: 3.80 (n = 20) — second of the five boards, behind the Federal Board at 3.30. The mean conceals a bimodal distribution rather than a moderate middle.

The board splits into two groups with almost nothing between them. Geography and chemistry are the most female-absent books in the entire corpus — Geography 6 and Geography 7 contain zero feminine tokens across a combined 39,609 words, and the two chemistry volumes contain five between them across 107,025 words. Against that, Pakistan Studies 10 narrates a woman as head of government, which no other board’s book in this corpus does at comparable length.

Balochistan names no woman scientist in any science book. Across roughly 218,000 words of English-medium science its only Curie occurrence is Curie temperature, the physical constant. Sindh returns the same null. This is not, however, a corpus-wide silence — see §2.5.

2.5 Class 10 completion — the matriculation year

Feminine:masculine token ratio is a structural indicator, not the score.

Women scientists — where they are named, and by whom

Counting these requires care, because most occurrences of the string “curie” in this corpus are not the person: they are the curie, a unit of radioactivity, and the Curie temperature, the point at which a magnet demagnetises. A count that does not separate the woman from the unit will produce a wrong answer in either direction.

Distinguishing them, the corpus contains:

The corrected finding is sharper than the false one. Two of four boards name a woman scientist and two name none. KP — the board with the worst gender score in the corpus (A7 4.31) — is the only board that names two, credits one with a major discovery in her own right rather than as a co-worker, and prints her portrait. Punjab names Curie once, in a list, as a person who studied what a man discovered. Sindh and Balochistan name none.

That inverts the simple reading. A board can be the most gender-conservative in the corpus on aggregate measures of presence and agency and still be the only one to show a girl reading its physics book a woman who explained nuclear fission. Aggregate ratios cannot see this, which is precisely why the per-book tables exist.

What survives of the original claim. Named women scientists remain extremely rare: four mentions across eleven matriculation science volumes and roughly 530,000 words, two boards with none at all, and no woman scientist named anywhere below class 9. The silence is real. It is not total, and this study should not have said it was.

2.6 Corpus expansion — and the corpus’s first female-majority science book

Added 23 Aug 2026. Feminine:masculine token ratio is a structural indicator, not the score.

FBISE A7 mean, English-medium: 3.30 (n = 10) — the lowest (least gender-conservative) board mean in the corpus, below Balochistan’s 3.75.

Three results from this batch are worth separating out.

Sindh’s General Science 7 is the first book in this corpus where feminine tokens outnumber masculine ones (ratio 1.03, across 48,708 words). It is the only book of 165 to score 1 on this axis. Whatever produced it, it demonstrates that the corpus-wide pattern is not a constraint of the subject or the language.

Punjab’s Geography 6 and Geography 7 contain zero feminine tokens — not a low ratio, zero, across a combined 35,509 words. Together with Balochistan’s two geography volumes, which do the same, this makes geography the most female-absent subject in the study, more so than physics or chemistry. That is not an artefact of technical vocabulary; it is a subject about places and people written as though only men inhabit them.

The federal board splits by subject exactly as the provinces do. Its Pakistan Studies 10 and English 10 score 2, while its Physics 9 and Chemistry 9 score 5 at a ratio of 0.11. The corpus-wide finding holds here too: no woman scientist is named in any federal science volume.

3. Vertical progression, class 6 → 10

3.1 Mean A7 by class

The pooled figure is nearly flat, and that flatness is the wrong summary. The three boards move in opposite directions between class 8 and class 9, and the direction is predicted by curriculum vintage, not by grade.

3.2 The pattern, stated properly

• Punjab improves into class 9 (4.17 → 4.00, and its two class-9 English-medium humanities books both score 2). Its class-9 English reader and its class-9 Pakistan Studies are the two post-2020 books in the corpus with dedicated gender content, and its class-9 reader ratio rises from 0.338 to 0.792.
• Sindh decays into class 9 (4.00 → 4.33) — but only in Islamiyat and in the 2004 Pakistan Studies. Its English reader improves (0.379 → 0.486, A7 3 → 2). Sindh’s real discontinuity is at class 8, not class 9: its 2015-curriculum Social Studies 7 scores 2 and its early-2000s federal-reprint Social Studies 8 scores 4. A Sindhi pupil’s most gender-critical year is class 7, and it is followed immediately by the corpus’s most conservative civics text.
• KP decays hardest and most consistently (4.17 → 3.83 → 4.67 → 4.71). It is the only board whose reader pronoun ratio falls at class 9 (0.352 → 0.282, the lowest reader figure anywhere), and it is the only board with no post-2020 book at class 9.

3.3 Does the Sindh Islamiyat decay pattern hold elsewhere? — Yes, in three of four places

The known lead: Sindh Islamiyat runs exactly one woman per volume at classes 6/7/8, always first in the personalities chapter, then zero in the IX–X volume. Confirmed at stbbislamiyat-6-en · PAGE 81 (Khadijah), class-7 · PAGE 78 (Ayesha), class-8 · PAGE 67 (Fatima al-Zahra), against class-9 · PAGE 157, 164, 169, 172 (Imam Hussain, Abu Ubaidah bin Jarrah, Jabir bin Hayyan, Muhammad Rashid Raudhe Dhahni — all male).

Testing the same shape across the corpus:

Reading. The decay is real and is the default: five of nine strands lose their women between middle school and matriculation. But it is not a property of adolescence — it is a property of the 2004/2006 curriculum generation. Every strand that decays is on that base. Every strand that inverts is a post-2020 revision (Punjab NCP-2023 English 9, Punjab 2023 Pakistan Studies 9, Sindh 2026 English 9). The class-9 collapse is what happens when nobody has revised the class-9 book.

4. Cross-board comparison, vintage-controlled

4.1 The vintage map for gender

Every book in the corpus with a dedicated gender unit is either a post-2014 provincial curriculum or a post-2020 national one. There is no 2004/2006-generation book in this corpus with a gender unit, in any province.

4.2 Vintage-matched pairs

These are the comparisons that survive RUBRIC’s vintage rule.

(a) 2022-curriculum Islamiyat, Punjab 8 vs KP 8 — no board difference. Both carry a 2022 national-curriculum stamp (track islamiyat §1). Punjab retains its standing Ummahat sub-unit; KP’s rebuild profiles Juwayriyah and Safiyyah, 2 of ~12. Both score 4. When the curriculum is matched, the Punjab–KP gap in Islamiyat disappears.

(b) 2004/2006-generation Pakistan Studies, Sindh vs KP — no board difference. Sindh’s book is approved 09-10-2004; KP’s class 10 is NC-2006 (NOC 2011). Both score 4. Sindh reaches it by prescription (5-pre: the “four walls” passage) and KP by thinness. Punjab’s Pakistan Studies 9 scores 2 — but Punjab’s is a 2023 book. The 142-vs-1 “women” gap between Punjab 9 and KP 9 is a vintage gap, not a province gap, and any write-up that presents it as evidence about Punjab’s politics versus KP’s is making a confounded claim.

(c) 2006-generation Islamiyat 6–7, Sindh vs KP — the same book. track islamiyat §6.1 establishes these are the same syllabus in two languages: identical five-chapter architecture, identical hifz surahs, identical four-person personality rosters in identical order. Per RUBRIC §5 the score is reported once: A7 = 4 at both classes, shared. No board difference exists to report.

(d) 2006-generation Urdu 9, Punjab vs KP — the same book. track urdu §5.1 establishes 11 of 11 prose lessons and all 8 poetry selections match, in near-identical order. Shared score A7 = 5, reported once. Both are all-male anthologies; this is a single editorial decision made once at the federal level and reprinted twice, not two provincial decisions.

(e) Readers, unmatched and confounded. Punjab 6–8 are SNC-2022, Sindh 6–8 are undated SPELT books, KP 6–9 are NC-2006. The reader comparison (Punjab 0.539 > Sindh 0.403 > KP 0.335) is the strongest measured board difference in this document and it cannot be vintage-controlled. Flag as confounded: KP’s low ratio is consistent with a board effect and with the fact that its reader is a 2006-base classic anthology (Sindbad, Grimm, Rumi, Dickinson) whose male-default narration was never commissioned to a gender quota.

4.3 What genuinely distinguishes the boards, after vintage control

Two things survive:

1. Punjab has the largest single investment and the smallest distribution. One chapter of one book (Pakistan Studies 9, pp.97–108) carries almost the whole load. Remove that chapter and Punjab’s mean rises from 4.15 to 4.23 and its best English-medium book becomes History 8.
2. Sindh is the only board that teaches gender critically — i.e. that names discrimination as a defect of society rather than describing women’s contributions. Social Studies 6’s media unit and Social Studies 7’s Gender-inequality section are the only places in the corpus where pupils are asked to identify and criticise gendered expectations, rather than to admire named women. That is a different pedagogical act from Punjab’s, and it is a Sindh-specific one.

5. Punjab — per-board assessment

Mean 4.15. Shape: one very high peak, a broad floor, and a self-contradiction inside the peak.

5.1 History 8 — women of the Pakistan Movement

history-8-en (native text, printed pp.62–63) carries a section headed “Role of Women and Students in Pakistan Movement” with seven named women. They act:

“Fatima Sughra, who installed the flag of the Muslim League on the Civil Secretariat was an active member of the Pakistan Movement. She was only 14 years old at that time.”

“Begum Raa’na Liaquat Ali was the first lady of Pakistan… She was the first woman Governor of Sindh. Before the establishment of Pakistan, she founded a women’s organization, All Pakistan Women’s Association (APWA). She also served as Pakistan’s ambassador to the Netherlands and Italy.”

This is genuine agency — founding, governing, representing. But note the framing of every entry: sister of the Quaid, wife of Pakistan’s First Prime Minister, Begum Maulana Muhammad Ali Jauhar, wife of Allama Iqbal’s close friend Barrister Shah Nawaz. Six of the seven are identified by a male relative before anything else. Scored 3, not 2, for that reason.

5.2 Pakistan Studies 9, Chapter 4 — the corpus’s high-water mark, and its internal split

pctbpakistan-studies-9-en · PAGE 97–108. It is the only book in the corpus at pronoun parity (1.02). What it does that nothing else does:

• Names living/recent women in office, with their offices: Benazir Bhutto (PM, twice), Shamshad Akhtar (Governor of the State Bank), Dr Nafis Sadiq (“the first woman in the world to hold such a high position in the United Nations”), Dr Fahmida Mirza (“the first woman from Pakistan to be the Speaker of the National Assembly from 2008 to 2013”), Samina Baig (Everest, seven summits), Arfa Karim, Bilquis Edhi (PAGE 102–103).
• Credits a woman with running a business, in a religious register no less:

  “Hazrat Khadija… was a wealthy lady in the Arabian Peninsula. She was the owner of a trade centre in Makkah and handled the business herself. Her merchandise was exported to far-away markets like Syria.” (PAGE 100)

• Names the violence and refuses to blame the victim. A Myth/Fact apparatus (PAGE 105):

  “Myth: … Fact: Some people argue that violence depends on the way the women dress up, their marital status, their way of living, their mindset etc. This argument is incorrect because it puts blame on the victim rather than the offender.”

• Explicitly rejects seclusion as a safety strategy (PAGE 105):

  “Myth: Violence against Women occurs when they are performing some role outside their house. Fact: In our society, the women’s life outside the house is usually unsafe. Therefore, women are forbidden or discouraged from stepping out into public places. This concept is incorrect… instead of restricting women from public places, emphasis should be placed on making all public places available and secure for women.”

• Names the Punjab Marriage Restraint Amendment Act 2015 and the Punjab Protection of Women Against Violence Act 2016 with its offence list (PAGE 106), the district Violence Against Women Centres with all-female staff, the 1043 helpline and the 8787 SMS line (PAGE 106–107), and CEDAW 1979 (PAGE 107).

And the same chapter, nine pages earlier, says the opposite:

“Islam is a religion that has not only dignified woman but has also provided her the protection in the form of veil and four walls.” (PAGE 99)

That single sentence is the exact proposition the Myth/Fact box on PAGE 105 exists to refute. This is the sharpest intra-book split message in the corpus — sharper than any of the cross-subject ones — and it is worth more than the cross-board comparison, because it shows the chapter was assembled from two sources with incompatible premises and never reconciled.

5.3 Pakistan Studies 10 — agency asserted, then naturalised away

pakistan-studies-10-en (native, no page markers; Chapter 8):

“Conservative families who never thought their daughters could pursue a career other than becoming doctors or teachers. Today, their daughters are showing their talents as lawyers, engineers, fashion designers, politicians, Air Force Pilots, Civil Service officers, Army officers and media anchorpersons etc.”

immediately preceded by:

“Distinguishing people on the basis of man or woman is called Gender Discrimination. Nature assigned separate roles for men and women. Their main purpose was to carry forward the human race.”

Also records women joining the Air Force “as flying fighters” and being “recruited as cadets and engineers for the first time in the army”. Scored 3: the professional list is the widest in the corpus, but it sits on a naturalised-roles premise the book does not examine.

5.4 The Punjab floor

Punjab’s three Geography volumes contain, between them, zero female-marked tokens in 52,827 words. History 6 and 7 add 27 female tokens against 729 male ones. Punjab’s SNC split of Social Studies into History + Geography (track social studies §5.2) put the entire gender load into History and left Geography with none — and History 6 and 7 spend their years before 1857, where the board has chosen to name no women at all.

6. Sindh — per-board assessment

Mean 3.91 — the lowest (most inclusive) of the three. Shape: critical where the curriculum is recent, prescriptive where it is old, and the two sit in the same building.

6.1 Social Studies 7 — the only critical gender text in the corpus

stbbsocial-studies-7-en · PAGE 161, under the heading Gender-inequality:

“In most cases, women do not get what they deserve in Asian counties. Women hardly get equal wages to that of men in different parts of Asia. Women are not allowed to pursue the career of their choice. They face a number of challenges when they come out of house. Even in some countries like Afghanistan, Pakistan and India, the conservative people don’t permit them to get education which is appalling. On some occasions, they are tortured and killed in the name of honour.”

“There is need to build more educational institutions for women… adequate representation of women in the law making bodies can help to get rid of male dominating society.”

And at PAGE 163:

“Women are denied from their right of education for centuries. Although, they contribute more than 50% of population but their ratio in education is very low. There are multiple reasons behind this issue like so called cultural traditions, poverty, non-availability of girls’ schools and female teachers, migration etc.”

No other book in the corpus — Punjab’s women’s chapter included — names honour killing, or the phrase “male dominating society”, or attributes the deficit to “so called cultural traditions”. Punjab describes what women achieved; Sindh 7 describes what is done to them and calls it a defect. Scored 2.

The same book also credits medieval Islamic medicine with “female doctors and nurses” (PAGE 35) — one of the corpus’s rare attributions of a profession to women in a historical setting.

6.2 Social Studies 6 — the split message, confirmed and located

The lead is confirmed, and it is sharper than reported. The book contains both of the following.

Uncritical statement of conduct norms. In the “Customs and Traditions” column of a table of South Asian countries (PAGE 194):

“Heads of women are to be covered / Men like to shake hands with other men. Women do not shake hands. They are permitted to nod.”

and, three pages later, under a heading Norms in a section on “Factors that Influence Attitudes and Values of the People of South Asia” (PAGE 197):

“Norms: Girls are not encouraged to move freely at night, whereas boys can go freely everywhere.”

Neither carries a critical apparatus, an activity, or a question. The PAGE 197 passage is especially load-bearing because it is not presented as one country’s custom but as a general norm of South Asian society, in a list alongside “Respecting the elders”.

Explicit critique of exactly the same kind of expectation. In the Media unit (PAGE 100):

“Gender Stereotyping: There are some common stereotypes of men and women, such as: Men are the ‘bread winners’, that is, they earn the money. Women are ‘housewives’, they stay at home. Men are the “backbone” that is they are strong and do all the work. Women are ‘dumb’ that is they are not as smart as men.”

with cultural stereotyping exemplified by “Girls are not good at sports” (PAGE 101), and pupil activities: “List the gender stereotypes promoted by the media in Pakistan” (PAGE 101) and “Give three example of how women are stereotyped in Sindh” (PAGE 103).

The civics unit adds a third register — women as under-represented citizens:

“Identify the problem women citizens have in participating in the political process and suggest ways to address them” (PAGE 79); “Make an inquiry to find out the problems women face in casting [votes]” (PAGE 81)

and the book states flatly, on PAGE 24, “Men and women are equal.”

Assessment. This is not hypocrisy so much as a units-authored-separately artefact: the Media and Governing Ourselves units were written to a competency spec, and the South Asia unit was written as descriptive ethnography with no gender spec at all. The result is that a Sindhi eleven-year-old is taught in Unit 4 to detect the belief that women belong at home, and taught in Unit 5 that girls do not go out at night. Scored 3 — the critical material is real and substantial, and it is what keeps this book off a 4.

6.3 General Science 7 — the highest ratio in the corpus, and why it means less than it looks

general-science-7-en scores 1.188, the only English-medium book above parity. It gets there on ~35 tokens, and Sindh’s science books do place girls in the investigator role by name:

“She got confused that some flowers have large and wavy stamens, whereas some have small stamens. She also… coloured petals. She took pictures of the two flowers. You should help Sadia to explore the type of flower she…”

That is a girl doing science. But the book’s other female exemplar is used for heat radiation:

“Fatima is a nature lover, she… seasons. She is always… She selects beautiful colours for her dresses that enhance her mood to enjoy nature. She always selects light… she avoids dark colours…”

— i.e. a physics principle taught through a girl’s choice of clothing. Sindh is the only board that puts girls in science activities at all; it is also gender-typing them while doing it. Scored 3, with the ratio flagged as uninterpretable.

Sindh Biology 9 is the corpus’s one science book that discusses gender as a subject:

“3. Inequality — Because of a cultural preference for men over women in many developing countries, many women risk malnutrition throughout their lives. The risk for malnutrition in girls begins at an early age. Although nutritional needs are the same for boys and girls in the first 10 years of life, boys often get more food than girls do.”

6.4 Pakistan Studies 9–10 — the prescription case (5-pre)

stbbpakistan-studies-10-en · PAGE 138, section “ii. Status of Men and Women”, taught as a description of Pakistani culture:

“Male member has occupied a unique status in Pakistani culture. He is the head of the family. He is the dominant member. But a woman is also considered an important part of the family who governs and manages all family affairs within the four walls. Household keeping and upbringing of children is entrusted to her in a family. She has the right to education, right to property and right to business in accordance with the principles of Islam.”

This is A7’s prescriptive limb exactly: roles stated as bounded, no critical apparatus, and — because it is in the Culture chapter — presented as a shared national heritage rather than a contested practice. It is the strongest single counter-example in the corpus to any claim that Sindh is uniformly the most liberal board; it is also a 2004 book, and its own board’s 2015 Social Studies 7 contradicts it directly.

Offsetting it, the same volume’s later civics chapters give the Women Ombudsperson and the Protection Against Harassment of Women at the Workplace Act 2010 a full treatment (PAGE ~8331–8345 lines), including how to file a complaint. Net score 4.

7. KP — per-board assessment

Mean 4.35 — the highest (most conservative) on every measure. Shape: not split, just thin.

KP is the only board with no internal contradiction to report, because it has almost nothing on either side of the argument. Its position is produced by three facts:

1. One gender-focused selection in four English readers (Naseem Hamid, class 7). Punjab has three and Sindh two in the same span (track english §2).
2. The lowest reader pronoun ratio in every single class (0.346 / 0.360 / 0.352 / 0.282), and the only board where the ratio falls at class 9.
3. Pakistan Studies 9 mentions women once. The apparent 0.217 ratio is almost entirely the grammatical personification of nations and parties (§1.3a).

7.1 KP’s genuine exceptions, and where they are

KP has three, and they matter precisely because the board is otherwise the weakest.

(i) The corpus’s only women rulers. Both in Urdu-medium History:

• Razia Sultana (History 6, pp.60–61) — treated sympathetically at length, her fall attributed to the nobles’ refusal to accept a woman ruler.
• Nur Jahan (History 7, pp.26–28) — an unusually long treatment as a political actor.

Punjab’s History 7 profiles Noor Jahan too but as its single woman; no Sindh book teaches either figure. So the answer to “does any board profile a woman as a ruler?” is: yes, exactly one board, in exactly two books, both Urdu-medium, both about medieval figures whose authority the text itself records as contested (track social studies §1.10, §1.12; low–medium confidence, Urdu OCR).

(ii) The corpus’s only women authors. Julia A. Carney (English 6), Christina Rossetti (English 7) and Emily Dickinson (English 8 and 9) — all three, in KP, with About-the-Poet biography boxes. See §8.2(a). No Punjab or Sindh book credits a woman with authoring any selection.

(iii) KP Physics 9 is the most gender-balanced science book in the corpus (0.727), because its worked problems use female actors as a matter of routine — a driver who brakes, a girl whose mass is 60 kg, a woman applying 400 N of force. That is a deliberate authoring habit, it costs nothing, and no other board does it.

These three exceptions do not rescue KP’s score, because none of them is about gender: two are side-effects of an old anthology’s reading list, one is a physics-problem convention, and the two women rulers are never referred to again after class 7. But they do complicate the simple story. KP is the board least likely to teach about women and, in two narrow respects, the most likely to let a woman speak or act without comment.

7.2 The irony worth recording

kptbbpakistan-studies-9-en · PAGE 100 carries the author biographies. One of them reads:

“Lecturer in Geography, Jinnah College for Women, Peshawar University. The authoress is attached with teaching in a well-known college of ‘Peshawar University for about 9 years. She has rendered valuable services in the field of Geography. She is an expert teacher of Geography.”

A woman co-wrote the book that mentions women once. She is the only woman credited as an author of anything anywhere in this corpus — and she is credited in an author-biography appendix, not as the author of a selection. (Compare track urdu §4: across all twelve Urdu readers, no woman is credited as author of any selection.)

8. Cross-subject delivery — what a pupil in each province is actually told

This is the question the lens is for. Take a pupil through one board’s whole year and ask what composite picture of women they receive.

8.1 The four registers, and who uses them

Across 74 books, women appear as agents in exactly four registers, and no others:

8.2 Where women are never agents — the three empty rooms

(a) No woman is credited as the author of any selection in any Urdu reader. track urdu §4 establishes this for all twelve (medium–high confidence: every legible author-bio page names a man).

The English readers are the one place the corpus does credit women as authors — and all of it is in KP:

Three women, three books, one board — and it is the board that scores worst on every other gender measure in this document. The mechanism is KP’s 2006-base classic-anthology reading list (track english §4), which was never commissioned to a representational spec and therefore admits women by accident, where the theme-commissioned SNC books admit them only by quota and only as subjects. Punjab and Sindh credit no woman as the author of any selection in any medium.

The nearest Punjab equivalent is a passing reference, not a selection: Punjab English 7 names “‘The Story of My Life’ by Helen Keller” in a list of books of literary value (english-7-en line 1356); Sindh English 9 teaches Keller as a subject at unit 6.4, not as an author.

(b) No woman is profiled as a scholar in any Islamiyat volume, in any board. Confirmed in track islamiyat §6.6 and re-checked here against the four English-medium Sindh volumes. The scholar slot in all three boards is filled by hadith and fiqh imams, Sufis, and Ibn Sina / Jabir bin Hayyan / Ibn Khaldun — every one of them male.

(c) No woman is named as a scientist in twelve English-medium science books. This is a new finding from this lens and it is the cleanest one in the document. Across ~500,000 words of Punjab, Sindh and KP science, the named scientists are Newton (160), Rutherford (44), Ibn Sina (44), Hooke (36), Bohr (36), Galileo (32), Jabir (29), Einstein (24), Linnaeus (17), Dalton (16), Thomson (13), Mendeleev (13), Archimedes (10), Leeuwenhoek (9), Schrödinger (7), Faraday (7), Crick (6), Chadwick (6), Watson (5), Lavoisier (5), Pasteur (4), Mendel (4), Darwin (3), Al-Razi (2), Fleming (2), Abdul Qadeer (2), Copernicus (1) — all male.

There is exactly one exception in the corpus, and it is instructive:

“Cm is the chemical symbol for curium, named after the famous scientist Madam Curie. Why wasn’t the symbol C, Cu or Cr used instead?” — kptbbchemistry-9-en · PAGE 40

Marie Curie appears once, in a short-answer question, as the source of a naming puzzle — never as a scientist who did something. And the counter-case is exact: Watson and Crick are credited with the DNA double helix in Punjab General Science 8, Punjab Biology 9 and Sindh General Science 8, and Rosalind Franklin appears zero times in the entire corpus.

The agent-vs-subject finding in one sentence: all three boards teach about women in at least one subject, and no board in any subject presents a woman as an author, a scholar or a scientist — so the three roles through which a pupil would learn that women produce knowledge are empty in every province, while the roles through which women are honoured are full.

8.3 Punjab’s composite message: concentrated, and contradicted

A Punjab pupil at class 9 reads, in the same year:

• Pakistan Studies: women as Prime Minister, central-bank governor, UN under-secretary-general; violence typology; victim-blaming named as a myth; public space asserted as a right (PAGE 105).
• The same chapter, PAGE 99: Islam “provided her the protection in the form of veil and four walls.”
• English: Hazrat Asma and Women Empowerment through Stories of Entrepreneurship, adjacent.
• Islamiyat (Urdu): a Sahabiyat section and a lesson on the Prophet’s good treatment of women — the women’s-rights-as-conferred frame, not the women’s-rights-as-claimed frame.
• Urdu (2006 book): twenty selections, twenty male authors, no woman.
• Biology, Chemistry, Physics: 13 female tokens in 145,565 words, of which zero refer to a named woman.

So Punjab’s message is split by concentration: one chapter says everything, contradicts itself once, and nothing else in the pupil’s timetable reinforces it. Remove Chapter 4 of Pakistan Studies 9 and Punjab is indistinguishable from KP.

8.4 Sindh’s composite message: split by vintage, inside the same subject

A Sindh pupil gets the corpus’s sharpest whiplash, and it runs within one subject across consecutive years:

• Class 6 Social Studies (2014 curr.): “Men and women are equal” (PAGE 24); identify how advertisements stereotype men and women (PAGE 100–101); identify the problems women citizens have in political participation (PAGE 79) — and “Girls are not encouraged to move freely at night” as a stated norm (PAGE 197).
• Class 7 Social Studies (2015 curr.): honour killings, unequal wages, “male dominating society” (PAGE 161).
• Class 8 Social Studies (early-2000s federal reprint): the gender-critical apparatus vanishes; a short equal-opportunity paragraph remains (PAGE 35).
• Class 9–10 Pakistan Studies (2004): the man is “the head of the family… the dominant member”; the woman manages “within the four walls” (PAGE 138).
• Islamiyat (2006 base, English medium): one woman per volume at 6/7/8, first position; zero at IX–X.
• English 8 and 9: two full gender units, one of them hidden from its own contents page.

Sindh therefore teaches gender equity and prescriptive domesticity in the same subject, three years apart, without ever acknowledging the contradiction — the precise pattern the brief asked about, and it is Sindh’s, not Punjab’s. track social studies §5.1 reaches the same conclusion from a different direction (“Any provincial characterization of ‘Sindh’s textbooks’ is false at the grade level”).

8.5 KP’s composite message: coherent, and coherently thin

KP is the only board whose subjects agree with each other. One gender lesson in four English readers; one woman per Islamiyat volume then (provisionally) none; one word for women in Pakistan Studies 9; nothing established in three Urdu geographies. Its two women rulers sit in Urdu-medium History at classes 6 and 7 and are never referred to again. There is no split message to report because there is no second message.

9. Notable findings

1. Absence, not prescription, is the dominant mechanism — overwhelmingly so. 31 of the 74 books score 5, and every one of the 31 is 5-abs. They break down as 15 science books, 6 geography volumes, 4 class-9 Urdu readers, 2 Islamiyat IX–X volumes, Punjab’s two pre-1857 History books, KP History 8 and KP Pakistan Studies 9. Not a single book in the corpus reaches a 5 by prescribing women’s roles. Only two books prescribe at all — Sindh Pakistan Studies 9–10 (scored 4-pre) and one sentence inside Punjab Pakistan Studies 9 (scored 2) — and both of those books also contain substantial women’s-rights material, which is exactly why neither reaches the pole. The corpus’s gender conservatism is almost entirely a matter of what is not written, not of what is forbidden.
2. The science books are the floor, and they are the same floor in all three provinces. Mean A7 for the 18 English-medium science books is 4.78 — 15 of the 18 score 5. Two contain zero female-marked tokens (Punjab Chemistry 9, Sindh Chemistry 9) and five more contain exactly one (Punjab General Science 7, Punjab Biology 9, KP General Science 7, KP General Science 8, Sindh Physics 9). This is the largest single block of books in the corpus and the one where the three boards are least distinguishable from each other.
3. Punjab’s geography is the extreme case: three volumes, 52,827 words, zero female-marked tokens. Not one she, her, woman, girl. Exactly five English-medium files in the entire corpus contain no female-marked token at all, and three of the five are Punjab’s geography books (the other two are the Punjab and Sindh chemistry volumes).
4. The 142-vs-1 Punjab/KP gap is a vintage artefact. Vintage-matched Pakistan Studies (Sindh 2004 vs KP 2006) shows no board difference at all: both score 4.
5. The class-9 collapse is real, general, and curricular. Five of nine subject strands lose their women between class 8 and class 9–10. Every one of them is on the 2004/2006 base. Every strand that inverts is a post-2020 revision.
6. Sindh’s one-woman-per-volume-in-first-position Islamiyat rule is confirmed and is unique to the 2006 base, which KP shares — and Punjab, which does not run the four-profile format, is the only board whose Islamiyat does not drop to zero at class 9.
7. Sindh Social Studies 6 teaches a critical vocabulary for gender stereotyping on page 100 and states an uncritical gender norm on page 197. Both re-verified verbatim.
8. Punjab Pakistan Studies 9 contradicts itself within nine pages — “veil and four walls” (PAGE 99) against “instead of restricting women from public places, emphasis should be placed on making all public places available and secure for women” (PAGE 105). This is the sharpest split message found, and it is intra-book.
9. Women scientists are named four times in the entire English-medium science corpus. Marie Curie appears in KP’s Physics 10 (p. 244, with dates) and Punjab’s Physics 10 (p. 174); Madam Curie in KP’s Chemistry 9 (p. 40), as the answer to a chemical-symbol puzzle; and Lise Meitner in KP’s Physics 10, credited with the theoretical explanation of nuclear fission (p. 252) and given a portrait at Figure 18.12 (p. 253). Sindh and Balochistan name none; no board names one below class 9. Care is needed in any recount: most occurrences of the string “curie” in this corpus are the unit of radioactivity or the Curie temperature, not the person. Rosalind Franklin still has zero occurrences corpus-wide while Watson and Crick are credited for DNA in three books.
10. The only three women credited as authors of any selection in the corpus — Julia A. Carney, Christina Rossetti and Emily Dickinson — are all in KP’s English readers, i.e. in the board that scores worst on every other gender measure. Punjab and Sindh credit none, in any medium. The mechanism is KP’s 2006-base classic-anthology reading list, which was not commissioned to any representational spec and therefore admits women by accident, where theme-commissioned SNC books admit them by quota and only as subjects. This is the single most counter-intuitive result in the document, and it is the one place where the vintage effect runs opposite to its usual direction.
11. A woman co-authored KP Pakistan Studies 9, the book that mentions women once.
12. KP Physics 9 routinely uses female actors in worked problems and is, for that reason alone, the most gender-balanced science book in the corpus (0.727). It costs nothing and no other board does it.

10. Caveats

1. Two-thirds of the confidence in this document rests on English-medium files. 24 of 74 books are Urdu-medium and are scored at low confidence from track urdu, track islamiyat and track social studies. Per RUBRIC §4, absence-based scores on those files are the weakest claims here. The specific ones that should not be relied on are KP History 8, KP Urdu 8 (the most degraded file in the corpus) and KP Geography 6/7/8, all scored 5 on absence that may be OCR failure. They are marked as such in §2.3.
2. KP Islamiyat IX–X is scored provisionally. track islamiyat §7 establishes that roughly the back half of that scan is missing and that no personalities chapter is visible in the surviving pages. The 5 is a placeholder; the class-9 decay finding for KP Islamiyat is stated as “probably”, not “confirmed”, and would be settled by a complete PDF.
3. The pronoun ratio is a token count, not a role count (as track english §8.5 already warned). This document adds two specific failure modes — nation-as-her personification and OCR-fabricated tokens — and marks 24 of the 43 ratios printed in §2 with † as uninterpretable. Every daggered figure was manually inspected before the A7 score was set, and where the number and the reading disagree, the reading was used (KP Pakistan Studies 9 is the clearest case: apparent ratio 0.217, actual human-female content one word).
4. The reader comparison cannot be vintage-controlled and is flagged confounded in §4.2(e). KP’s low ratios are consistent with both a board effect and a 2006-anthology effect and this corpus cannot separate them.
5. Scores are integers on a 5-point axis over heterogeneous book lengths. A 49-page KP Geography 8 and a 232-page Sindh Social Studies 6 each contribute one score to their board’s mean. The board means in §0 are therefore book-weighted, not page-weighted or pupil-hour-weighted, and the subject-matched figure (English + Islamiyat + Urdu, 12 books each) is the more defensible comparison.
6. Board means are composition-sensitive. Punjab and KP each publish three science and three geography volumes per middle-school year where Sindh publishes one integrated Social Studies book; since science scores near 5 everywhere, that composition alone moves Punjab and KP up relative to Sindh. This is the main reason the subject-matched comparison is reported alongside.
7. No claim here has been checked against a printed copy. The Punjab Pakistan Studies 9 internal contradiction (§5.2) and the zero-women-scientists finding (§8.2) are the two results most worth confirming against physical editions before publication.
8. Sindh’s General Science 6 (Urdu) is scored by analogy, not by evidence. It is the single weakest row in the table and should be treated as a placeholder.
9. “No woman as author” rests on track urdu’s medium–high confidence, not on certainty: a female author whose name OCR’d to noise could have been missed. The claim is strongest for Punjab Urdu 9 and KP Urdu 9, both of which were fully reconstructed from clean author-biography pages.