*Agent-readable manuscript for `english-for-homeschooling`. Headings match the book. Not a PDF dump.* # English for Homeschooling # English for Homeschooling Michael Gannotti ------------------------------------------------------------------------ This illustrated edition is the teaching-manual of 31 August 2026. Endnotes run in one series. Chapter-opening images are original still-lifes, not portraits of persons or children. Documents still constrain the facts. ------------------------------------------------------------------------ # Contents - Welcome - How to Use This Book This Week - If You Only Remember Five Things - The English Hour - Chapter 1 — Grades 1–2: Sounds, letters, and first sentences - Chapter 2 — Grades 3–5: Fluency, knowledge, and paragraphs - Chapter 3 — Grades 6–8: Longer books, argument paragraphs, grammar in the sentence - Chapter 4 — Composition - Chapter 5 — Literature - Chapter 6 — Language - Chapter 7 — Research and rhetoric - Chapter 8 — Speaking, listening, and media - Chapter 9 — Records - Chapter 10 — Resources - A Note on Sources # Welcome This book exists because the English hour at your table is the whole program. There is no department down the hall and no specialist waiting after lunch. There is you, a student, and today’s idea — a word to look through, a sentence to write, a page to read aloud. That is enough, if you know what to do with the hour. I wrote this for a capable, busy, willing parent. You may be teaching two ages at once. You may be fitting reading and writing between a job, a toddler, and a grocery run. You may have loved school English, or you may remember grammar as a fog and phonics as something other people taught. Many adults feel rusty when they sit down to teach the code or the sentence. That feeling is common. It is not a verdict. We will move on from it. This book will make you fluent enough in today’s idea to notice a wrong turn and ask a good question — without taking over the pencil. You do not need to be an English specialist. You do need to understand this week’s idea well enough to hear a picture-guess as “not reading yet,” and a topic sentence that is only a fact as “not yet an argument.” You need a session shape you can run on a Tuesday. You need a few sentences that actually help. That is the job. A path through English — from first-grade decoding through a college-ready high-school sequence — is possible at a kitchen table. The path is not a personality trait, a Lexile, or a percentile. It is a sequence of ideas, practiced until they hold, with you in the chair. About 3.4 percent of U.S. students ages 5–17 were homeschooled in 2022–23 — roughly 1.765 million children.[^1] That figure is context, not a ranking. You are one of those tables. Homeschooling is legal in all fifty states and the District of Columbia. The paperwork is not one load: Texas asks almost nothing of the state agency; New York asks for a written plan and regular reports.[^2] There is no national homeschool diploma and no national English credit a registrar already files.[^3] Four English credits are typical of state high-school maps.[^4] The titles that travel — Language Arts, English 9, English 10, American Literature, British Literature, World Literature, Composition, Speech — are names a stranger can read. The hour in front of you is the work. ## What a good English hour looks like You sit down already knowing today’s idea. The child warms up on words and sentences they can already get right. You model one move out loud, short: look through this word; write this kind of sentence; read this paragraph as it should sound. Then they try, and you wait. You ask one good question — not “did you get it?” but “how did you know that was *ship* and not *shop*?” Practice mixes the new move with last week’s. The hour ends with one short task they do alone: read this sentence aloud, write this sentence, tell what the paragraph said. You stop talking sooner than feels polite. The child holds the pencil, or the pointer. You hold the key. That is the hour. The next long piece of this front matter, *The English Hour*, will teach it in full. Later chapters will not reinvent it. Reading is the product of decoding and linguistic comprehension. If either is missing, the child is not yet reading that page, no matter how well they talk about the story.[^5] Writing is taught, not assigned. Model a sentence. Write one together. Then they write alone. Knowledge and vocabulary feed comprehension; a content-free “reading strategy” hour is not the path. Fluency is accurate, reasonably paced, expressive reading of text they can already decode — not a race. Grammar earns its keep in the sentence on the table. Those are the moves. The chapters will give you wording. The hour will stay this shape. ## What you will actually get Each teaching chapter does eight jobs, always in the same order, so you are never hunting for the move. You will learn why this week’s idea is worth the struggle — what it unlocks later. You will understand the idea yourself, in plain language, with one everyday picture and one precise picture, and with the wrong answers you should be able to hear. You will get a session you can run this week, including exact wording. You will get practice that builds learning, not a pile of the same worksheet. Your student will get a short section of their own. If it isn’t clicking, you will get three diagnostics and a next move for each, with no shame in the room. Tools, including AI, stay optional and adult-side. And you will get a plain checklist for “done enough,” so you can place by skill rather than by birthday. The eight teaching chapters follow the work, not twelve thin grade labels: the code in grades 1–2; fluency and reading to learn in grades 3–5; longer literature, argument paragraphs, and grammar in the sentence in grades 6–8; then high-school composition, literature, language, research and rhetoric, and speaking, listening, and media. Records and resources come last. This week you can learn the session shape and today’s idea well enough to hear a wrong turn. Today the child can try one word, one sentence, or one paragraph without you holding the pencil. ## What this book will not do This book will not hand you 180 days of worksheets. A year of photocopies is not a teaching method, and I will not pretend it is. It will not sell you a curriculum. Later, a short resources chapter names common programs by fit — parent load, how they teach the code, how they teach writing, how they treat a book, how they place a student — so you can choose. All About Reading, Logic of English, IEW, Well-Trained Mind, Brave Writer, Teaching the Classics, AmblesideOnline, and a textbook-anthology path are tools. None of them is this book, and this book is not a catalog in disguise. A family may combine. Combining does not turn a brand into a course name. The transcript still says English 9. It will not promise a score. There is no guaranteed percentile, no guaranteed Lexile, no guaranteed SAT Reading and Writing number, and no guaranteed college letter inside these pages. What this book can promise is a path: the ideas in order, at the skill the child actually has, until they can do the next idea unaided. NAEP Proficient is not “on grade level,” and there is no national homeschool reading score in these pages to wave.[^6] It is not a Common Core tract, for or against. Standards are a map you can steal clusters from. They “define what all students are expected to know and be able to do, not how teachers should teach.”[^7] They are not a homeschool mandate and they are not a method. Texas, Virginia, and Florida print their own maps. A publisher’s “grade 6” book is a scope, not a legal grade. English 9 is a body of skill, not a fourteenth-birthday present. It is not a science-of-reading war. Decoding is taught because it is how English writing works. You get the move. You do not get a culture fight. And this is not a sequel. *Autonomous AI and Education* and *Critical Thinking for Life* are other books. AI is a lane here: one English-hour box in this front matter, then a short tools box in each teaching chapter. Evaluation of claims lives inside the English work — the documented paper, the speech, the paragraph — not as a course titled Critical Thinking 101, and not as a course titled AI English. This is not a book that lectures your child about you, or you about your character. The student is a person, not a percentile. When this book speaks to them, it speaks with respect. No baby-talk. No research sermon. ## The promise If you only remember one sentence, remember this: a path, not a percentile. The work is to take this child through English — the code, fluency, knowledge feeding comprehension, sentences that become paragraphs that become essays, books taught rather than listed, language in the draft, a documented paper, a talk a person can hear — without skipping the code because they are “a good reader,” and without parking them in a book that is too easy because a catalog printed a grade on the cover. You will sometimes slow down. You will sometimes skip ahead. Both are teaching. Birthday is not placement. Three things have to hold. You understand this week’s idea well enough to hear a wrong turn. The child does the reading and the writing before the tool. Any helper — including an AI agent — stays on your side of the table: a supplement you host and constrain, not a substitute for the child’s work, and not a school-safe product by default. You can do this. You do not have to know next year’s idea today. You have to know this week’s idea well enough to sit still while they struggle, then ask one good question. Start here. Read *How to Use This Book This Week*, then the one-page list of five things, then *The English Hour*. After that, open the chapter that matches the skill in front of you. You will know more after one chapter than you know this morning. Your child will have something to try today. We can do this. # How to Use This Book This Week Start at the skill in front of you, not on page one of Grade 1 because a catalog, a birthday, or a well-meant relative said so. This book is a handbook you open to this week’s band, not a novel you read cover to cover. The teaching chapters run: grades 1–2 (the code — phonemic awareness, phonics, decoding, handwriting, oral language, first sentences); grades 3–5 (fluency as a bridge, reading to learn, paragraphs, spelling that maps to print, knowledge and vocabulary); grades 6–8 (longer literature and informational text, argument paragraphs, grammar as craft in the sentence they are writing); Composition; Literature; Language; Research and rhetoric; Speaking, listening, and media. Records and resources come last. Grade labels here are sequence anchors. They tell you where an idea usually sits. They do not tell you where *this* child sits. This book does not replace the program already on the shelf. If All About Reading, Logic of English, IEW, a Well-Trained Mind sequence, Brave Writer, Teaching the Classics, AmblesideOnline, or a literature anthology is already working, keep it. Use this book to hear a wrong turn, to run the hour, and to name the credit a stranger can read. The transcript still says Language Arts or English 9, not a brand. ## Pick the chapter by skill, not by birthday Open the chapter you think is right. Skip to **What “done enough” looks like** at the end of the *previous* chapter, or to **If it isn’t clicking** in the one you opened. If the child can already do those checks unaided, you are too early. If the checks from two chapters back are still failing, drop back. If they still guess from pictures, they are still in the code, whatever their age. A sixth-grader who cannot look through a word is not “behind in literature.” They are still in a code program. Guessing from pictures is not reading.[^8] If they cannot yet hold a sentence — if an essay prompt produces a list, a copied paragraph, or a stare — they are still in taught writing, whatever the transcript year. Model a sentence. Write one together. Then they write alone. Assigning the essay first is how composition becomes a fog. If they can decode connected text, read it aloud with accuracy and expression, and write a paragraph with a claim and a reason, and they are still in a “grade 4” reader only because the cover says so, skip ahead. Instruction for a six-year-old need not be “first grade”; it should match what the student actually needs and already knows.[^9] A publisher’s “Level 4” is a scope, not a legal grade. A ten-year-old can sit in three different “grades” in three programs on the same morning. Use this book’s checklists, or a placement test from the program you own, and then teach. Two children may need two chapters. That is ordinary. The session shape in *The English Hour* still holds; the idea on the table changes. You may run a twenty-five-minute code session with one child and a forty-five-minute composition session with the other. You do not need two personalities. You need two first problems. ## How you use the parent half Most of each teaching chapter is for you. Read it *before* the lesson, not over the child’s shoulder. **Why this matters** tells you what this idea unlocks. You will teach with more patience when you can see the later door. **For the parent: understand it yourself** gives one everyday picture, one precise picture, and three to five wrong answers you should be able to hear. Sit with those. If “look at the picture” still feels like reading help, that is the code idea, and it is worth five quiet minutes now. **How to teach it this week** assumes the session shape from *The English Hour*. It will not rebuild the hour. It will give you the first problem, the wording for *this* idea, and how to fade your help. **Practice that actually builds learning** is what goes on the page after the model: a short blocked set of the new move, then mixed work so the child has to choose, plus a little retrieval of words and sentences they already know. Knowledge from history, science, and literature read aloud can feed the hour. It does not replace looking through the word, writing the sentence, or the unaided exit. **Tools, including AI** is optional, short, and for the adult. The rules live once in *The English Hour*. **What “done enough” looks like** is how you leave. You are looking for unaided work, not a perfect Tuesday. Placement is by skill, not birthday. You do not have to read the whole chapter tonight. You do have to read the parent half of *this week’s* idea before you sit down with the child. ## The five-minute parent warm-up Make this a habit. Five minutes. Child not yet in the chair. Phone face down. 1. Read today’s idea until you can say it in one sentence. 2. Do the move yourself, on paper: look through the word; write the sentence; mark the claim in the paragraph. Stay off the key until you finish. 3. Glance at the “wrong answers you should be able to hear.” Name the one you would have given at fourteen. 4. Write one sentence you will actually say. Not a speech. Example: “Look through the whole word.” Or: “Show me the because.” Or: “Read it as it should sound.” 5. Close the book to the student page. You are ready. That five minutes is how you stay out of their working memory. If you are learning the idea *while* they are stuck, you will talk too much or take the pencil. Prepare first. Then sit still. You need today’s sentence, not a seminar. ## How the student uses “For the student” Every teaching chapter includes a short section written to the student, not about them. One or two pages. Warmer. Direct. What the idea is, a tiny worked example, two tries, an “explain it back” prompt, and one challenge. Hand them that section when it is time for their attempt — after your short model, not instead of it. They may read it silently or out loud. You stay in the room. You do not add a lecture about high school. The two tries are theirs. “Explain it back” checks that they have a picture, not only a copied step. The challenge is optional that day; it is there so a ready student does not sit in busywork. The student page is not something to send off with an unsupervised chatbot. The attempt is still theirs. You still hold the key. ## When to skip ahead Skip ahead when this chapter’s “done enough” checklist is already true *unaided*. Lingering past that is not thoroughness. It is boredom wearing a virtue. Slow down when the same wrong turn repeats after a clear model and a real attempt. **If it isn’t clicking** will give you three likely causes and a next move: a missing piece of the code, a sentence they cannot yet hold, or a page they can decode but cannot yet understand because the knowledge is missing. Missing letters or missing sentences are skill problems, not character problems. Ten minutes of struggle on a well-posed new idea is ordinary work. Diagnose decoding and the held sentence before you buy the literature anthology or start “the research paper.” Accelerate a child who has the prerequisites. Hold still for a child who does not. Both are the same kindness. If you need a human tutor, that is a normal high-school plan, not a failure of the kitchen. For this week: pick the chapter by skill. Do the five-minute warm-up. Run the hour as *The English Hour* describes it. Let the student page be theirs. Stop talking sooner than you want to. That is how you use the book. # If You Only Remember Five Things Keep this page. The chapters will add wording, pictures, and practice. They will not replace these. **1. The code is taught.** Decoding is how English writing works. Reading is the product of decoding and linguistic comprehension; if either is missing, that page is not yet being read.[^10] Teach the sounds, map them to letters, look through the word. Guessing from pictures is not reading. Fluency is what happens next: quick, accurate, expressive reading of text the child can already decode — not a race, and not optional. Knowledge and vocabulary feed comprehension. A content-free “reading strategy” hour is not the path. You do not wait for “readiness” as a personality trait. You teach the code, then you use it. **2. Writing is taught, not assigned.** Model a sentence. Write one together. Then they write alone.[^11] Grammar earns its keep in the sentence on the table — combining, expanding, fixing the draft in front of you — not as a worksheet island of parts of speech. An essay prompt is not a writing lesson. Struggle before rescue: ask, wait, hint, then model. The child holds the pencil. You do not finish the paragraph. **3. The child does the work before the tool.** The student reads the page and writes the sentence. You hold the key. A tool may explain today’s idea *to you*, make extra practice with answers hidden, write a hint script, or help you diagnose work already done. It may not sit in the chair during the attempt, paste the essay, or rewrite the passage so they skip decoding. The full rules are in *The English Hour*, once. Later chapters will point back rather than preach. **4. Transcript titles a stranger can read.** The work can be All About Reading, IEW, a living-books year, or a literature anthology. The credit is Language Arts, English 9, English 10, American Literature, British Literature, World Literature, Composition, or Speech.[^12] Never a brand. Never “Critical Thinking 101.” Never “AI English.” There is no national homeschool diploma. The record has to travel without you in the room to explain it. **5. A path, not a percentile.** The promise is this child through English, at the skill they actually have, until they can do the next idea unaided. A Lexile, a NAEP bar, an SAT Reading and Writing number, and an AI rewrite are not the path. Birthday is not placement. You will sometimes slow down. You will sometimes skip ahead. Both are teaching. If this week needs a compass, this is it. Five things. Then sit down and teach. # The English Hour The hour has a shape. Learn it once. Later chapters will give you today’s idea, today’s first word or sentence, and today’s wording. They will not rebuild this hour. When a chapter says “run the session,” it means this. You do not need a school bell. You need a beginning, a short model, a real attempt, one good question, mixed practice, and an unaided check. Younger children may finish in twenty-five minutes. Older students may need forty-five. The shape does not change. The idea on the table does. Sit down having already done the five-minute parent warm-up from *How to Use This Book This Week*. You know today’s sentence. The child has the pencil, or the pointer. You have the key. ## 1. Warm-up Three to five minutes. Words, sounds, and sentences they can already get right. Mixed, not a page of the same item. This is retrieval, not a test of character. Say: “We’re going to start with things you already know.” “Tap the sounds in *ship*. Now write it.” “Read me yesterday’s sentence.” If they already get these right untimed, you may use a brief check of pace on a passage they can already decode. Say: “You already know these words. Read it as it should sound.” If they do not already get them right, skip the timer. Time only what they can already produce. Fluency is accurate, reasonably paced, expressive reading of text they can already decode — not a race and not optional.[^13] You are building retrieval, not an identity. Cold decoding is not a race. Keep this short. The warm-up is not the lesson. When you feel yourself teaching a new pattern here, stop. That pattern belongs in the model. ## 2. Short model Five to eight minutes. One new move, visible. Then a second example with holes they fill. Then you stop. New, dense material — first look-through of a vowel team, first because-sentence, first claim with a reason — is where a short model earns its keep. A novice staring at a blank page spends working memory searching, not learning. Show the move. Then fade. Say: “Watch this one. I’ll look through the whole word out loud.” “I’m writing one sentence of the kind you’ll write. Listen to the because.” “Hear how this paragraph should sound. Then you’ll try it.” Then, on the second example: “Your turn to tell me what I write next. I’ll hold the pencil this time. You do the thinking.” “I’m leaving the last word blank. You look through it.” Include one incorrect example when the idea is new, and ask: “What did they think this word was? How can you tell?” That question is worth more than another correct copy. When they can already execute the move, stop showing full examples. Extra modeling after they can do it is noise. Fade: full example, completion, conventional problem. Then they work. The model is short on purpose. If you are still talking at minute twelve, you are giving a lecture. Close the model. Hand them the first word, the first sentence, or the first paragraph. ## 3. Student attempt This is the center of the hour. Eight to twelve minutes. Maybe longer for a paragraph or a short paper. The first problem is theirs. Hand them “For the student” if the chapter has it, or the first task from **How to teach it this week**. Then you talk less than you want to. Say: “This one is yours. I’ll be quiet.” Then be quiet. If they stall, use this order: ask, wait, hint, then model. Not the reverse. Ask: “Show me what you tried.” “Look through the word. What’s this part?” “Where is the because?” Wait. Count a slow ten in your head. The silence is the work. If you fill it, you took the problem back. Hint, one hint: “You know that vowel team. Say the sounds.” “Start the sentence with *Although* and finish it.” “Point to the sentence that gives the reason.” Then, if they are still stuck after a real try: “I’m going to show you this one step. Then you take it from here.” Show the step. Return the pencil. If they are missing a piece — they cannot hear the three sounds in *cat*, they cannot hold a kernel sentence — that is a hole, not useful struggle. Fill the hole with a short retrieval or a picture of the move, then return to the task. Struggle on a well-posed new idea, with the pieces in place, is the lesson. Struggle on missing prerequisites is just being lost. Keep your hands off their pencil. Taking it feels kind. It teaches them to watch. Finishing the paragraph for them feels kind too. Same lesson. If you feel the urge, put your hands in your lap and say the ask-sentence again. The child holds the pencil. You hold the key. When they are stuck on a word, the prompt is “look through the word,” not “look at the picture.” Pictures remain useful for meaning *after* the word is read.[^14] ## 4. One good question Two or three minutes. Not “did you get it?” They will say yes. You will learn nothing. A good question makes them generate the idea again, in words or in a second try. Say: “How did you know that was *ship* and not *shop*?” “Read me the sentence. Where is the because?” “What was this paragraph about? Two sentences. Book closed.” “Explain it so I could do the next one.” One question. Maybe a follow-up. Then stop. This is not a seminar. You are checking that they have a picture — of the word, of the sentence, of the page — not collecting a speech. If they cannot explain it back, they copied a step. Return to the move — look through the word, write the kernel, mark the claim — and give one more attempt. Copying is what a stuck novice does when the model was too fast or the attempt was too short. Slow the model tomorrow. Keep today’s pencil theirs. ## 5. Mixed practice Ten to fifteen minutes. First a short blocked set of the new move — four to six words, sentences, or short reads that use today’s idea — so they learn to execute. Then mix. Yesterday’s pattern. Last month’s word. An old sentence among new ones. The new move sitting next to an old one so they have to choose. Blocked practice teaches execution. Mixed practice teaches *when* to use the move. A page of twenty identical items feels like mastery at 9 p.m. It often is not still there in two weeks. Put today’s correspondence next to yesterday’s. Put an old sentence in this week’s paragraph. Wide reading of already-decodable text sits here. It is not the whole program.[^15] Say: “These first few are today’s move. Get the feel of it.” Then: “These are mixed on purpose. First decide what kind of job it is. Then choose.” “You already know this pattern from last week. Find it among the new ones.” Fluency work in this block is oral, repeated, with feedback, on text they can already decode. Knowledge lives in the rest of the day — history, science, literature read aloud above decoding level — not as a strategies workbook. Writing time here includes instruction: a modeled sentence, a shared sentence, then theirs. Extra writing alone, with no teaching in it, is a small effect. A journal is not the lesson.[^16] Keep practice short enough that attention holds. More pages are not more learning if the last ten are sloppy. Stop while they can still think. ## 6. Unaided exit ticket Three to five minutes. One short task. Book closed. No hints. No “For the student.” No chatbot. No looking at the worked example. This is the check. Tonight’s helped practice, with you in the chair and a tool in reach, is not this check.[^17] Say: “One task. Book closed. I won’t help. That’s the point.” “Wrong is useful. We’ll look after.” “You don’t need a perfect score. I need to see what you can do alone.” Read this sentence aloud. Spell this word. Write one sentence that starts with *Although*. Tell me in a sentence what this paragraph was about. Then you are quiet again. When they finish, you look. If it is right, you are done enough for today. If it is wrong, you diagnose — which wrong turn from the parent section is this? — and you note tomorrow’s first move. You do not redo the whole hour. You do not punish. You find the picture that was missing. An exit ticket that always needs you is not an exit ticket. Fade until it doesn’t. ## When to stop talking After you ask, wait. If you fill the silence, you took the problem back. During the attempt, your job is almost nothing. Presence, not narration. If you hear yourself explaining while they write, you have moved back into the model. Stop. Let them write. Let them look through the word. Instead of “this is the hard part” or “I was never good at grammar,” say today’s sentence: “Look through the word.” Or: “Show me the because.” The first sentence fills their working memory with your feeling. The second sentence teaches. Praise a clear look-through, a sentence they can reread, a catch of their own error — not speed. “You noticed *shop* didn’t match. That’s the reading I want.” Stop talking when the next sentence would be a rescue. Hands in your lap. Count ten. Then one hint, or one step, and the pencil goes back. If you remember only one move from this page, remember that one. ## The AI rules, once Later chapters will point here in one sentence. They will not reprint this box. > **Child first. You hold the key. AI is optional for the adult.** > > The child reads the page and writes the sentence before any tool is in the room. You keep the answer key — on paper, in a closed tab, in your head after the warm-up. A chatbot does not sit beside them during the attempt. > > Use a tool, if you use one at all, for four jobs: explain today’s idea *to you* before the lesson; make extra practice with the answers hidden from the child; write a hint script you can read from; diagnose work they already produced. > > Never paste the essay. Never say “fix this to an A.” Never leave an unsupervised chatbot running during the attempt. Never photograph the worksheet for a key. Never rewrite the passage so the child skips decoding. > > In a high-school math study, an unguarded chatbot raised assisted practice about 48 percent and then cut the unaided exam about 17 percent; a hint tutor that withheld full solutions raised practice without that exam drop. That study is mathematics, not a writing trial. The crutch still governs writing: if the model produces the sentence, the child is a spectator.[^18] > > Hermes 4 is a chat model. Hermes Agent is a different runtime. Grok the assistant is not Grok Bot. Grok Bot has no education product and is not a school-safe tutor. If you use any of these, they live on your machine or your account. They are not the child’s study buddy, and they are not a companion. > > Detectors fail often enough that they are not a verdict on a child’s page.[^19] > > You can teach this entire book with no AI. Many families will. The hour still has the same shape. If a product claims it “does 100 percent of the teaching,” treat that as advertising. You still need to hear a wrong turn. You still hold the key. ## A Tuesday, said plainly Here is an illustration, not a reported family. A parent has spent five minutes saying “look through the whole word” and writing *ship* with the three sounds marked. The child warms up on two words already mapped. The parent models one new word out loud, then leaves the last grapheme blank. The child looks through *shop*, slowly. The parent asks, “How did you know it wasn’t *ship*?” Mixed practice: two new words, one old sentence, one short read-aloud of a paragraph they can already decode. Exit ticket: “Read this sentence. I won’t help.” Right or wrong, the hour had a shape. Tomorrow’s first move is already visible. On a writing day the same shape holds. Warm-up: three spellings already taught. Model: one because-sentence, thinking aloud. Attempt: they write one. Question: “Where is the because?” Mixed: yesterday’s kernel next to today’s expansion. Exit: one unaided sentence, book closed, no rewriter in the room. You can run that hour. You do not need a degree. You need today’s sentence, a pencil you refuse to take, and the willingness to stop talking. Dose is not a 180-day script. Phonemic awareness is a short stretch of hours across a semester, with letters, not a year of oral games.[^20] Connected text every day once there are words they can decode. Fluency: a short oral-rehearsal piece, daily, on text they can already get right slowly. Knowledge: the rest of the day. Spelling, if you set a clock, is a short daily block or three longer ones in a week — professional judgment, not a trial.[^21] The hour is this shape. The minutes follow the child. Later chapters will say: run the session as in *The English Hour*. That sentence is this page. The idea will change. The hour will not. # Chapter 1 ## Sounds, letters, and first sentences Wooden letter tiles, a blank index card, and a yellow pencil on a kitchen table. ## Why this matters This year is not a waiting room until “real reading” begins. The work of these years is the code of written English: hearing the sounds inside spoken words, mapping those sounds onto letters, looking through print instead of around it, forming letters by hand, talking in full sentences, and writing the first sentences those letters can hold. The sentence that surprises people is this one: talking about a story is not the same as reading it. Philip Gough and William Tunmer put reading as a product. Reading equals decoding times linguistic comprehension, or (R = D C), each factor running from zero (nothing) to one (complete).[^22] Linguistic comprehension here is listening comprehension — understanding language by ear — not “reading comprehension” as a test score. If either factor is zero, reading is zero. A child who can retell a picture-book plot but cannot map the print to sound is not yet reading that book. A child who can sound out every word and does not know the language of the page is not reading it either. Teach both. You do not wait for one to finish before starting the other. Hollis Scarborough later unpacked those two factors as a rope: three strands of word recognition twisting with five of language comprehension.[^23] Fluency is not a ninth strand. It is a sign that the rope is beginning to run. An hour that only drills sounds, or only talks about books, is teaching a frayed rope. Oral language and first sentences keep the language side in the room. What you are teaching is phonemic awareness finishing its job; a planned phonics sequence used in reading and writing; decoding of connected text; handwriting that is print in this band; oral language in full sentences; and the first written sentences, with you modeling and writing together before they write alone. Grade labels are sequence anchors, not birthdays. Guessing from pictures is not reading. A ten-year-old who still guesses may sit here. Treat the Common Core English map as a map, not a homeschool mandate. Many families do not use it. Texas never adopted it as TEKS. Florida’s B.E.S.T. introduction presents itself as eliminating Common Core. Virginia’s 2024 standards are a separate document. What the map is good for is saying what the English *is*. The document’s own limit: “The Standards define what all students are expected to know and be able to do, not how teachers should teach.”[^24] Foundational Skills — print concepts, phonological awareness, phonics and word recognition, fluency — run kindergarten through Grade 5. Phonological awareness as a heading stops after Grade 1. Those skills “are not an end in and of themselves; rather, they are necessary and important components of an effective, comprehensive reading program.”[^25] Texas names encoding and decoding as reciprocal.[^26] Virginia keeps phonemic awareness through Grade 2 and starts a transition to cursive then.[^27] Florida keeps phonological awareness in the regular progression through Grade 1, keeps Grade 2 handwriting as print, and is explicit that many high-frequency words are irregular or only *temporarily* irregular — the decodable ones belong with phonics.[^28] There is no national cursive law. The Common Core PDF does not use the word *cursive*. Your Tuesday is not a statute. The National Reading Panel, from thirty-eight studies, found systematic phonics a faster start than instruction that taught little or no phonics. Linnea Ehri and colleagues’ published account of that meta-analysis reports an overall mean of (d = +0.41); for kindergarten and first grade, (d = 0.55).[^29] This is a useful study, not a promise that every home will see the same result. Then teach. Phonics is a means. The Panel itself warned that letter-sound work that never gets used in reading and writing is unlikely to be very effective. For older children, phonics still helped decoding; it did not, in that analysis, raise later comprehension on its own. The code is the on-ramp. Knowledge, vocabulary, and language are the road. Phonemic awareness instruction moves reading and spelling when it is brief, focused, and tied to letters. Ehri, Nunes, Willows, Schuster, Yaghoub-Zadeh, and Shanahan, reporting the Panel’s phonemic-awareness meta-analysis, found (d = 0.86) on phonemic awareness itself, (d = 0.53) on reading, and (d = 0.59) on spelling. Effects were larger when awareness was taught with letters, when one or two skills were taught rather than many, and when instruction lasted between five and eighteen hours rather than longer.[^30] A year of oral sound games with no letters is not what the Panel found. Fifteen minutes with letter cards, until the child can segment a one-syllable word with ease, is closer to the shape. Words become sight words by orthographic mapping, not by guessing from pictures. Ehri describes four overlapping phases: pre-alphabetic (a logo, a color, a picture cue), partial alphabetic (the first letter, maybe the last), full alphabetic (complete letter–sound connections), and consolidated alphabetic (chunks, syllables, morphemes).[^31] A sight word, in that account, is a word already mapped: the letters connected to the sounds, stored in memory, read at a glance because the mapping was done. “All words when practiced become read from memory by sight.” Irregular words still have mappable parts. *Said* has /s/ and /d/. The Institute of Education Sciences K–3 guide discourages guessing unknown words from beginning letters or pictures.[^32] When the child is stuck, the prompt is look through the word. Ohio Revised Code § 3313.6028 and Louisiana Act 517 have withdrawn three-cueing as a method of identifying words.[^33] Pictures, syntax, and meaning remain how a reader *checks* a decoded word. They are not how the reader *finds* it. Teach through the print. Invented spelling is a stage, not a destination. Let a six-year-old write *KAT* while you are teaching /k/ /a/ /t/. Hold *cat* to standard spelling once those three letters have been taught.[^34] English spelling is a morphophonemic code — it maps to sound and to meaningful parts — not a chaotic list. That sentence is planted here. The next chapter owns the Latin and the syllables. What this band unlocks is fluency as a bridge, reading to learn, and written paragraphs that can be revised. Skip the code, and later “comprehension strategies” arrive as a bag of tricks with no print underneath. Skip oral language, and decoding lands on words the child does not yet own by ear. Skip handwriting and first sentences, and composition in grade 4 has nowhere to sit. You do not need to be an English specialist. You do need to hear “kitten” for the printed word *cat* as a picture-guess, not as reading, and *KAT* after *c-a-t* has been taught as a stage that should be receding. ## For the parent: understand it yourself Many adults feel rusty on how English writing actually works. That is ordinary. Five minutes of this section, then the warm-up at the end, is enough for tomorrow. **Everyday picture.** Three taps on the table. The spoken word *cat* is three sounds: /k/ /a/ /t/. Three taps, then three letter cards in a row, then the blended word the child already knows by ear. The drawing of a cat on the facing page is not the word. The word is the print. If you cover the drawing and the child can still say *cat* from *c-a-t*, they are reading the word. If you cover the print and they say *cat* from the drawing, they are naming a picture. **Precise picture.** A phoneme is the smallest unit of spoken language. *Go* has two. *She* has two. *Cat* has three. *Ship* has three: /sh/ /i/ /p/. Letters and letter teams — graphemes — map onto those sounds. Decoding is looking left to right through the print, saying the sounds, and blending them into a spoken word. Encoding is the reverse: hearing the sounds and writing the letters. Texas is right that those two are reciprocal.[^35] A ten-minute sitting that only “sounds out” and never writes the word is half the move. A sitting that only copies a word and never says the sounds is the other half. Gough and Tunmer’s product is the precise picture of the whole act. Decoding without language is noise. Language without decoding is listening. Hoover and Gough later tested the view with English–Spanish bilingual children in grades 1–4: the product improved prediction beyond simply adding the two.[^36] You need the hearing: a fluent retell of a read-aloud is language. A finger under *sat*, sounds blended, is decoding. Reading is both in the same sitting. Phonemic awareness is the hearing of those units. It is not phonics. Phonics is using letter–sound maps in reading and writing. Awareness that uses letter cards to blend or segment overlaps with phonics, and that overlap is the useful one. The Panel distinguished both from auditory discrimination (same word or different word). If the child cannot yet hear that *cat* and *hat* differ by one sound, you are not ready to expect independent decoding of a page of *at* words. If the child can already fully segment and blend with letters, more oral sound games are not what Ehri and colleagues found. Timothy Shanahan, writing six years after the Panel as a member of it, put the duration in household terms: the amount associated with the greatest reading gains was between five and eighteen hours; about fifteen minutes a day for a semester of kindergarten, adjusted diagnostically, is a wise shape.[^37] Segmenting and blending gave the greatest reading advantage. Continue until single-syllable words can be fully segmented with ease. Phonics in this band is a planned sequence, taught explicitly, practiced in connected text. Grade 1 on the Common Core map: common consonant digraphs (*sh, ch, th, wh*); regularly spelled one-syllable words; final silent *e* and common vowel teams; the idea that every syllable has a vowel sound; two-syllable words following basic patterns; inflectional endings (*-s, -ing, -ed*); grade-appropriate irregulars.[^38] Grade 2: more vowel teams; two-syllable words with long vowels; common prefixes and suffixes; inconsistent but common correspondences; more irregulars. Texas names the syllable types in Grade 1: closed, open, vowel-consonant-*e*, vowel teams including diphthongs, *r*-controlled.[^39] You do not need those names on a quiz. You do need a sequence. Synthetic phonics (letters to sounds, then blend) and analytic phonics (analyze sounds inside known words) were both effective in the Panel’s set; neither significantly outperformed the other. What mattered was a plan, not a personality. High-frequency words belong in that plan, mapped, not parked on the fridge as unmappable blobs. Florida’s own clarification is the household rule: many such words are irregular or only temporarily irregular; the decodable ones should be introduced with phonics.[^40] *The* has mappable parts. *Said* has mappable parts. The handful that must be introduced as wholes before a text — the IES guide allows a few essential ones — are still looked through, not guessed from a drawing.[^41] Handwriting this year is print. Print all upper- and lowercase letters. Write words and sentences that a stranger can read. Cursive, if you start it, is a state-map choice, not a national year. The next chapter will pick it up where Texas, Virginia, and Florida actually place it. Steve Graham and colleagues, in an elementary writing meta-analysis, found that teaching transcription skills — handwriting, spelling, sentence construction — had an effect size of 0.55 on writing quality.[^42] That is why the pencil stays in the child’s hand. A parent who forms the letters “so we can go faster” is doing the transcription. The child is a spectator. Oral language is not an extra period. The National Early Literacy Panel, as summarized by Shanahan and Christopher Lonigan, found simple vocabulary tests a modest predictor of later reading (about (r = .33)). Complex oral language — grammar, defining words, listening comprehension — correlated (r = .70) with later reading comprehension.[^43] Talk in full sentences. Read aloud books above the child’s decoding level. The read-aloud is the language-comprehension hour. The decodable page is the decoding hour. Both are English. First sentences are taught, not assigned. The maps already want opinion, information, and story pieces in Grade 1, with guidance: a topic, some development, a sense of closure.[^44] Model one sentence out loud as you write it. Write a second together. Then they write one alone. Name a capital letter and a period in the sentence on the table. That is already the later finding that conventions earn their keep in the sentence being written, not on a worksheet island.[^45] **Wrong answers you should be able to hear** 1. *The printed word is* cat*. The child looks at the drawing and says “kitten,” or “puppy.”* That is picture-naming. It is not reading. Cover the drawing. Put the finger under the letters. “Look through the word. Say the sounds.” If they can do it with the picture gone, the code is at work. If they cannot, the picture was doing the job. 2. *The printed word is* house*. The child says “horse” from the first letter and a drawing of a building, or from a memory of the book.* Partial alphabetic guessing. Ehri’s middle phase: salient letters plus context, not yet checked against the whole spelling. The IES panel’s discouraged move. The next prompt is still through the word: /h/ /ou/ /s/. The picture may confirm *after* the blend. 3. *You ask for the sounds in* sat*. The child recites letter names: “ess-ay-tee.”* Letter names are useful. They are not phonemes. Say: “The sounds, not the names. Watch my mouth: /s/ /a/ /t/.” Then they tap. Then they write. 4. *The child writes* KAT\* for *cat* in October, then still writes *KAT* in May after *c-a-t* has been taught and practiced.\* Invented spelling as a stage is expected early. Invented spelling as a destination, for a pattern already taught, is the miss. Hold the taught pattern to standard spelling in writing. Leave untaught patterns as invented spelling a little longer. Louisa Moats, citing Paula Mehta and colleagues on a Houston grades 1–4 sample, is the warning that spelling will not ride along by itself: passage comprehension stayed near the middle while spelling percentiles dropped.[^46] 5. *The child “reads” a patterned book by reciting it from memory, eyes off the print, or by using the refrain and the pictures, and cannot point to* the\* when you ask.\* Memorized performance. Print concepts are still the job: a sentence has a first word, a capital, an ending mark; a word is a letter sequence with space on either side.[^47] Finger under the word. One-to-one pointing. A short decodable sentence with the same words in a new order is the check. A sixth you will also hear: every unfamiliar word is skipped, or replaced with “whatever,” while the child talks about the picture. That is the three-cueing prompt arriving as a habit. The statutes named above withdrew it as a method of finding the word. Your sentence is still: “Look through the word.” Then wait. **Five-minute parent warm-up** Do this before the lesson, on a scrap of paper and three letter cards, no child in the room. Minute 1. Say *cat* slowly. Tap three times. Write *c a t* with a space between letters. Blend it back. Say out loud: “Three sounds. Three letters. The picture is not the word.” Minute 2. Write *said*. Underline *s* and *d*. Say: “The mappable parts are here. The middle is the part we learn as a pattern.” Write *the*. Same move. You are installing mapping, not a blob list. Minute 3. Write *cat* on a card. Imagine a drawing of a kitten next to it. Decide, without performing, what “kitten” would mean if the child said it. Decide what covering the drawing and waiting would mean. You are not taking a test. You are installing the miss so you can hear it tomorrow. Minute 4. Write one short sentence: *The cat sat.* Point under each word. Notice the capital and the period. That sentence is tomorrow’s connected text, not a paragraph. Minute 5. Write the sentence you will actually say: “Look through the word. Say the sounds.” Put the pencil down. That sentence is the lesson. If you can do those five minutes, you are ready to sit down. The child looks through the print. You hear. ## How to teach it this week A good English hour this week has a shape. The front of this book’s English-hour chapter is the full version, including the AI rules. Here is the shape scaled to this band. **Warm-up (3–5 minutes, unaided).** Two or three words the child can already map. One short sentence using those words, finger under print. Paper. No device. No picture doing the work. **Short model (5 minutes).** One new move, one picture of the move, one written word or sentence. A new digraph on cards, then in a word, then in a sentence. You talk for a few minutes. Then you stop. **Student attempt (10–15 minutes).** Three to five words, then one or two sentences. The child points and writes. You wait. Struggle before rescue: ask, wait, hint, then model. Leave the pencil, and the pointer, in the child’s hand. **One good question.** Not “what does the picture tell you?” A question that names the idea: “How many sounds?” “Show me the letters that make /sh/.” “Look through the word — what do you hear at the end?” **Mixed practice (5–10 minutes).** Yesterday’s pattern next to today’s. An older short-vowel word next to a new silent-*e* word. A taught high-frequency word in a new sentence. If the book is almost entirely one pattern — most early readers are — you pull one older word onto the page. **Exit ticket (5 minutes, unaided).** Two words and one sentence isomorphic to what you taught. Phone in another room. You listen against a key you hold. The child reads. The child writes one of the words from dictation. That shape is a practice you impose on whatever book is on the table. It is not a 180-day plan. The IES K–3 guide wants connected text every day in this band, and it wants the child reading words, not guessing them.[^48] **Exact wording you can say** On sounds: “Tap the sounds. Not the letter names — the sounds.” “How many sounds in *ship*? Watch: /sh/ /i/ /p/. Three.” “Say it slowly. Now say it fast.” On mapping: “Look through the word. Start at the first letter. Say the sounds. Blend them.” “Show me the letters that make /ch/.” “The middle of *said* is the part we learn. The /s/ and the /d/ you already know.” On a stuck word: “Your pointer. Look through the word.” “Cover the picture. What do the letters say?” “You said *horse*. Check the rest of the word. What is the last sound?” On writing: “Hear the word. Say the sounds. Write the letters.” “That pattern we already learned. In our writing we spell it this way.” “Your pencil. I’ll ask a question.” On a first sentence: “A sentence starts with a capital and ends with a mark. You write it. I’ll wait.” “Tell me the sentence out loud. Now write the first word.” When you are about to take over: “Your pointer. Your pencil. I’ll listen.” Then wait. Ten seconds of silence is teaching. Taking the pencil is the high-help path. Preparation, then generation. You cannot “help less often” as a daily homeschool strategy if you have not prepared. Prepare so you can hear “kitten” for *cat*. Then leave the pencil. **First problem for the student** Write this on a small card, exactly, no picture: sat Say: “Look through the word. Say the sounds. Blend them.” Wait. If the child says *sit*, they used the first and last letters and guessed the middle. If they recite letter names, they have not yet switched to sounds. If they say *sat*, they looked through it. If they stare, show a fully worked different word — *mat*, said as /m/ /a/ /t/, blended — then a second word they complete with you, then return to *sat*. That is fading: full example, then a completion, then the item.[^49] Later the same week, the diagnostic item: A sentence on a strip, a small drawing at the side that does *not* match a guess: The cat sat. If they say the sentence with their eyes on the drawing, cover the drawing and ask them to point under each word. If they can, the code is in use. If they cannot, the drawing was doing the reading. After the child has attempted, you may show a worked *incorrect* example: a card that shows *house* with a drawing of a horse, and a speech bubble that says “horse.” Ask what that person used to find the word. Generation first, then the named error. The child names the miss. You do not narrate it as a verdict on talent. **How to fade help** Day of a new correspondence: you model one word fully. The child explains a second. A third is missing only the last sound. A fourth is conventional. When the child is accurate, stop showing full examples. Bring the letter cards back when a new team arrives, or when two-syllable words first appear. Fading is for accuracy, not for a calendar date. **When to stop talking** After the question. After the hint. After the model of a *different* word. If you hear yourself sounding out the child’s current word while their pointer is idle, you have started doing the work. Stop. Point to the first letter. Wait. Print concepts sit in the same week. A sentence has a first word, a capital, an ending mark. Words have spaces. Grade 1 finishes those sentence features.[^50] If the child is still sweeping a finger under a whole line while reciting, restore one-to-one pointing on a short sentence. Two-syllable words follow basic patterns: *napkin, robot, cupcake.* Every syllable has a vowel sound. Look through each syllable the way you looked through *sat*. Prefixes, suffixes, and inflectional endings are chunks you still look through: *jump, jumps, jumped, jumping.* Irregulars are not a separate religion. A short daily handful, mapped as far as they will map, mixed with decodable words, is enough. Texas asks for at least a hundred high-frequency words from a research-based list in Grade 1; Florida wants automaticity with the clarification already quoted.[^51] The list is not the method. Mapping is the method. A child who recites *The cat sat* from the pattern of yesterday’s book has a memorized performance. A child who points, maps, and then can say what happened has the product in miniature. Your job is to hear which. ## Practice that actually builds learning **Blocked, for a new move.** The day you introduce /sh/, give a short set that is only that: *ship, shop, fish, wish*, cards then writing then one sentence. The day you introduce silent *e*, a short set of pairs: *hat/hate, kit/kite, hop/hope.* The day you introduce pointing under a sentence, a few sentences using only known words, pictures covered or absent. **Mixed, for when to use it.** Later the same week, the new move reappears between unlike words: a short-vowel word, a silent-*e* word, a taught irregular, a two-syllable word you have already mapped. Mixing is the practice of choosing a map.[^52] A purchased reader that is almost entirely one pattern can still be mixed by you pulling yesterday’s word onto the same page. **Brief retrieval of words already mapped.** Two or three minutes, mixed, after the child can read those words correctly untimed. Cover any pictures. You listen. Overlearning a word list until tonight’s score is perfect has limited additional benefit compared with meeting those words again next week in a new sentence. Retrieval is a second meeting, not a race. **One incorrect example to diagnose.** After the child has attempted *house*, show a card that guessed “horse.” Ask: “What did this person use to find the word?” After a letter-name recitation, show *s-a-t* said as names and ask for the sounds instead. The child names the miss. **Dictation.** You say a taught word. They write it. Then a short taught sentence. Encoding is the check that the map runs both ways. Invented spelling is allowed for sounds not yet taught. Taught patterns are held to standard spelling in the dictation book. **Connected text every day.** A sentence, then a short decodable page, then a slightly longer page as the code grows. Guided oral reading with feedback already belongs in miniature: the same short sentence a second time, then a third for smoothness.[^53] Fluency as a named bridge is the next chapter. Silent independent reading as the *only* method is not what the Panel could causally support for children who have not yet got word reading. Wide reading of rich books *to* the child is the language side, and it is not the same sitting as the decodable page. **Kitchen, talk, and making.** Talk in full sentences about things in the world. Retell a short event with a beginning and an end. That talk is complex oral language, not a quiz of ten rare words. A grocery list in invented spelling is useful encoding. It does not replace dictation of taught patterns. **First sentences this week.** One modeled sentence. One shared sentence. One independent sentence. Capital letter, spaces, ending mark. The content can be a true opinion (*I like the cat.*), a fact (*The cat sat.*), or a tiny narrative (*The cat sat. Then it ran.*). You are not assigning a paragraph. You are teaching a sentence the child can hold. **Read-aloud above decoding level.** Ten or fifteen minutes, you reading, they talking. New words live here. Syntax lives here. The code hour does not have to carry the whole language load. Scarborough’s language-comprehension strands grow in this sitting while the word-recognition strands grow on the decodable page.[^54] A weekly shape: day of a new correspondence, a short blocked set. Same week: mixed retrieval of older words. Later this week and next: the new correspondence between unlike words, and in a sentence. Daily: connected text, even if it is three lines. You impose that shape on the book you already have. ## For the student You are learning how print works. Letters stand for sounds. You look through the word. You say the sounds. You blend them. That is reading a word. A picture can be interesting. A picture is not the word. Three taps. The word *cat* is /k/ /a/ /t/. Three sounds. Three letters. You can write it. You can read it with the picture covered. **Tiny worked example** Look at this word, with no picture: sat Start at the first letter. The first sound is /s/. The next is /a/. The last is /t/. Say them slowly: /s/ /a/ /t/. Now say them fast: *sat*. Check: cover anything except the letters. The letters still say *sat*. That is looking through the word. **Two tries** 1. Look through this word. Say the sounds. Blend them. map 2. Look through this word. Say the sounds. Blend them. Then write the word on your paper. ship For the second one, /sh/ is two letters that make one sound. Then /i/. Then /p/. Write your answers. Check each one by putting your finger under the letters and saying the sounds again. **Explain it back** Tell someone at the table, in your own words, how you read a new word. Then point to *sat* and say the three sounds. Then point to a picture in a book and say whether the picture is the word or not, and why. **Challenge** The cat sat. Some people look at a drawing and say a word that fits the drawing. Some people look at the first letter and guess. What would those two moves mean? How do you look through each word instead? Read the sentence with your finger under the print. Then tell what happened, in a sentence of your own. You are allowed to struggle. You may use letter cards or a tap on the table. You point. You write. If you get stuck, ask for a hint — not the word. Then try again. A sight word is a word you have already mapped, so you can read it quickly. It is not a word you guessed from a picture. *Said* still has /s/ and /d/. You can catch the parts. Then you remember the middle. When you write, a sentence starts with a capital letter and ends with a mark. You hold the pencil. One sentence is enough for today. ## If it isn’t clicking Three diagnostics. Each one has a next move. None of them is a verdict on talent, and none of them is a reason to wait for a birthday. **1. The child cannot hear that *cat* is three sounds, or cannot blend /s/ /a/ /t/ into *sat*.** Phonemic awareness is not yet installed. Next move: smaller work, with letters. Two-phoneme words first (*go, at, me*), then three. One or two skills — blend and segment — not a circus of substitutions, deletions, and pig-latin. Letter cards under the sounds. Short daily sittings. Five to eighteen hours is the research window, not a year of sound games with nothing to look at.[^55] If this is still the bottleneck after several weeks of short daily work, stay here. Phonics worksheets on top of a child who cannot hear the units will not hold. A human tutor who will sit with sounds and letters, not with pictures to guess, is a reasonable next step if you have tried the small-set work and the blend still does not come. **2. The child guesses from pictures, or from the first letter, after you have already taught looking through the word.** The cue is still the drawing, or the first-letter-plus-context habit. Next move: cover the pictures. Short decodable sentences. Finger under each word. The prompt is always through the print. After an attempt, one incorrect example to explain (“this person said *horse* for *house* — what did they use?”). Slow down the “interesting” books until the map is hearable; a child who can recite a patterned reader and cannot point to *the* is not ready to skip this. Go ahead once they can read a short new sentence of known patterns with the pictures covered. A tutor is useful if the guess remains the default after a couple of weeks of daily covered-picture work *and* the hour has become a fight. **3. Accurate decoding is not growing, even after systematic teaching of the code, and the difficulty is unexpected given what the child understands by ear.** This chapter does not diagnose anyone. The International Dyslexia Association’s 2002 definition describes a specific learning disability, neurobiological in origin, marked by accurate and/or fluent word-recognition difficulties and poor spelling and decoding, typically from a phonological-language deficit, unexpected given other abilities and effective instruction.[^56] Gough and Tunmer locate the proximal problem, in the simple view, as an inability to decode.[^57] The first move remains systematic teaching of the code. The IES guide notes that students who struggle persistently with phonological awareness often benefit from one-on-one or small-group intervention, and that early intervention can often remedy a phonological core deficit that otherwise may lead to the single-word decoding difficulty that is a hallmark of reading disabilities.[^58] Next move: keep the code work daily, smaller steps, more mapping, less picture. Get a specialist when the difficulty is unexpected, persistent after that teaching, and interfering with work the child can understand by ear. A specialist is a next step, not a verdict on the sitting. **When to slow down.** Phonemic awareness missing. Picture-guess still the default. Letter-sound maps unused in connected text. Handwriting so effortful that a sentence cannot be finished. Those are brakes. “Not ready” because of age is the brake this book will not use. The Panel rejected the claim that kindergarten children are not ready for phonics if the teaching is appropriate.[^59] “Not ready” because the prerequisite is missing is a real brake. **When to go ahead.** Segmenting and blending with letters is easy. The child looks through the word when stuck. Short decodable sentences are accurate with pictures covered. Print is legible. A sentence can be written from dictation using taught patterns. Oral language is in full sentences, and a read-aloud above decoding level is a normal part of the day. Then the next chapter’s fluency, multisyllabic words, and paragraphs have somewhere to sit. Being “good at books” because a child tells wonderful stories from pictures is not a reason to skip the code. **When to get a human tutor.** You have run the small-set sound work, or the covered-picture decoding work, or the daily dictation of taught patterns, for a stretch of daily sittings, and the same diagnosis is still the one in the room, and the hour has become a fight. A tutor is a release valve, not a failure of the sitting. Keep yourself as the person who can still hear “kitten” for *cat* and *KAT* after the pattern was taught. Outsourcing the hearing is the thing to avoid, not asking for help. A child who can tap sounds and will not look at print is missing one strand. A child who can recite letter–sound chants and cannot understand a read-aloud is missing another. Hearing which is thin is enough. If the child is missing the sounds, model tapping a two-phoneme word, then guided practice, then independent. If the child has the sounds and the task is a first silent-*e* pair, a short generation with what they already know, then a clear model, is the second sequence. ## Tools, including AI Optional helpers for you, the adult. The full rules live in the English-hour box in the front of this book. This chapter is a pointer, not a second policy paper. The short version: the child attempts first. You hold the answer key. A tool may explain today’s idea *to you*, make extra word lists and sentences with answers on a separate page you keep, write a hint script you read aloud, or help you diagnose work the child has already done. The child looks through print on paper. The child writes on paper. Leave these out of the hour: pasting the page and taking the words; “what’s the word”; photographing a worksheet for a key; an unsupervised chatbot during the attempt; rewriting the passage so the child does not have to decode it. Text-to-speech is a documented help for some students with reading disabilities, as a way to *access* a chapter they cannot yet decode. Sarah Wood, Jennifer Moxley, Elizabeth Tighe, and Richard Wagner’s meta-analysis found an average weighted ({d} = 0.35) on comprehension for students with reading disabilities.[^60] That is compensation, not a phonics program. TTS that removes decoding is why it can help a documented print disability reach a science chapter, and why it does not teach the code. A language model will happily read the sentence for them, or rewrite it simpler. Performance during that kind of help is not the same as learning. The unaided word on the card tomorrow morning is the test.[^61] Unguarded solvers in a high-school math study raised assisted practice and then cut the unaided exam; that paper is mathematics, not a reading trial. The crutch is still the sitting condition. If the tool produces the word, the child is a spectator.[^62] Letter cards, a pointer, a pencil, and a strip of paper with no picture are tools too. Use them, then fade them. ## What “done enough” looks like Placement is by skill, not birthday. A “grade 1” reader is a publisher’s scope, not a legal grade. All About Reading, Logic of English, and a well-used older phonics series can place the same child in three different rooms, because they cut the grain differently.[^63] Placement tests exist because the labels lie. Choose by fit. This book names those programs as options. It does not rank them. **Checklist before moving on** - The child can segment and blend a one-syllable word with letters, including a consonant blend (*stop, hand*) and a digraph (*ship, chat*). - When stuck, the child looks through the word. Guessing from a picture, or from the first letter plus a drawing, has receded as the way to *find* the word. - Regular one-syllable words with taught patterns are decoded in isolation and in a short sentence. Silent *e* and the vowel teams you have taught are in use. Two-syllable basic patterns can be looked through syllable by syllable. - Taught high-frequency words are mapped as far as they will map and recognized quickly. The fridge is not a wall of unanalyzed shapes. - Invented spelling is receding for taught patterns. Dictation of those patterns is mostly standard. Untaught patterns may still be invented. - Print is legible: all letters, words in a sentence, a capital, an ending mark. Cursive, if you have started it, is extra in this band, not a national requirement. - The child can write one independent sentence after a model and a shared sentence. You did not hold the pencil. - Oral language is in full sentences. A read-aloud above decoding level is a normal part of the week. The child can retell a short heard story with a beginning and an end. - Connected text is read every day, even if it is five lines. Guided oral rereading of a short sentence is ordinary, not a race. - You can hear a picture-guess and a letter-name recitation, and you can ask a good question, without taking the pointer. If most of that list is true, go on to fluency as a bridge, multisyllabic words, and paragraphs, even if the birthday says otherwise. If the birthday says “third grade” and the pictures are still doing the reading, stay. The next chapter is the bridge. A path through English is the promise. A Lexile is not. # Chapter 2 ## Fluency, knowledge, and paragraphs An open children’s book, a handwritten page, and a yellow pencil on a wooden table. ## Why this matters Last chapter installed looking through the word. Keep that. This chapter is the bridge: from sounding out a sentence to reading in order to learn, and from a first sentence to a paragraph you can revise. The sentence that surprises people is this one: a child who can decode every word on a page may still not understand the page, and a child who “gets” a story when you read it aloud may still stumble when the same language is in print. Gough and Tunmer’s product still holds. Reading is decoding times linguistic comprehension. Fluency is what lets those two run together without using up the whole of working memory on the next word.[^64] The National Reading Panel found that guided repeated oral reading, with feedback from a teacher, a peer, or a parent, had a significant positive impact on word recognition, fluency, and comprehension.[^65] Timothy Rasinski later named the miss you will hear at the table: emphasizing speed at the expense of expression is “a corruption of the concept of reading fluency.”[^66] This is not a race. It is also not optional. Foundational phonics is not finished on the first day of a “grade 3” book. All four maps this book uses — Common Core English, Texas TEKS, Virginia 2024, Florida B.E.S.T. — still specify multisyllabic decoding, morphology, and fluency through Grade 5. Grade 5 is the last Common Core grade with a Foundational Skills strand.[^67] After that, the map assumes the code is in use. If it is not, a sixth-grader is not “behind in literature.” They are still in a code program. Last chapter’s move does not expire because a catalog printed a 3 on the cover. What you are teaching, as English, is fluent reading of prose and poetry the child can already decode; unfamiliar multisyllabic words taken apart by letters, syllables, and meaningful parts; paragraphs of opinion, information, and story; spelling that maps to print, not only to a weekly list; and knowledge and vocabulary feeding comprehension. Keyboarding minima appear on the Common Core map: one page in a sitting by Grade 4, two pages by Grade 5.[^68] Drawing evidence from texts — Writing standard 9 — begins in Grade 4.[^69] Cursive becomes usable on Texas, Virginia, and Florida maps in this band. The Common Core PDF is silent on cursive. There is still no national year. Daniel Willingham’s account is the one to keep in the kitchen: background knowledge speeds inference, chunking, and memory for new information.[^70] Donna Recht and Lauren Leslie’s baseball study, as Willingham describes it, is the picture. Junior-high students who knew baseball understood a baseball passage better than “good readers” who did not, even when the passage was written at an early fifth-grade level.[^71] Strategy instruction has a place and a ceiling. Willingham and Gail Lovette put it as a bag of tricks, quickly learned, not requiring a lot of practice; a handful of sessions may confer the same benefit as fifty. “When it comes to improving reading comprehension, strategy instruction may have an upper limit, but building background knowledge does not.”[^72] The Panel found evidence for seven strategies and stronger effects when several were taught together. Shanahan, reading the same studies as a Panel member, is more bullish on the weeks those studies actually used.[^73] Keep both lanes. Teach a strategy briefly, in the text on the table. Spend the year on knowledge and words. A content-free “reading strategies” hour is not the path. English spelling is a morphophonemic code. Richard Venezky mapped spelling first onto an intermediate level of sound-and-meaning units, then into sound.[^74] Louisa Moats, citing Hanna, Hanna, Hodges, and Rudorf, reports about half of English words spelled from sound-symbol correspondences alone, another 34 percent with one error on that basis (84 percent mostly predictable), and about four percent truly irregular.[^75] Invented spelling, useful in the last chapter, recedes as studied patterns are held to standard spelling. A fourth-grader still writing *KAT* for a word already taught is not being creative. They are missing the map. Grade labels remain sequence anchors. A child may sit this band at calendar grade 2 or calendar grade 6. Move a child who has last chapter’s code, pencil, and sentence. Hold the bridge for a child who does not. Skipping fluency because a child is “a good reader” of patterned books is the miss. Skipping knowledge because a workbook of main-idea circles is easier to schedule is the other miss. You do not need to be an English specialist. You do need to hear a racing, expressionless minute as a fluency problem, not as success, and “I don’t get it” on a baseball paragraph as a knowledge problem, not as a missing strategy worksheet. ## For the parent: understand it yourself Many adults feel rusty on fluency, on why a “simple” paragraph is hard, and on why English spelling looks unfair. Five minutes at the end of this section is enough for tomorrow. Last chapter installed three sounds, three letters, look through the word. Keep all three. **Everyday picture.** A short paragraph about a steal of home. If you know baseball, you see the runner, the pitcher, the catcher, the risk. If you do not, the words may all be readable — *lead, pick-off, slide* — and the play still does not happen in your head. The paragraph was not written in a secret code. The knowledge was the missing half of Gough and Tunmer’s product. Recht and Leslie put that picture in a study. You only need it as a hearing: when the child can say every word and cannot say what happened, ask what they know about the world of the page before you hunt a new “comprehension skill.” **Precise picture.** Fluency is accurate, reasonably paced, expressive reading of text the child can already decode. Accuracy is looking through the words. Rate is not a contest; it is enough speed that the sentence hangs together in memory. Expression — Rasinski’s prosody — is the music that shows the child is parsing phrases, not barking words. Guided oral reading is the method the Panel could support: the child reads a short passage aloud, more than once, with feedback. Independent silent reading as the *only* method is not what the Panel could causally support. Wide reading still matters as a way to meet words and worlds. It is not a substitute for the sitting in which you listen. Multisyllabic words in this band are taken apart, not guessed. Grade 3 on the map: common prefixes and derivational suffixes, common Latin suffixes, multisyllable words, grade-appropriate irregulars.[^76] Grades 4–5: combined letter-sound knowledge, syllabication patterns, and morphology — roots and affixes — to read unfamiliar multisyllabic words in and out of context. *Unhappiness* is *un- + happy + -ness*. *Transportation* is *trans + port + ation*. The child looks through the parts the way they once looked through *sat*. Florida puts Greek and Latin morphology in the vocabulary strand from Grade 3.[^77] Texas Grade 4 still has a “beginning reading and writing” expectation covering syllable types, syllable division, prefixes, suffixes, and matching spelling.[^78] Virginia drops the phonemic-awareness heading after Grade 2 and keeps phonics and word analysis through Grade 5.[^79] RF is not finished. Knowledge and vocabulary are the comprehension engine. Willingham’s mechanisms are ordinary: you infer what a sentence does not say; you chunk new information onto what you already have; you remember more of a page about a world you already know. Direct teaching of a few words, in the text, beats copying definitions. The Panel, as Shanahan reports, found vocabulary instruction improved comprehension; copying dictionary definitions was repeatedly the weaker comparison; words should be taught both directly and indirectly.[^80] Isabel Beck, Margaret McKeown, and Linda Kucan’s three tiers, as Shanahan presents them, are a useful household cut: everyday words the child already owns; the high-utility academic words worth teaching thoroughly; the rare domain words that come with the science or history chapter. Successful programs, in that account, teach a few hundred words thoroughly, not a thousand on a quiz.[^81] Morphology is vocabulary you can see: *struct* in *construct, structure, destruction*. A strategy is a brief, modeled move inside a real page. Question asking, monitoring, summarization, question answering, story mapping, graphic organizers, and working with a partner are the seven the Panel found evidence for.[^82] Reciprocal teaching — Palincsar and Brown, as Shanahan recommends — is I do, we do, you do, with predicting, questioning, clarifying, and summarizing in a shared text.[^83] Teach one, in the chapter you are already reading, for a handful of sittings. Then go back to the chapter. Shanahan’s longer window (daily lessons for four or more weeks in the studies) and Willingham’s brief-dose ceiling are not a contradiction if you refuse to replace knowledge-building with a strategies hour. Florida’s “Knowledge Matters” premise is the same hearing from a state map: comprehension depends more on relevant background knowledge than on mastery of reading strategies.[^84] It is a map rhyme, not a slogan to wear. Paragraphs are taught. An opinion paragraph in this band has a point of view, reasons, linking words, and a closing. An informative paragraph has a topic, facts, and a sense of ending. A narrative paragraph has a sequence and some development. From Grade 4, the child also draws evidence from a text they have read — a sentence from the page, in their own paragraph.[^85] Model one paragraph. Write one together. Then they write one alone. Steve Graham, Debra McKeown, Sharlene Kiuhara, and Karen Harris found strategy instruction a large effect on elementary writing quality (1.02), transcription skill 0.55, extra writing alone a smaller 0.30.[^86] Writing is taught, not assigned. A journal dumped on the table with no model is extra writing. It is not the whole hour. Spelling maps to the same mental representation as reading. Catherine Snow, Peg Griffin, and M. Susan Burns, quoted by Moats: “Spelling and reading build and rely on the same mental representation of a word.”[^87] Moats’s professional-judgment sequence is a useful grain: position patterns and inflections in Grade 2; syllables and common affixes in Grade 3; Latin in Grade 4; more Latin in 5–6.[^88] You do not need her book on the table. You need the hearing: *jumped* is not a new blob; it is *jump + ed*. *Define* and *definition* share a part you can point to. Invented spelling recedes. The IES K–3 guide already said so for the last chapter; this chapter enforces it.[^89] **Wrong answers you should be able to hear** 1. *A one-minute read, many errors ignored, no retell, pride in “how fast.”* Rate without accuracy and without expression. Rasinski’s corruption. Next sitting: a shorter passage the child can already decode, read aloud three times, with you supplying a sound if they stall, then “tell me what it said.” Time, if you time at all, is a check on a passage they can already get right slowly. 2. *Unhappiness read as “unhappy” plus a guess, or skipped, or replaced with “sad.”* Morphology not yet in use. Cover nothing. Point to *un-*, *happy*, *-ness*. “Look through the parts.” The next chapter of the science book will be full of these. Guessing from the first chunk is last chapter’s first-letter habit in longer clothes. 3. *A page of “find the main idea” paragraphs about nothing the family studies, twenty minutes, no history, no science, no story.* A strategies hour. The bag of tricks was emptied and then the hour kept going. Put the strategy inside a page that is about something. Ask one good question. Then learn the something. 4. *becuz, wuz, thay* in a Grade 4 paragraph for words already taught.\* Invented spelling that has not receded. Hold taught patterns to standard spelling in the day’s writing. Untaught Latin may still be approximated. Mehta and colleagues, as Moats reports, saw spelling percentiles drop across grades 1–4 while comprehension held — spelling will not ride along.[^90] 5. *Every hard word on a history page is “I don’t know this,” and the child waits for you to say it, even when the word is* rebuilding\* or *unfairness*.\* The habit of being rescued. Struggle before rescue. Ask, wait, hint (“look at the first part”), then model a *different* word. Leave the pointer in their hand. A sixth, related: the child can retell a read-aloud richly and cannot read the next paragraph of a “grade-level” book. Listening comprehension is ahead of decoding, or fluency has not yet bridged them. That is last chapter’s product, still. Go back toward the code and the oral reread, not toward a harder literature list. **Five-minute parent warm-up** Do this before the lesson, with a short paragraph and a scrap of paper, no child in the room. Minute 1. Read a paragraph you know well — a recipe, a box score, a psalm — and notice how much you bring to it. Then imagine the same paragraph given to someone who lacks that world. That is Recht and Leslie, without a lab. Minute 2. Write *unhappiness*. Draw two lines: *un-* *happy* *-ness*. Say: “Look through the parts.” Write *transport*. Same move if you know *trans* and *port*. Minute 3. Take a four-sentence paragraph. Read it aloud once, carefully. Read it a second time with expression. Notice that the second time is not a race. It is the same words, more available to meaning. Minute 4. Write one sentence you will not accept in this week’s writing: a taught word in invented spelling. Write the standard spelling beside it. That is the receding stage. Minute 5. Write the sentence you will actually say: “Read it with your voice, then tell me what it said.” Put the pencil down. If you can reject a racing minute as fluency, and you can hear “I don’t get it” as maybe knowledge, you are ready to sit down. ## How to teach it this week A good English hour this week has the same shape as last chapter. The English-hour chapter in the front of this book is the full version. Here is the shape scaled to this band. **Warm-up (3–5 minutes, unaided).** A short list of already-mapped words, including one or two with affixes. One sentence from yesterday’s paragraph, read aloud. Paper. No device. **Short model (5 minutes).** One new move. A prefix in a word from the week’s science page. A sentence that will become the topic sentence of a paragraph. A phrase read with expression. You talk for a few minutes. Then you stop. **Student attempt (10–15 minutes).** The longest block. Connected text read aloud, or a paragraph written, or both on alternate days. The child holds the pencil and the pointer. You wait. **One good question.** Not “did you get it?” A question that names the idea: “What do the parts of this word tell you?” “What did that paragraph actually say?” “Where is your reason?” **Mixed practice (5–10 minutes).** Yesterday’s affix in a new word. Last month’s silent-*e* word next to this week’s Latin root. A sentence from the history page next to a sentence from the story. If the book is blocked — most spelling books are — you pull one older pattern onto the page. **Exit ticket (5 minutes, unaided).** Two items isomorphic to what you taught: a new word with the week’s affix, read and spelled; or two sentences that retell the paragraph without looking. Phone in another room. You hold the key. The IES K–3 guide’s Recommendation 4 still applies at the start of this band: connected text every day; no guessing from pictures.[^91] The Panel’s fluency sitting is oral, repeated, with feedback, on text already in the child’s code. **Exact wording you can say** On fluency: “This is not a race. Read it so it sounds like talking.” “Again. Same words. I’ll listen for the phrases.” “Tell me what it said, with the book closed.” On a multisyllabic word: “Look through the parts. Where does the prefix end?” “Cover *ness*. What is left? Now put it back.” “You know *port*. What is *transport* doing?” On knowledge: “What do you already know about this? Tell me before we read.” “The word is easy. The play is not, if we don’t know baseball. Let’s learn the play.” On a paragraph: “What is this paragraph trying to say? Say it in one sentence. That sentence can go first.” “Where is your reason? Point to it.” “Find one sentence in the book that belongs in your paragraph. Copy it with quotation marks, then say it your way.” On spelling: “That pattern we already learned. In our writing we spell it this way.” “Show me the part that *define* and *definition* share.” When you are about to take over: “Your pointer. Your pencil. I’ll ask a question.” Then wait. If you hear yourself finishing their paragraph, or supplying every hard word, you have started doing the work. Stop. **First problem for the student** Write a short paragraph of four or five sentences on a strip, at a level the child can already decode slowly. No helpful drawing of the hard words. Say: “Read this out loud to me three times. Then close it and tell me what it said in two sentences.” Wait through the first reading. Supply a sound if they stall, not the whole word if the parts are known. After the third reading, close the page. Two sentences, unaided. If they race and cannot retell, the minute was speed. If they retell richly after three oral readings, fluency did its job. If they cannot decode the words, this is last chapter’s sitting, not this one. Later the same week, the diagnostic item: a word such as *unhappiness* or *rebuild* in a sentence, pictures absent. Hear a skip. Hear a first-chunk guess. Hear looking through the parts. After the child has attempted, you may show a worked incorrect example: a passage timed at a gallop, errors unchecked, a blank stare when asked what it said. Ask what that person thought fluency was. Generation first, then the named error. **How to fade help** Day of a new affix: you model one word fully, parts then whole then a sentence. The child explains a second. A third is missing only the last part. A fourth is conventional. When the child is accurate, stop showing full examples. Bring the part-lines back when a new Latin root arrives, or when a science chapter dumps ten new words in a page. Fading is for accuracy, not for a calendar date. **When to stop talking** After the question. After the hint. After the model of a *different* paragraph. If you hear yourself summarizing the child’s current page while their book is closed and their mouth is idle, you have started doing the comprehension. Stop. Point to the sentence. Wait. Opinion, information, and narrative all live in this band. You do not need three separate years. One rich task can carry several standards — the Common Core note, and a household mercy.[^92] A history paragraph that states a claim, gives two facts from the page, and closes is opinion and information and W.9 in one sitting. Keyboarding is production, not English-as-typing-class. One page in a sitting by Grade 4 on the map, two by Grade 5, three by Grade 6. Hunt-and-peck that makes a paragraph impossible is a transcription brake, the same family as handwriting in last chapter. Teach enough keyboarding that the sentence can get onto the page. The sentence is still the work. Cursive, if your map or your family wants it, becomes usable here: Texas, complete assignments in cursive by Grade 4; Virginia, all letters in Grade 3 and maintain print and cursive in 4–5; Florida, all letters in Grade 3, legible in 4, fluent and legible in 5.[^93] It is still not a national law. Print that a stranger can read remains acceptable English. ## Practice that actually builds learning Understanding and practice grow together. Wait weeks to put a new affix into a paragraph and the idea stays a card. Drill fluency minutes with no retell and you get a race. **Blocked, for a new move.** The day you introduce *un-*, a short set that is only that: *unhappy, unfair, unkind, unpack*, read and written. The day you introduce a topic sentence, two or three paragraphs that only practice stating the point first. The day you introduce oral rereading, one short passage, three times, with you listening. **Mixed, for when to use it.** Later the same week, the new affix in a science sentence next to an older short-vowel word next to a Latin root you taught last month. A retell of today’s paragraph next to a spelling dictation of yesterday’s pattern. Mixing is choosing a map. A purchased spelling book that is almost entirely one list can still be mixed by you pulling an older pattern onto the same dictation. **Brief retrieval of words and facts already known.** Two or three minutes of already-mapped words, including last month’s affixes. Two oral questions about yesterday’s history page. Cover the answers. You listen. Retrieval is a second meeting, not a quiz show. **One incorrect example to diagnose.** After the racing minute, the gallop. After *unhappiness* guessed as “sad,” the skip. After *becuz* in a taught word, the invented spelling that did not recede. The child names the miss. **Guided oral reading, not a stack of silent pages.** The same short passage, repeated, with feedback. Poetry and a paragraph to a parent are rehearsal for meaning. A stopwatch, if you use one, is a check on a passage already accurate untimed. The Panel did not crown independent silent reading as a causal fluency program. It also did not forbid a child to read a book in a chair. Hear the difference. **Knowledge in the rest of the day.** History, science, literature read aloud above decoding level, nature, scripture, a trade, a sport — these are not “extras” that steal from English. They are the C in (R = D C). A strategies workbook cannot substitute for them. Kitchen talk still helps oral language. Kitchen talk does not, by itself, install *photosynthesis* or the causes of a war. Read the page. Talk about the page. Then write from the page. **Spelling as the code, not as a list.** Dictation of taught patterns. Word sums you can see: *hope + ing = hoping*, with the *e* dropping. Contrast *hoping* and *hopping* so the child hears why the consonant doubled. That is morphophonemic, not a trick. A grocery list is useful practice. It does not replace the dictation. **Paragraphs this week.** One modeled. One shared. One independent. Product goal said in advance: “three sentences, first sentence the point, one reason from the book.” Graham and Perin’s later adolescent meta-analysis found product goals a 0.70 effect on writing quality; the elementary paper found 0.76.[^94] A goal is not a rubric with fourteen boxes. It is a sentence you both can hear. A weekly shape: day of a new affix or a new paragraph move, a short blocked set. Same week: mixed retrieval. Daily: connected text, some of it oral and repeated. Two or three days: writing with a model. You impose that shape on the book you already have. ## For the student You are learning to read so that the page can teach you something, and to write a paragraph that can hold a thought. Fluency means you can read a page you already know how to decode, accurately, at a pace that sounds like talking, with expression. It is not a race. Some pages are easy because you already know the world they are about. A paragraph about a sport you play is easier than the same level of words about a sport you have never seen. That is not a trick. That is knowledge doing part of the reading. You can grow that part by learning things, not only by practicing “find the main idea.” Long words are often parts you already know, stuck together. *Unhappiness* is *un-* + *happy* + *-ness*. Look through the parts. **Tiny worked example** Look at this word: rebuild Cover *re-*. What is left? *Build*. You know that. *Re-* often means “again.” So *rebuild* is “build again.” Now a tiny paragraph: The town had to rebuild the bridge. The old one fell in the storm. Workers used new wood. Read it out loud. Again. One more time, like talking. Close it. What did it say? Two sentences of yours, not the book’s. **Two tries** 1. Look through this word. Show the parts. Say what it means. unfairness 2. Read this paragraph out loud three times to a person. Then close it and write two sentences about what it said. Ants live in colonies. Each colony has workers, and many have a queen. Workers find food and care for the young. For the second one, the hard word *colonies* is *colony* with a plural ending. Look through it. Then read for meaning. **Explain it back** Tell someone at the table what fluency is, and what it is not. Then point to *rebuild* and say how the parts help. Then say why a page about a game you know is easier than a page about a game you do not. **Challenge** Some people read a paragraph as fast as they can and cannot tell you what it said. Some people skip every long word. What would those two moves mean? Read a short paragraph three times, looking through the long words. Then write one paragraph of your own: first sentence your point, next sentence a reason, last sentence a close. Hold the pencil. No one types it for you. You are allowed to struggle. You may mark the parts of a long word. You read aloud. You write. If you get stuck, ask for a hint — not the answer. Then try again. Spelling is the same map as reading, running the other way. *Because* has a pattern you can learn. *Becuz* is a stage you are leaving. ## If it isn’t clicking Three diagnostics. Each one has a next move. None of them is a verdict on talent, and none of them is a reason to wait for a birthday. **1. The child still guesses from pictures or from the first chunk, or cannot decode the week’s “grade-level” paragraph at all.** The code is not done. This is last chapter’s sitting. Next move: go back. Cover pictures. Look through the word. Short decodable text. Phonemic awareness with letters if the sounds were never installed. A “grade 4” comprehension book on top of guessing will not hold. Slow down. Grade 5 is the last Foundational Skills year on the Common Core map; the skill, not the birthday, is the gate. A tutor who will teach the code, not a packet of main-idea circles, is a reasonable next step if you have already rerun last chapter’s daily work and the guess is still the default. **2. Accurate on first reading, but slow and exhausting, or fast and empty.** Fluency is the thin strand. Next move: shorter passages the child can already decode. Guided oral rereading with you, three times, then a retell. Poetry. A paragraph performed to a person. Leave the public race out. If the first reading is full of errors, the passage is too hard — that is accuracy, not fluency. Drop to a page they can get right slowly, then reread. Go ahead to longer chapter books once a short paragraph can be read a third time with expression and a retell. A tutor is useful if daily rereading for a couple of weeks never yields a retell, or if the child refuses to read aloud and the hour has become a fight. **3. Every word comes out, the retell is thin, and the page is about a world the child has never met.** Knowledge and vocabulary are the thin strand, or language comprehension more broadly. Next move: build the world before and during the page. Read aloud above decoding level on the same topic. Teach a handful of words from the page, in the page, not as a disconnected list. Ask what they already know. A strategies worksheet will not install baseball, or the water cycle, or the plot of the book you skipped. Hugh Catts, Suzanne Adlof, and Susan Ellis Weismer later found eighth-grade poor comprehenders with language-comprehension deficits and normal phonological processing — profiles already visible, looking back, in kindergarten.[^95] You do not wait until grade 8 to notice a language problem you can hear in a retell today. Slow down the “grade-level” independent reading until the child has something in the bank. Go ahead on decoding-heavy work if that side is strong; keep feeding knowledge every day. A specialist is reasonable if language comprehension is weak in listening as well as in reading, after you have been reading aloud and talking in full sentences for a stretch of months. **When to slow down.** Code still guessing. Fluency a race or a halt. Paragraphs assigned with no model. Invented spelling of taught words as the house style. Those are brakes. “Not ready” because of age is not a brake this book will use. “Not ready” because last chapter’s code is missing is a real brake. **When to go ahead.** Looking through multisyllabic words by parts. A short paragraph reread with accuracy, expression, and a retell. A paragraph written with a point and a reason, unaided, after a model. Taught spelling patterns held in writing. Knowledge being built on purpose — history, science, stories — not only a strategies hour. Then middle-school literature and argument paragraphs have somewhere to sit. Being “good at reading” because a child races a timed sheet is not a reason to skip knowledge, or to skip the paragraph. **When to get a human tutor.** You have rerun the code work, or the oral rereading, or the modeled paragraph, for a stretch of daily sittings, and the same diagnosis is still the one in the room, and the hour has become a fight. A tutor is a release valve, not a failure of the sitting. Keep yourself as the person who can still hear a gallop, a skip, and a *becuz*. A child who can decode and will not reread is missing one strand. A child who can race and cannot retell is missing another. A child who can retell a read-aloud and cannot decode the page is back in last chapter. Hearing which is thin is enough. ## Tools, including AI Optional helpers for you, the adult. The full rules live in the English-hour box in the front of this book. This chapter is a pointer. The child attempts first. You hold the answer key. A tool may explain an affix or a paragraph move *to you*, make extra passages and word sums with answers on a separate page you keep, write a hint script you read aloud, or help you diagnose a paragraph the child has already written. The child reads the page. The child writes the paragraph. Leave these out of the hour: rewriting the assigned passage so the child does not have to decode it; pasting the prompt and taking the paragraph; “fix this to an A”; an unsupervised chatbot during the attempt; photographing a spelling sheet for a key. Text-to-speech remains compensation for a documented print disability, not a fluency program.[^96] An AI-polished paragraph is performance. The unaided retell, and the unaided sentence the next morning, are the test.[^97] The high-school math study in which an unguarded tutor raised assisted practice and then cut the unaided exam is still mathematics, not an English trial. The crutch is the sitting condition.[^98] A recording of the child reading, which you listen to later, can be a tool if you are teaching two ages at once. You still supply the feedback. A stopwatch can be a tool if the passage is already accurate. It cannot be the identity of being a good reader. ## What “done enough” looks like Placement is by skill, not birthday. A “grade 4” reader is a publisher’s scope, not a legal grade. All About Reading, a literature-rich year, and a textbook anthology can place the same child in three different rooms, because they cut the grain differently.[^99] Place by whether the child looks through long words, can reread a paragraph and retell it, and can write a paragraph with a point. Choose by fit. This book names programs as options. It does not rank them. **Checklist before moving on** - Unfamiliar multisyllabic words are looked through by letters, syllables, and meaningful parts, in and out of context. Guessing from pictures or from the first chunk has receded. - Grade-level prose and poetry the child can decode are read aloud with accuracy, reasonable rate, and expression, on successive readings, and then retold. Fluency is not a race. - A short opinion or informative paragraph can be written unaided after a model and a shared write: a point, a reason, a close. From the Grade 4 map onward, one bit of evidence comes from a text. - Taught spelling patterns, inflections, and common affixes are held in writing. Invented spelling has receded for those patterns. Spelling is treated as the same map as reading. - Vocabulary is growing from morphology and from knowledge-building text, not only from a quiz of rare words. - Knowledge is being built on purpose in history, science, and literature, including read-alouds above decoding level. A content-free strategies hour is not the English program. - Keyboarding is far enough along that one page (Grade 4 map) or two (Grade 5 map) can be produced in a sitting when you need typing. Handwriting — print, and cursive if you have chosen it — is still legible. - You can hear a gallop, a skip, a *becuz*, and a thin retell, and you can ask a good question without finishing the paragraph. If most of that list is true, go on to longer literature and argument paragraphs, even if the birthday says otherwise. If the birthday says “sixth grade” and the pictures are still doing the reading, or the minute is still a race with no retell, stay. The next chapter assumes a reader. This chapter’s job was to make fluency, knowledge, the paragraph, and the spelling map hearable. A path through English is the promise. A Lexile is not. # Chapter 3 ## Longer books, argument paragraphs, grammar in the sentence A closed hardcover, handwritten index cards, and a black pen on a wooden desk. ## Why this matters Last chapter built the bridge: fluency, knowledge, a paragraph, spelling as a map. Keep all four. This chapter is the first long stretch of English that looks like what a stranger means by the word. Novels and substantial informational text. An argument paragraph that can grow into an essay. Grammar as craft in the sentence the student is already writing. The sentence that surprises people is this one: a sixth-grader who still guesses from pictures is not “behind in literature.” They are still in a code program. The Common Core English map has no Foundational Skills strand in grades 6–12.[^100] Florida keeps 6–12 foundation standards for intensive intervention electives, not for core ELA.[^101] If last chapter’s checklist is not true, stay there. Birthday is not placement. Compacted acceleration into high-school literature on a fragile code is the analogue of starting Algebra I on a broken fraction foundation. The book in front of you can wait. The code cannot. What you are teaching, as English, is longer literature and informational text that the student can actually read; citing a bit of the page, then several bits, then using them; theme and central idea; evaluating an argument on the page; writing claims with reasons and evidence, and by grades 7–8 naming another view; grammar and sentence combining in the draft; research that continues a habit begun in the early grades, not a paper that is born in eighth grade; and enough keyboarding to produce three pages in a sitting on the Common Core map.[^102] Florida names rhetorical appeals in Grade 6, so students will have a handle on logos, pathos, and ethos when they start argumentative writing.[^103] One sentence of that naming is enough for this chapter. Then teach claims in ordinary language. A claim is a sentence that needs a reason and a bit of the text. A counterclaim is not a quarrel. This is not a Critical Thinking 101 course, and it is not a transcript title. The Common Core map still wants literature. The famous 70 percent informational figure is a share of student reading *across the grade*, not a command to empty the English-class novel list. The document says so in a footnote: senior English teachers are not required to devote 70 percent of reading to informational texts.[^104] Texas will later put American, British, and world literature *inside* English I, not as three middle-school years.[^105] Virginia wants graduates exposed to Homer and Shakespeare among others, and still prints no prescribed list.[^106] Florida prints sample texts by benchmark in its own standards — state exemplars, not a federal canon.[^107] Teach the book on the table. Teach a substantial article on the table. Both. Writing in this band is taught, not assigned. Steve Graham and Dolores Perin’s meta-analysis of grades 4–12 found strategy instruction a 0.82 effect on writing quality, summarization 0.82, sentence combining 0.50, and grammar instruction as tested a −0.32.[^108] Grammar in that analysis is explicit teaching of parts of speech and sentence types, often versus a treatment that itself taught writing. The negative sign means those grammar conditions lost to the comparison, not that no one should ever learn a participle. Sentence combining is the nearby practice that moved quality. Self-regulated strategy development, as a subset of strategy instruction in that paper, was 1.14 versus 0.62 for other strategy instruction. This is a useful study, not a promise that every home will see the same result. Then teach. The middle-school lever is a planning strategy for this genre, plus sentence work in the draft. A worksheet island of circling nouns is the sitting the meta-analysis already tested and did not reward. Hugh Catts, Suzanne Adlof, and Susan Ellis Weismer found eighth-grade poor comprehenders with language-comprehension deficits and normal phonological processing, and poor decoders with the opposite pattern. Those profiles were already visible, looking back, in kindergarten.[^109] You do not wait until grade 8 to notice a language problem you can hear in a retell, and you do not wait until grade 8 to notice a code problem you can hear in a skipped word. Last chapter said this. It still holds. Grade labels remain sequence anchors. A child may sit this band at calendar grade 5 or calendar grade 9. Move a student who can reread a paragraph, retell it, look through long words, and write a paragraph with a point. Hold the long book for a student who cannot. Skipping argument because “we do novels” is a miss. Skipping novels because “we do articles” is the other miss. You do not need to be an English specialist. You do need to hear a claim with no reason as unfinished, a counterclaim as a view rather than a fight, and a grammar error as a craft problem in this sentence, not as a reason to buy a new workbook. ## For the parent: understand it yourself Many adults feel rusty on argument, on why a long book is worth the weeks, and on what to say about a run-on besides “fix it.” Five minutes at the end of this section is enough for tomorrow. Last chapter installed fluency as accuracy plus expression plus a retell, and a paragraph with a point. Keep both. **Everyday picture.** Two people at the table disagree about whether a character should have told the truth. One says “because I would have.” The other says “because in chapter four she promised, and here is the sentence.” The second person has made a claim and attached a bit of the text. That is the work. The first person has a feeling. Feelings can start an argument. They cannot finish one on a page a stranger will read. **Precise picture.** A claim is a debatable sentence. “The bridge is made of wood” is a fact, or a detail. “The town was right to rebuild the bridge first” is a claim. It needs a reason (the because-link) and evidence (a bit of the page, a fact already established, a result you can point to). By Grade 7 and 8 on the Common Core map, the writer also names an alternate or opposing claim.[^110] Florida’s ladder is the same shape in state language: Grade 6, claim and evidence from multiple sources; Grade 7, acknowledge a counterclaim; Grade 8, rebut.[^111] A counterclaim is the other reasonable view, stated fairly, not a cartoon of a fool. Rebutting it is answering it, not mocking it. Rhetoric, when Florida names appeals in Grade 6, is this work with three ordinary nicknames: a reason that follows (logos), a feeling the audience already has (pathos), the writer’s worthiness to be believed (ethos). You may use the nicknames once. Then go back to claim, reason, evidence, other view. Longer literature is a book you teach, not a book you assign and quiz. What happens. How it is made. What it is saying. Those three questions, which a later chapter will own as a method, already work on a middle-school novel, a short story, a poem, and a play. Informational text in this band is not a war on novels. It is a chapter, an article, a report, a historical document, read with the same three questions plus one: what is the author’s claim, and does the evidence hold. Evaluating arguments is a Reading Informational standard on the map.[^112] The student cites the text in Grade 6, cites several pieces in Grade 7, and uses evidence more firmly in Grade 8.[^113] Finger still under the sentence when they quote. The quote is on the physical page, not in a chatbot’s memory. Grammar earns its keep in the sentence on the table. The maps still test conventions. SAT Reading and Writing and ACT English will later test them too; those are signals, not this year’s course.[^114] The teaching move is not a year of parts-of-speech worksheets. Graham and Perin already measured that sitting against writing quality and saw it lose. The teaching move is: find the run-on in this sentence; combine these two short sentences three ways and pick the best; keep subject and verb together in the sentence you just wrote. Sentence combining, 0.50 in that analysis, is grammar as craft.[^115] Name a participle when one is misbehaving in the draft. Leave the island of circling them. Research is not born this year. Grade 1 already had shared research on every map this book uses. What changes is independence, a standard citation format by grades 7–8, and the habit of a source card that is not a bibliography copied from a screen.[^116] A short inquiry that generates a further question is enough. The documented high-school paper is a later chapter. Keyboarding: three pages in a sitting by Grade 6 on the map. That is production. The argument paragraph is still the work. A student who cannot get a page out by hand or by keyboard has a transcription brake, last chapter’s sitting in older clothes. **Wrong answers you should be able to hear** 1. *A paragraph that restates the topic and never makes a claim: “This essay is about whether the town should rebuild.”* A topic is not a claim. Ask: “What do you actually think, in one sentence that someone could disagree with?” The first sentence of the paragraph should be that sentence, or should lead to it by the second line. 2. *A claim with a feeling and no bit of the text: “She is brave because I like her.”* Missing evidence. “Point to a sentence in the book that made you think that. Put it in the paragraph, then say why it counts.” The because-link is the warrant, in ordinary language. You do not need the word *warrant* until a later chapter, and you may not need it then. 3. *A “counterclaim” that is a sneer: “Some people are just dumb and think the opposite.”* Not a view. A quarrel. Ask: “What would a reasonable person on the other side say, in a clause we could put after *although*?” Grade 7 is acknowledge. Grade 8 is rebut. Fairness first. 4. *A grammar worksheet finished in ink, then a draft full of the same run-ons, untouched.* The island. Circle the run-on in *this* draft. Combine the two clauses. The worksheet was performance. The next morning’s unaided sentence is the test. 5. *A six-hundred-page book assigned, never discussed, a packet of chapter questions, the student lost at chapter two and silent.* Assignment without teaching. Slow down. Read some of it aloud. Talk after a sitting. One scene, one sentence that does work, one claim about the character. A later chapter will teach a book in full. This band already needs the habit. A sixth: every quotation in the paragraph was supplied by a chatbot, including a page number that is not in the book. Language models invent titles, pages, and critics. Treat model-supplied quotations as untrusted until they are checked against the physical book. The student finds the sentence. The student copies it. Then they interpret it. **Five-minute parent warm-up** Do this before the lesson, with a short article or a page of the novel, no student in the room. Minute 1. Write one claim about the page that a reasonable person could disagree with. Not a topic. A claim. Minute 2. Underline one sentence on the page that could serve as evidence. Write a because-sentence that is not a restatement of that evidence. Minute 3. Write one other view in a single clause beginning with *although*. Minute 4. Find one sentence in last week’s student draft that is two sentences glued with a comma, or a modifier hanging off the wrong noun. Rewrite it. That rewrite is tomorrow’s grammar lesson. Minute 5. Write the sentence you will actually say: “What is your claim, and where is the bit of the text?” Put the pencil down. If you can tell a topic from a claim, and a sneer from a counterclaim, you are ready to sit down. ## How to teach it this week A good English hour this week has the same shape. The English-hour chapter in the front of this book is the full version. Here is the shape scaled to this band. **Warm-up (3–5 minutes, unaided).** Two sentences from yesterday’s book, read aloud. One already-taught combining move: two short kernels into one sentence. Paper. No device. **Short model (5–8 minutes).** One new move. A claim written in front of them, then a because, then a quoted bit with a page. Or two kernel sentences combined three ways, the best one chosen out loud. You talk for a few minutes. Then you stop. **Student attempt (15–20 minutes).** The longest block. Reading a scene or a section of an article, marking one sentence, or writing the paragraph. They hold the pencil. You wait. **One good question.** Not “did you like it?” A question that names the idea: “What is the claim?” “Where is that on the page?” “What would the other view say?” “Which combined sentence is best, and why?” **Mixed practice (5–10 minutes).** Yesterday’s combining move in today’s paragraph. Last month’s character next to this week’s article. A taught Latin root in a word from the science chapter. If the grammar book is blocked — most are — you pull one older sentence problem onto the draft. **Exit ticket (5 minutes, unaided).** One claim sentence and one evidence sentence, book open, tool closed. Or two kernels combined two ways, the better one starred. Phone in another room. You hold the key. **Exact wording you can say** On a claim: “Say it so that a reasonable person could disagree.” “That is a topic. Shrink it to a claim. One sentence.” “What is the because?” On evidence: “Finger on the sentence. Now write it with quotation marks and a page.” “That quote is a fact. Your next sentence has to say why it counts.” On the other view: “Although who, and although what?” “State it fairly. Then we can answer it.” On grammar in the draft: “Read this sentence out loud. Where did you need to breathe? That might be two sentences.” “Here are two short sentences. Combine them three ways. Pick the best.” “The -ing phrase is sitting on the wrong noun. Who is actually doing that?” On a long book: “What happened in this sitting, in three sentences.” “Point to one sentence the author made work. What is it doing?” “What is this chapter saying, not just what happened?” When you are about to take over: “Your pencil. I’ll ask a question.” Then wait. If you hear yourself finishing their argument, or rewriting their sentence while they watch, you have started doing the work. Stop. **First problem for the student** After a short shared reading — a scene, or a one-page article — write this prompt at the top of a blank page: Write one paragraph that states a claim, gives one piece of text evidence, and names one other view in a clause. Book open. No chatbot in the room. Say: “A claim is a sentence someone could disagree with. Evidence is a bit of the page. The other view is a fair although.” Wait. If they write a topic, they have not yet claimed. If they write a feeling with no quote, they have not yet used the page. If they sneer at the other view, they have not yet named it. If they produce a fluent paragraph that appeared faster than their handwriting usually allows and they cannot point to their quote in the book, the tool may have done the sitting. Ask them to point. Ask them to say the because in their own mouth. Later the same week, the diagnostic item: two kernel sentences on the table — *The town rebuilt the bridge. The old one fell in the storm.* — combine three ways, pick the best, say why. Hear a comma splice. Hear a long ramble. Hear a clean subordinate clause. After the student has attempted, you may show a worked incorrect example: a paragraph that begins “This paragraph is about bravery” and never claims, or a “counterclaim” that calls the other side stupid. Ask what is missing. Generation first, then the named error. **How to fade help** Day of a new genre move: you model one paragraph fully, claim then because then quote then although. The student explains a second, already written. A third is missing only the although-clause. A fourth is conventional. When they are accurate, stop showing full examples. Bring the skeleton back when a new kind of text arrives — a poem, a report, a speech — or when the first multi-paragraph essay is asked. Fading is for accuracy, not for a calendar date. **When to stop talking** After the question. After the hint. After the model of a *different* paragraph. If you hear yourself interpreting the novel while their copy is closed and their pencil is idle, you have started doing the reading. Stop. Point to the line. Wait. Several standards can live in one rich task. A week on a short novel can carry citing the text, a claim paragraph, a combining lesson in the draft, and a short discussion. That is the map’s own mercy, and Graham and Perin’s lever: a strategy for this genre, in this text, plus sentence work where the sentence lives. Literature and informational text both belong. A year of only novels with no argument is a thin C in Gough and Tunmer’s product for the kind of page that carries a claim. A year of only articles with no story is a thin literacy-knowledge strand and a thinner habit of attending to how a thing is made. Alternate. A novel alongside a related article is one good month. If the student is still guessing at words, the long book is the wrong object. Return to last chapter, or to the first. Florida’s intervention foundations exist because some twelve-year-olds still need them. That is not an insult. That is placement. ## Practice that actually builds learning Understanding and practice grow together. Wait weeks to write the argument paragraph after “learning about claims” and the idea stays a speech. Drill grammar items with no draft and you get a worksheet score. **Blocked, for a new move.** The day you introduce a claim, three claim-sentences about the same short text, only that. The day you introduce although-clauses, a handful of fair other-views, only that. The day you introduce combining with *because* or *although*, a short set of kernel pairs. **Mixed, for when to use it.** Later the same week, a claim paragraph that also needs a combined sentence and an although-clause, next to a handful of older kernel pairs, next to a quote from last month’s story. Mixing is choosing a move. This chapter does not import a mathematics interleaving coefficient. It borrows the practice of putting unlike tasks on the same page. **Brief retrieval.** Two already-made claims from earlier in the month, said again without notes. Two combining moves already taught. A Latin root from last chapter in a new word. Cover the page. You listen. **One incorrect example to diagnose.** After a topic-masquerading-as-claim, the “this essay is about” opener. After a sneer, the cartoon other-side. After a comma splice in the draft, the two clauses with a comma and no *and*. The student names the miss. **Reading that is actually reading.** A sitting with the book open, a marked sentence, a spoken retell, then a written claim. Read-alouds still belong, especially for books whose language is ahead of fluency. Independent silent reading of a long book with no check is last chapter’s Panel caution in longer clothes. You do not have to quiz every chapter. You do have to talk, and you do have to see a marked page. **Sentence combining in the draft.** Two kernels from the student’s own paragraph, combined three ways, the best put back. That is Graham and Perin’s 0.50, scaled to Tuesday. Parts of speech named as needed, in that sentence. The Language strand on the maps is real. The island is the miss. **Knowledge still feeds comprehension.** Middle-school pages assume worlds: myth, scripture, civics, science, a period of history. Willingham’s baseball still applies, only the baseball is now the French Revolution or cell division or the Odyssey.[^117] A strategies hour still has a ceiling. Build the world. Then read the page. Then write from the page. **Inquiry that continues.** A question, two sources, notes in the student’s hand, credit given, a further question. Standard citation format by 7–8: pick one style and keep it. The documented paper is later. Starting it as if elementary inquiry never happened is the miss the maps already refuse. A weekly shape: two or three days in the long book or the article, with talk and a marked sentence. One day a modeled then independent argument paragraph. One day sentence work in that draft. Mixed retrieval of older moves. You impose that shape on the books you already have. ## For the student You are learning to read longer work, and to write a paragraph that argues something a reasonable person could disagree with. A claim is not a topic. “Bravery” is a topic. “The character was right to tell the truth in chapter four” is a claim. Someone could disagree. That is how you know. Evidence is a bit of the page — a sentence you can point to — plus a because that is not just the same sentence in new clothes. Another view is not an insult. It is the although. *Although she had promised to be silent, …* Then you answer it. Grammar is how your sentence does what you meant. Combining two short sentences can make the because show up in the grammar, not only in your head. **Tiny worked example** Two kernels: The town rebuilt the bridge first. People needed a way to cross for food. Combined one way: *The town rebuilt the bridge first because people needed a way to cross for food.* A claim paragraph, miniature: The town was right to rebuild the bridge first. The article says families could not reach the market until the crossing was back (page 2). Although some people wanted the school repaired sooner, a town that cannot move food cannot run a school. Claim. Evidence with a page. Although. That is the shape. **Two tries** 1. Turn this topic into a claim. Then add one because. Topic: the character’s promise 2. Combine these two sentences three ways. Star the one you would actually put in a paragraph. The old bridge fell in the storm. The town had no crossing for a month. For the second one, you might use *because*, *after*, *which*, or a semicolon if you know it. Say why the starred one is best. **Explain it back** Tell someone at the table the difference between a topic and a claim. Then say what a fair although-clause does. Then point to a sentence in your book and say what you would use it as evidence *for*. **Challenge** Write one paragraph that states a claim, gives one piece of text evidence, and names one other view in a clause. Book open. No chatbot in the room. Then take two short sentences from your own paragraph and combine them. Put the better sentence back in. Some people write “This paragraph is about…” and never claim. Some people skip the page and write only what they feel. What would those two moves mean? How does your paragraph avoid them? You are allowed to struggle. You may mark the book. You write. If you get stuck, ask for a hint — not the paragraph. Then try again. Rhetoric, if someone uses the word, is just this work with a fancier label: a claim, a reason, a feeling the reader already has, a writer worth believing. You do not need a separate course called Critical Thinking 101. You need a paragraph you wrote. ## If it isn’t clicking Three diagnostics. Each one has a next move. None of them is a verdict on talent, and none of them is a reason to wait for a birthday. **1. The long book is a wall: guessing, stalling, no retell, or a retell that matches a summary on a screen rather than the page.** The code, or fluency, or honesty of the sitting, is the thin strand. Next move: check last chapter’s checklist. If pictures or first-chunks are still doing the reading, go back. If the first reading is accurate but exhausting, shorter daily sittings of oral rereading, then the novel in smaller chunks, some of it read aloud by you. If the retell matches an online summary and not the chapter, close the search and open the book. A still-guessing sixth-grader is in a code program. Florida’s intervention foundations are for this student, as an intensive path, not as a shame label.[^118] A tutor who will teach the code or listen to oral reading is a reasonable next step if you have already rerun the bridge and the wall is still the wall. **2. The paragraph will not claim, or the claim will not attach to the page, or the other view is a sneer.** Argument is the thin strand. Next move: smaller. One claim sentence a day for a week, about the text already read. Then one quote with a page. Then one because-sentence that is not a restatement. Then an although-clause, fair. Model, shared write, independent. Product goal said in advance: “four sentences: claim, quote, because, although.” Slow down the “essay” until the paragraph holds. Go ahead to two paragraphs, then to a counterclaim answered, once the four-sentence shape is unaided. A tutor is useful if daily modeled paragraphs for a couple of weeks still produce only topics, and the hour has become a fight. **3. Conventions worksheets are fine, and the draft is still full of run-ons, fragments, or modifiers on the wrong noun.** Grammar is still an island. Next move: stop using the draft as a place that never gets edited, and stop using the worksheet as the English hour. Take two kernels from *this* paragraph. Combine them. Fix the one run-on in front of you, not the twenty in a packet. Graham and Perin’s grammar-as-tested sitting lost on writing quality; sentence combining did not.[^119] Name the rule in the sentence. Retrieval of a named rule can be brief, later, mixed. A specialist is reasonable if language in listening and in writing is disordered after months of this work — Catts and colleagues’ poor-comprehender profile is a language profile, not a laziness profile — and you want a second pair of ears.[^120] **When to slow down.** Code still guessing. Fluency still a race with no retell. Claims that are topics. Books assigned and never taught. Grammar only on islands. Those are brakes. “Not ready” because of age is not a brake this book will use. “Not ready” because the paragraph and the page are not yet connected is a real brake. **When to go ahead.** A long work can be read, talked about, and written about with a finger on the sentence. A paragraph claims, quotes, and names another view, unaided after a model. Sentence combining shows up in the draft. Inquiry has notes and credit. Then high-school composition and literature courses have somewhere to sit. Being “good at English” because a student talks well at dinner is not a reason to skip the unaided paragraph. **When to get a human tutor.** You have run the smaller claim work, or the oral rereading of the long book, or the combining in the draft, for a stretch of daily sittings, and the same diagnosis is still the one in the room, and the hour has become a fight. A tutor is a release valve, not a failure of the sitting. Keep yourself as the person who can still hear a topic, a sneer, and a run-on. A student who can argue at dinner and will not look at the page is missing one strand. A student who can decode and will not claim is missing another. A student who can claim and cannot hold a sentence is missing craft. Hearing which is thin is enough. ## Tools, including AI Optional helpers for you, the adult. The full rules live in the English-hour box in the front of this book. This chapter is a pointer. The student attempts first. You hold the answer key. A tool may explain a claim or a combining move *to you*, make extra kernel pairs and short articles with answers on a separate page you keep, write a hint script you read aloud, or help you diagnose a paragraph the student has already written. The student reads the book. The student writes the paragraph. Leave these out of the hour: pasting the prompt and taking the essay; “fix this to an A”; an unsupervised chatbot during the attempt; rewriting the chapter so the student skips the long book; photographing an editing sheet for a key; a companion persona. Quotations the model supplies are untrusted until they are found in the physical book. Detectors fail often enough that they are not a verdict on a student’s honesty. A detector is not part of the English hour.[^121] An AI-polished paragraph is performance. The unaided paragraph the next morning, book open, tool closed, is the test.[^122] The high-school math study in which an unguarded tutor raised assisted practice and then cut the unaided exam is still mathematics. Writing is even easier to outsource, because the model’s native output is a fluent paragraph. The paper itself notes that math has objective scoring which is unavailable in creative subjects such as writing. That is a caution, not a writing trial.[^123] If the tool produces the paragraph, the student is a spectator. ## What “done enough” looks like Placement is by skill, not birthday. A “grade 8” literature book is a publisher’s scope, not a legal grade. A literature-rich year, a textbook anthology, and a writing-first program can place the same student in three different rooms.[^124] Place by whether they can read a long work with their finger on a sentence, write a claim paragraph unaided, and combine sentences in the draft. Choose by fit. This book names programs as options. It does not rank them. The transcript line a stranger can read is still Language Arts, or English 8, not a brand, and not Critical Thinking 101. **Checklist before moving on** - A novel, a substantial short-story stretch, a poem or play, and informational text have all been on the table this year. The English hour was not emptied of literature, and it was not only literature. - The student can cite a bit of the text in Grade 6 shape, several bits in Grade 7 shape, and use evidence more firmly toward Grade 8, with the sentence findable on the physical page. - An unaided paragraph states a claim, gives evidence, and names another view in a clause. By Grade 8 shape, the other view can be answered, not only acknowledged. - Grammar work happens in the sentence being written. Sentence combining is in use. A worksheet island is not the program. - Keyboarding or handwriting can produce the equivalent of three pages in a sitting when you need a longer piece. Production is not the same as argument, but argument has to get onto the page. - Inquiry has a question, notes, credit, and a further question. Standard citation format is begun. The high-school documented paper is not being asked to be born here from nothing. - A still-fragile decoder is in a code or fluency program, not only in a literature packet. - You can hear a topic, a sneer, a run-on, and a quotation that is not in the book, and you can ask a good question without finishing the paragraph. If most of that list is true, go on to high-school composition and to teaching a book as a book, even if the birthday says otherwise. If the birthday says “ninth grade” and the student still cannot hold a claim to a page, stay. The next chapters assume a writer who can claim and a reader who can point. This chapter’s job was to make the long text, the argument paragraph, and grammar as craft hearable. A path through English is the promise. A Lexile is not. # Chapter 4 ## Composition A marked manuscript stack, a fountain pen, a crumpled draft, and a coffee mug. ## Why this matters High-school writing is the year a sentence has to carry a claim, a paragraph has to hold a reason, and an essay has to keep a line of thinking without you sitting in the chair. It is not a personality trait and it is not a brand. It is work that is taught: a model you can see, a page you write together, then a page the student writes alone. That is worth the struggle because almost everything later hangs on it. The documented paper in the next stretch of this book is an essay with sources, not a new species. A literature course that asks for an analytical paragraph is asking for the same control, now about a book. College composition, if it comes, will assume the student can already hold a thesis, use evidence, and revise. A year that assigns five “essays” and never teaches a sentence is writing in name. Name the year by what was taught. Typical U.S. high school is four English credits, not a federally named American–British–World sequence. Nearly every state in a 2018 federal table asked for four English credits; California’s state minimum in that table was three, a few states left the count to local boards, and Texas, Florida, and Virginia each asked for four.[^125] Course titles a registrar actually sees vary. Texas names the legal courses English I, II, III, and IV; Virginia’s Board names English 9 through 12; local catalogs often subtitle a junior year “American Literature.” Those are packaging facts. The writing work inside them is the same job: sentence, paragraph, essay, with argument and informative writing carrying more of the year as the student gets older. The Common Core writing map, which many families never use and which this book does not treat as a homeschool law, still names the emphasis clearly: “the overwhelming focus of writing throughout high school should be on arguments and informative/explanatory texts.”[^126] Narratives do not vanish. They stop being the whole of the writing hour. Texas English I already asks for literary, informational, argumentative, and correspondence writing in one course; English III adds literary analysis and rhetorical analysis as composition genres.[^127] Florida’s writing ladder keeps narrative, argumentative, and expository every year, with more control of counterclaim, sources, and tone as the grades rise.[^128] Steal the cluster. You do not have to steal the logo. Writing is taught, not assigned. Steve Graham and Dolores Perin, looking at experimental and quasi-experimental writing instruction in grades 4–12, found that teaching a strategy for the kind of writing on the table moved quality (weighted average 0.82); summarization did the same (0.82); a peer helping as the student wrote (0.75); a clear product goal (0.70); word processing (0.55); sentence combining (0.50). A process-writing package averaged 0.32, and nearly zero without professional development. Traditional grammar instruction, as tested in that literature — parts of speech and sentence types, often against a treatment that itself taught writing — averaged −0.32.[^129] Self-regulated strategy development, Harris and Graham’s more fully specified version of strategy instruction, averaged 1.14 in that subset, against 0.62 for strategy instruction without those extra pieces.[^130] Those are averages across heterogeneous studies, often on researcher-made quality measures close to what was taught, often with students who struggle. This is a useful pattern, not a promise that every kitchen will see the same number. Then teach. What changes in this band is control, not the invention of writing. The student who cannot yet hold a sentence is not “in high-school composition.” That student is still in taught writing, whatever the birthday and whatever the transcript year. The student who can already plan a claim, write a paragraph from an outline, and combine two short sentences into one that actually says more, is ready for the essay as a week’s work, not as a mystery. Transcript the year as **English 9**, **English 10**, or **Composition** — titles a stranger can read. The NCAA sample transcript uses English 9–12 and Composition I and II; the worksheet wants the course name to match the title on the page.[^131] A year of a named writing method is work done *inside* a course. It is not the course name. Never write a brand in the title column. Never write “AI English.” College-ready overlays sit on top of this band. They are not the kitchen method. AP English Language is an introductory college composition course: rhetorical analysis of nonfiction, evidence-based analytic and argumentative writing, no prerequisite course.[^132] CLEP College Composition is a first-year composition exam: fifty multiple-choice items and two timed essays, with ACE recommending six semester hours at a score of 50, and each college setting its own policy.[^133] SAT Reading and Writing (the current section name) and ACT English are signals, not instruction: College Board’s readiness benchmark for Reading and Writing is 480; ACT’s English benchmark is 18.[^134] Those two numbers are not equated to each other. They are publisher-defined probabilities of a grade in a first-year course, not the meaning of English. The door is skill. Can the student write a sentence that is actually a sentence? Can they write a paragraph with a claim and one piece of support? Can they revise one sentence by combining or subordinating, not by hoping a tool will make it sound older? If those three are fragile, stay there. The rest of this chapter teaches the course. This page names why the course is worth a hard year, and why the unaided draft — not the polished file — is the work. ## For the parent: understand it yourself You do not need to be an English teacher. You do need to understand today’s idea well enough to hear a fact offered as a thesis, and a chatbot paragraph offered as the student’s voice, not as cute slips. **Everyday picture.** Think of giving directions to a house the other person has never seen. “There is a house” is a fact. It does not get anyone there. “Turn left at the oak, then the blue door is the third on the right” is a claim about a route, with reasons you can check. High-school writing is directions for a thought. The reader was not in the room when the thought formed. The page has to carry them. **Precise picture.** A *topic* is what the piece is about. A *thesis* is an arguable claim about that topic, the kind of sentence a reasonable person could disagree with. “The school day starts at eight” is a topic sentence that is a fact. “Starting the school day at eight costs students attention they would have at nine” is a thesis. It needs reasons. Each reason needs evidence — a bit of a text, a number from a source you actually opened, a scene from a book that is on the table. A *warrant*, said in ordinary language, is the because-link: why that evidence is allowed to support that claim. “The passage shows the narrator lying twice in one page; a narrator who lies that often is not a guide we can take at his word” is a warrant. “Because I said so” is not. The sentence is the unit. A paragraph is a small essay: one claim, support, a turn if the genre wants one, a finish that is not a shrug. An essay is several of those, in an order the reader can follow, with a beginning that names the job and an ending that does not merely repeat the first paragraph in new clothes. High-school maps keep three types in view: arguments, informative/explanatory texts, and narratives. Argument and informative writing carry more of the year. Correspondence, literary writing, and (on some maps) rhetorical analysis sit in the same family. They are not three franchises. They are three jobs a sentence can do. The teaching sequence is older than any program on your shelf. Model a sentence so the student can see the move. Write one together, pencil passing or two hands on one page. Then they write alone. Fade: a complete model, a shared write with you doing less, a completion (topic sentence given, they write the support), then a conventional paragraph from an outline they made. When they can execute, stop showing full models of the same move. The model that helps a beginner can become a crutch later. A product goal is a sentence you both can check. “Write a paragraph that states a claim, gives one piece of text evidence in the student’s own words plus a short quotation, and names one other view in a clause” is a product goal. “Write a good essay” is a mood. Graham and Perin’s 0.70 for product goals is the reason the first kind of sentence is worth writing on a sticky note before you sit down.[^135] Sentence combining is grammar that earns its keep. Two kernels: *The river froze. The town lost its ferry.* Combined: *When the river froze, the town lost its ferry.* Or: *The river froze, so the town lost its ferry.* Or: *The town that lost its ferry had already watched the river freeze.* The student chooses. The parent does not pick a winner as “the sophisticated one.” Combining moved writing quality in that same literature (0.50).[^136] Isolated parts-of-speech worksheets, as tested, did not. Naming a participle in the sentence on the table is a different act from a year of circling them on a worksheet island. Chapter 6 takes the Language strand in full. This chapter only needs the nearby move: build the sentence they are actually writing. Strategy instruction, including the Harris-and-Graham version, is a planning and revising routine for *this* genre, practiced until the student can use it without you. The six stages, in kitchen language: know enough about the genre to see what “done” looks like; hear the steps (often a short mnemonic); watch you use the steps out loud; remember the steps; use them with you fading; use them alone, with a goal and a self-check.[^137] You do not need a training institute to steal that shape. You do need to stay with one genre long enough that the steps are not a chant. The Writing Revolution, the Hochman Method, is a named sequence that starts at the sentence and embeds writing in the content being studied: expansion with *because*, *but*, *so*; combining; topic sentences; outlines for a paragraph and then for several.[^138] It matches, at the level of ideas, sentence combining and product goals. It is a 2017 book, not a randomized trial. This book will not invent a number for it, and will not rank it against other writing methods families actually use. Those other methods are tools, not the course. Some families imitate from a source text and dress sentences with specified style moves. Some write in partnership while spoken fluency still outruns the pencil. Some narrate, then copy, then move into exposition from outlines. Some embed sentence work in history and science so English is not a separate island. The studies do not name a franchise as the winner. They support a simpler list: teach a strategy for the genre on the table; build sentences; set a product goal; the child writes. AP Language is not AP Literature. One sentence, because the two get mashed: Language is college composition and the rhetoric of nonfiction; Literature is literary analysis of fiction, poetry, and drama.[^139] Neither is required for a homeschool transcript. If the student sits an exam, the score is a separate line. The kitchen still teaches the sentence. ### Wrong answers you should be able to hear **1. A fact offered as a thesis.** The student writes “Shakespeare wrote *Hamlet*” as the claim of an essay, or “The Civil War happened in the 1860s,” and then summarizes. *What it usually means.* Topic and thesis have collapsed. The student knows what the piece is about and has not yet said anything that could be disagreed with. *What to say.* “That sentence is true. A thesis is a sentence a reasonable person could argue with. What do you want us to believe *about* Hamlet, or about that war? Write that sentence first. Then we find the support.” **2. The five-paragraph chant as the whole of writing.** The student can fill three “reasons” that are the same reason in three hats, or pads a conclusion that only restates the introduction, and freezes when the assignment is two pages, or one tight paragraph, or a letter. *What it usually means.* A school shape was learned as a ritual, not as one possible order for a line of thinking. The shape is not the enemy. Treating the shape as the thinking is. *What to say.* “Three body paragraphs are allowed when you have three different reasons. If you have one reason with two pieces of evidence, write one strong paragraph. The job is the line of thinking, not the count.” **3. “I wrote a lot, so it’s done.”** Length without a claim, or a claim that appears in the last three lines after a page of summary. *What it usually means.* Extra writing alone is a small lever. In the elementary meta-analysis, extra writing averaged 0.30; in the adolescent one, process-without-training was essentially zero.[^140] Time on the page is not the same as instruction. The student has practiced producing words, not producing a controlled piece. *What to say.* “Show me the sentence that is the claim. Circle it. Now show me the sentence that supports it. If the claim is hiding at the end, move it up. Then cut what does not serve it.” **4. Grammar as the writing program.** A stack of parts-of-speech worksheets, a clean quiz on predicate adjectives, and a paragraph that still runs on, or still cannot hold a because-clause. *What it usually means.* The grammar hour never touched the sentence the student was writing. In Graham and Perin, that kind of instruction lost to comparison treatments on writing quality (−0.32). Sentence combining, which is grammar in use, did not. *What to say.* “We will name the problem in *this* sentence. Is it two sentences smashed with a comma? Is the modifier attached to the wrong noun? Fix this one. Then write the next one on purpose.” **5. A fluent paragraph the student cannot read aloud as theirs.** The diction jumped a grade in twenty minutes. The student cannot say what the third sentence is doing. The outline, if it exists, does not match the page. *What it usually means.* The page was outsourced. A language model’s native product is English prose; writing is the easy subject to hand over. In a high-school mathematics study, students with an unguarded chatbot improved assisted practice and then scored lower on an unaided exam; they did not perceive the harm.[^141] That study is math, not a writing trial. The same crutch is available here, and harder to hear, because a finished-looking paragraph has no answer key. Current performance on a page with a model at the elbow is often an unreliable index of learning.[^142] The unaided draft is the index you want. *What to say.* “Read this out loud. Stop at any sentence you would not have written. We are going back to your outline. You write the paragraph. I will look at *your* sentences when you are done.” ### Five-minute parent warm-up Before the lesson, on paper, no student in the room: 1. Write one fact about a book or a topic you actually know. Under it, write a thesis a reasonable person could disagree with. Star the second sentence. 2. Take two short kernel sentences and combine them three ways. Pick the best. Say why in five words. 3. Write tonight’s product goal in one sentence you could check with a pencil: claim, support, and (if the genre wants it) one other view. If those three are fluent for you, you can hear a fact wearing a thesis hat. That is the job. The student still holds the pencil. ## How to teach it this week Run the door before you call the year Composition. **The sentence.** Can the student write a complete sentence, then another that is not a clone, then a because-clause or a so-clause that actually changes the meaning? A student who still produces fragments and run-ons as the default, and cannot combine two kernels on purpose, is not ready for a four-page essay, regardless of calendar grade. Stay at the sentence. Model. Shared write. Independent. Sentence combining in short bursts. Then return. **The paragraph.** Can they state a claim, give one piece of support, and finish? If the “paragraph” is a summary with no claim, or a claim with no support, the essay will only be a longer version of that hole. **The unaided habit.** Can they produce a short draft with the book open and the model closed? If every good page so far appeared after a tool pass, you do not yet know what they can do. If the sentence is fragile, the earlier writing chapters of this book are the repair, not a pep talk. Then return. High school is a band, not a birthday. When the gates are open, here is the shape of a sitting. Warm-up. Short model. Student attempt. One good question. Mixed practice. Exit ticket. The English hour in the front of this book already taught this shape. This chapter does not reinvent it. It fills the shape with composition. **Warm-up (3–5 minutes, unaided).** Yesterday’s sentence move, not tonight’s new genre. Two kernels to combine. One “fact or thesis?” item. One sentence from last week’s paragraph to tighten. No tool. No audience yet. **Short model (5–8 minutes).** One fully worked paragraph on paper, thinking aloud. Name the product goal first. Then: “Here is my claim. Here is the bit of the text I am using. Here is why that bit is allowed to support the claim. Here is one other view, in a clause, so I am not pretending I am the only person in the room. Here is my last sentence, which is not a restatement of the first.” Keep it short. You are showing a schema, not delivering a lecture. Fade as soon as you can. Second example: they explain what your sentence is doing. Third: topic sentence given, they write the support. Fourth: they write from an outline they made. Include at least one incorrect example per new move: a fact-as-thesis, a quote dumped without a because-link, a conclusion that only repeats. Ask “what did they think this sentence was for?” “They forgot to try” is not a diagnosis. “They treated the first sentence as a topic label, so there was nothing to prove” is. **Exact wording you can steal.** On a thesis: “A thesis is a sentence someone could reasonably disagree with. Read yours out loud. Could I argue with it, or can I only nod because it is a fact?” On evidence: “Point to the sentence in the book. Now say it in your words. Then, if you quote, keep it short and attach it to your sentence. A quotation sitting alone is not a reason.” On a warrant: “Because *why*? The because cannot be a restatement of the evidence. The because is the rule that lets this evidence count.” On a counterclaim: “Name one other view in a clause. You are not starting a quarrel. You are showing you have heard it.” On combining: “Two short sentences. Glue them three ways — *because*, *but*, *so*, or a who/which/that. Pick the one that says what you mean.” On the shared write: “Your hand on the pencil. I will say a sentence, you may change it, we write what we both can stand behind. Then you do the next sentence without me.” On fading: “I wrote the first paragraph yesterday. Today you write the twin of it from your outline. I am not going to talk during the attempt.” On the chatbot temptation, once, then stop talking about tools: “The draft I will grade is the one you write with the model closed. We can look at *your* sentences afterward.” **First problem for the student, after the model.** A paragraph, not an essay. Product goal on a sticky note: claim, one piece of evidence from a page they have actually read, one because-sentence, a finish. They make a three-line outline first. Then they write. You wait. Struggle before rescue. Ask, wait, hint, then model a sentence if needed. You do not finish the paragraph. **One good question.** After the attempt: “Which sentence is doing the claiming, and which sentence is doing the proving?” The student generates the why. You do not narrate it. **Student attempt, then mixed practice.** Two or three paragraphs of the new type, unaided, after the fade — not five essays. Then mix: yesterday’s informative paragraph, last month’s narrative scene, a sentence-combining trio, one “fact or thesis?,” one letter or short correspondence if that is on your map. A purchased book that is almost entirely blocked — twenty identical “write a thesis” drills — can still be interleaved by you pulling yesterday’s informative task onto the same page. **Exit ticket (5–10 minutes, unaided).** One short draft without notes from a tool and without a rewrite pass. Book open if the task is about a text; model closed. You grade the unaided page. An AI-polished paragraph is performance. The unaided sentence the next morning is the test. College Board still requires the student to produce unaided work in an AP room. CLEP’s essays are typed in a testing window, not at a kitchen chatbot. Every essay the transcript will claim needs an unaided draft the model did not open. That is the year, not a special integrity week. **How to fade help.** Complete model → student explains a second → shared write → completion problem → conventional paragraph → incorrect example named → student generates. When they can write the paragraph without your topic sentence, put the scaffold away. Bring it back when the next genre arrives, or when fact-as-thesis comes back. **When to stop talking.** After one model and one question. The pencil stays with the student. A parent who last wrote a paper years ago can still hear a fact in thesis clothing. Hearing is not the same as taking over the page. If spoken fluency still outruns the pencil, partnership on the shared write is honest: they talk the sentence, you may scribe *their* words for one round, then the pencil goes back. That is a fade, not a career as the family typist. Keyboarding is a production skill on the maps — one page in a sitting by the grade-4 cluster, two by grade 5, three by grade 6.[^143] By this band, typing a draft is normal. Typing is not thinking. The outline still comes first. ## Practice that actually builds learning Understanding and practice grow together. Practice a new genre this week, with a model and a why. A stack of “write an essay” with no strategy is assignment, not teaching. **Blocked, for a new move.** The day a student first writes a counterclaim clause, a short blocked set is honest: several paragraphs that all want that move, so execution can form. Same for a first because-but-so expansion, a first outline that is more than a list of topics, a first informative paragraph that explains a process rather than arguing. **Mixed, for when to use it.** Later the same week, and next week, the new skill reappears between unlike jobs. A page that is only “write a persuasive paragraph” teaches execution. A page that mixes “is this a fact or a thesis?,” “combine these kernels,” “write the argument paragraph,” “write the informative twin,” and “write a four-sentence summary with no extra opinion” teaches choosing. **Brief retrieval of moves already known.** Sentence boundaries, a because-clause, last month’s quotation-attachment, the product goal from Tuesday. A short mixed oral at the start of the hour is enough: “Fact or thesis: *Water boils at 100°C at standard pressure.* Now make it a thesis about boiling water in this kitchen.” Fluency here is quick, flexible control of a move they can already get right — not a race to fill pages. **One incorrect example to diagnose.** Show a paragraph whose first sentence is a fact, or a quotation with no because-link, or a conclusion that only repeats. Ask what they thought the sentence was for. Then an isomorphic paragraph the student writes unaided. **Compare methods after execution exists.** After they can write an argument paragraph from an outline, put an informative twin next to it on the same topic, then a short narrative that *shows* the same idea. “Which job is this prompt asking for, and how can you tell?” Flexibility is a third outcome, not a luxury. Kitchen, money, and making can motivate or give a second picture. A letter to a grandparent is real correspondence. A lab report in science is real informative writing. They do not replace the unaided English paragraph. A doubled recipe will not teach a warrant. Set a product goal every sitting. Read the draft aloud — the student reads, not you — because the ear catches the run-on the eye forgives. Then change one sentence by combining or subordinating. That is the day’s revision. A full rewrite every night is how the tool gets invited in. One improved sentence, on purpose, is instruction. Monthly: an unannounced mixed sitting the student does without notes and without a tool. One paragraph, one combining trio, one fact-or-thesis. That is the retrieval the year needs, not a stack of identical worksheets the night before. A reader helps. In the adolescent literature, peer assistance while writing averaged 0.75.[^144] At home the “peer” is often you, or a sibling, or a co-op partner. The job is to be a reader: “I cannot tell what you want me to believe.” It is not to become a co-author. You do not grab the pencil. You do not finish the paragraph. ## For the student This year your sentences have to carry a thought a reader was not in the room to hear. A *topic* is what you are writing about. A *thesis* is what you want the reader to believe about it — a sentence a reasonable person could disagree with. “The river froze” is a fact. “The freeze, not the council, ended the ferry” is a thesis. Your job is to prove the second kind of sentence. A paragraph is a small essay. One claim. Support — a bit of a text, a fact you can point to, a scene. A because-link that is not a restatement. A finish. An essay is several of those, in an order a stranger can follow. You will write arguments and explanations more than stories this year. Stories still matter. They are not the whole job. Nobody is born knowing this. It is taught. You will watch a paragraph get made. You will write one with a parent or teacher. Then you write one alone. The one that counts is the one you write alone. **Tiny worked example.** Prompt: Is the narrator of this chapter a person we should trust? Outline, three lines: - Claim: We should not take this narrator at his word. - Evidence: He tells two different times for the same door slamming. - Because: A person who cannot keep a small fact straight is not a safe guide to a large one. Paragraph: We should not take this narrator at his word. In the first telling, the door slams “before supper”; two pages later the same slam is “well after dark.” A person who cannot keep a small fact straight is not a safe guide to a large one. He may still be interesting. He is not a witness. The first sentence can be disagreed with. The middle points at the page. The because-sentence is the rule. The last sentence does not repeat the first in new clothes; it limits it. **Try 1.** Take these two kernels: *The lantern went out. The path disappeared.* Combine them three ways. Pick the best for a tense scene. If all three sound the same, you have only glued words, not chosen a relationship (*when*, *so*, *and with them*). **Try 2.** Write one unaided paragraph from an outline *you* made, about a page you have actually read. Product goal: a thesis someone could argue with, one piece of evidence in your own words, one short quotation if you need it, one because-sentence. Read it aloud. Then change one sentence by combining or subordinating. Do not paste it into a rewriter. **Explain it back.** In two sentences, tell someone the difference between a topic and a thesis, and how you would check a proposed thesis without writing the whole essay first. **Challenge.** Take a fact you know — a real one, from science or history or a book in the house. Write it as a fact. Under it, write three different theses that could be argued from it. Star the one you could actually support this week with a page you have. Then write that paragraph, unaided. If you cannot find a page to support it, the thesis is a wish. Choose a smaller one. You are allowed to struggle. A fact wearing a thesis hat is information, not a verdict. Circle the claiming sentence. Point to the proving sentence. Read it out loud. The tool does not take this test. You do. ## If it isn’t clicking Three diagnostics, and the next move for each. No shame. Placement is by skill, not by birthday. **Diagnostic 1 — the sentence is still the gate.** The student writes fragments as the default, or run-ons, or three short kernels that never touch. Combining produces *The river froze the town lost its ferry* with no punctuation and no relationship. Handwriting or typing is so slow that a paragraph never arrives. *Next move.* Pause the essay. Return to sentence construction: kernels, combining, because-but-so, boundaries. If the pencil itself is the problem, handwriting or keyboarding practice is the composition work this week; writing “places multiple simultaneous demands on the writer,” and fluency with those production skills frees attention for composing.[^145] A student who is calendar-grade 9 with sentences still fragile is not “behind a legal grade.” The student is at the door. Stay there until a paragraph with a claim and support is accurate unaided. Then this course. **Diagnostic 2 — a procedure without a picture.** The student can chant “introduction, three reasons, conclusion” and cannot say what the claim is. Can fill a worksheet on topic sentences and writes a fact as the thesis on a bad day. Can produce a long file after a tool pass and cannot read it aloud as theirs. *Next move.* Slow down on the new genre. One conceptual pass: thesis versus fact, evidence versus restatement, because-link versus “because I said so.” Then a handful of paragraphs, not an essay. Then a return to the distinction. Put the outline next to the paragraph and ask them to match. If they are already lost inside a first argument paragraph, start with a fully worked example, not with “see if you can figure it out.” Explicit teaching — a clear model, practice, think-aloud, feedback — is the lever the writing studies keep finding. Missing sentence skill is not a chance to discover. It is a fluency problem. If the fluent page appeared in twenty minutes with diction the student does not own, treat it as an outsourced draft, not as a talent. Go back to the outline. They write. You look at their sentences. **Diagnostic 3 — choosing is the hole, not executing.** The student can write an argument when the page says “argue,” an explanation when the page says “explain,” and freezes on a mixed prompt, or writes the same three-reason chant for a lab report, a letter, and a literary paragraph. *Next move.* Keep going on new content in short blocked bursts, and change the practice. Mix yesterday’s type with last month’s. Ask “which job is this?” before “where is the thesis?” An unannounced mixed sitting once a month. Pull a letter, a paragraph about a poem, and an informative explanation onto the same week’s page. Choosing is a skill. Blocked pages will not teach it. **When to get a human tutor.** You are the ceiling on tonight’s idea, and preparing the five-minute warm-up still leaves you unable to hear a fact in thesis clothing. Or the student has the sentence gate and weeks of faded models and mixed practice, and the unaided ticket is still a blank. A tutor, a co-op teacher, or a dual-enrollment composition course later is a normal high-school plan. Stay the supervisor who can still hear a missing warrant on the exit ticket, even if someone else taught Tuesday’s lesson. Dual enrollment, where a family can arrange it, is a college course with a college transcript; it is common as an *option* in public high schools, not a homeschool default, and not a substitute for documenting the student’s own writing.[^146] **When to go ahead.** Sentences are accurate unaided. A paragraph can hold a claim and support. The student can combine kernels on purpose. One genre is fluent and a second is in progress. The unaided ticket is mostly right, including on mixed jobs. Then lengthen: two paragraphs, then a short essay, then a paper with sources (the next writing-heavy chapter). Acceleration is a pathway choice, not a neighborhood race. Diagnose. Then move. ## Tools, including AI Optional helpers for you. The student attempts first. You hold the key. > **AI box — Composition** > > In a high-school math study, an unguarded chatbot raised assisted practice and then cut the unaided exam; writing is even easier to outsource, because the model’s native product is a paragraph. The exit ticket is a draft the model did not open. > > Use a tool to explain today’s idea to *you* (why a fact is not a thesis; why a quotation needs a because-link), to make extra kernels and fact-or-thesis items with answers hidden from the student, to write a hint script that withholds the finished paragraph, or to diagnose a draft already written. > > The student writes tonight’s paragraph unaided. No pasting the prompt for an essay, no “fix this to an A,” no unsupervised chatbot during the attempt, no photograph of a worksheet for a key. > > A detector is not a verdict on a child’s writing; the tools that have been measured mislabel human work. Read the unaided page instead. > > After the attempt, you may paste *their* sentences into a parent account and ask for questions, not a rewrite. The next sitting still begins with their pencil. Spell-check and a word processor are production tools, in the same family as the 0.55 for word processing in that writing literature.[^147] They are not a thesis teacher. Use them after the unaided draft exists. ## What “done enough” looks like A “grade 9 book” is a publisher’s scope, not a legal grade. Done enough is skill. **Checklist before moving on.** - Writes complete sentences on purpose, and can combine two kernels three ways and pick one. - Distinguishes a topic from a thesis; hears a fact in thesis clothing. - Writes a paragraph with a claim, support from a page actually read, and a because-link that is not a restatement. - Can name, in a clause, one other view when the genre wants it. - Revises by combining, subordinating, or cutting, not only by adding length. - Produces an unaided draft with the model closed; can read it aloud as theirs. - On a mixed page, chooses argument, informative, or narrative on purpose, not only the chant in the worksheet title. - Has at least several unaided paragraphs and one short unaided essay in a folder a stranger could skim. What can still be thin: a full documented paper with a standard citation format (that is the research chapter); rhetorical analysis as a named genre (some maps wait; Texas names it in English III); AP-style timed sophistication. Those can travel. The unaided paragraph with a real thesis cannot. **Transcript.** Write **English 9** or **English 10** or **Composition**. If the spine was a publisher or a writing-method name, add a parenthetical a stranger can map: “English 9 (composition; argument and informative essays).” Four English credits remain the typical high-school pattern.[^148] A year of sentence work that never reached a paragraph is still work; name it honestly rather than printing Composition on a folder of worksheets. If the student sits AP English Language or CLEP College Composition, put the exam score on a separate line. The kitchen course is still English. Authenticity is content, not a cover. A path through English is the promise. A chatbot paragraph, a percentile, or a brand is not. A parent who felt rusty should be able to say, after this chapter: I know a fact from a thesis. I know what to say tomorrow. I know what a fluent page that is not theirs usually means. The student still does the struggle. We can do this. # Chapter 5 ## Literature A stack of cloth-bound books, an open volume, and a brass desk lamp. ## Why this matters A literature year is how you teach a book, not how you reprint a nation’s reading list. The student reads what is on the table, learns to say what happens, how it is made, and what it is saying, and writes about it from the page — not from a study-guide brand and not from a model that has never opened the physical book. That is worth the struggle because almost everything later hangs on it. Composition in the last chapter asked for a claim and a because-link; literature is the same job pointed at a story, a poem, a play. Research in the next stretch will ask the student to handle a source they did not invent; a novel is already that kind of source. College literature, if it comes, will assume the student can read imaginative writing and write about it without a packet of “themes.” A year that watches the film, skims a summary, and fills character charts is literature in name. Name the year by the books that were actually read. There is no federal canon. The Common Core map, which many families never use and which this book does not treat as a homeschool law, names mythology, foundational U.S. documents, and Shakespeare as examples of content it mentions. It does not enumerate a required list, and it says plainly that the standards “do not—indeed, cannot—enumerate all or even most of the content that students should learn.”[^149] Virginia’s 2024 English standards say the state “does not provide a list of prescribed books” and still want graduates exposed to foundational authors such as Homer and Shakespeare, among works from different genres, cultures, geographies, and times.[^150] Florida prints sample texts by benchmark in its own standards, plus literary periods and a civic-literacy list; those are Florida’s samples, not a United States reading list.[^151] Texas names the legal high-school courses English I, II, III, and IV. English I already asks students to read and respond to American, British, and world literature *inside one course*. English III asks them to read and analyze American literature across literary periods; the credit is still English III.[^152] Steal the cluster. You do not have to steal a list. Typical U.S. high school is four English credits, not a federally named American–British–World sequence.[^153] American Literature, British Literature, and World Literature are **transcript titles** a stranger can read, and they are a true description of the books on the table this year. They are not a law that junior year must be American and senior year British. A family that reads a mixed list every year, and transcripts English 9 through 12 with the books named in the description, has done the same job. A family that titles a year American Literature because the shelf was Emerson, Douglass, Dickinson, and a novel of the Harlem Renaissance has also done that job. Local school catalogs sometimes subtitle English 11 “Survey of American Literature.” That is catalog language, not a national sequence. AP English Literature is not AP English Language. Literature is college-level literary analysis of fiction, poetry, and drama; Language is college composition and the rhetoric of nonfiction.[^154] Neither is required. If the student sits an exam, the score is a separate line. Close reading, in this chapter, is a move: slow down on a stretch of text, notice how it is made, and say what the making does. It is not a franchise, not a seminar brand, and not a 173-question product. Some families use a Socratic list sold as a literature method. The questions can help. The course name is still the books. What this year depends on is already in place: decoding and fluency, so the page can be read; the middle-school habit of pointing at a line of text; enough knowledge of the world to recognize a myth, a bit of scripture, a war, a custom, when a later work alludes to it. A student who still guesses through the word is not “behind in literature.” That student is still in a code program, whatever the birthday. A student who can decode every word and does not know the world of the book is not reading it either. Reading is the product of decoding and linguistic comprehension; if either factor is zero, reading is zero.[^155] Teach the book. Teach the world of the book. Both. Transcript the year as **English 9**, **English 10**, **American Literature**, **British Literature**, **World Literature**, or another title a registrar already has a slot for. Never write a brand in the title column. Never write “Critical Thinking 101.” The books belong in the description: titles, authors, what was written about them. The door is the book in front of you. Can the student retell what happens without the packet? Can they point to a sentence that is doing work and say, in their own words, what that sentence does? Can they write a paragraph about the book with the book open and a summary site closed? If those three are fragile, stay there. The rest of this chapter teaches the year. This page names why a real book is worth a hard season, and why a national list you print from memory is not the teaching. ## For the parent: understand it yourself You do not need to be a literature professor. You do need to understand today’s idea well enough to hear a plot summary offered as analysis, and a quotation the student cannot find in the book, not as cute slips. **Everyday picture.** Think of watching a sport you know. A double play is one event for you and a confusing sequence of throws for someone who does not know baseball. The same broadcast, two different games. A book is like that. Knowledge of the world of the book — the custom, the war, the myth, the house, the joke — is not extra credit. It is how the sentences mean. Junior-high students who knew baseball understood a baseball passage better than “good readers” who did not, even when the passage was written at an early fifth-grade level.[^156] You do not need a historian’s training to teach *this* book. You need enough of its world that an allusion is not a brick wall, and you need to notice when the student is standing in front of one. **Precise picture.** Three questions work on any story, play, or narrative poem. They are not a brand. What happens? Plot, situation, who wants what, what stands in the way, how it ends. If the student cannot retell this, stop. Everything later will be fog. How is it made? The writer’s choices: who is telling it, from where, in what order, in what kind of sentence, with what images, what pattern of repetition, what scene against what scene. A metaphor is a making. A chapter break is a making. A joke the narrator does not get is a making. What is it saying? Not “the theme is friendship,” as a poster. What claim about people, or power, or God, or luck, is this made thing urging, questioning, or laughing at? A reasonable person could disagree. That is how you know you have a claim and not a topic label. Those three are the year. Close reading is how you answer the second and third without abandoning the first. You pick a stretch — a paragraph, a sonnet, a page of dialogue — and you stay there long enough to see the joints. You mark a sentence that does work. You say what the work is. Then you step back and ask how that stretch fits the whole. Close reading is not a franchise that replaces finishing the book. A student who can annotate a stanza and cannot say what the poem is about has performed a ritual. A student who can retell the plot and cannot point to a line has performed a different ritual. You want both. Citing the text is the same move you already used in composition: a claim, a bit of the page, a because-link. “The narrator is not to be trusted because he gives two times for the same door” is literary argument. “Shakespeare uses imagery” is a topic label. “This symbolizes hope” with no line is a guess wearing academic clothes. Poetry, short fiction, a play, and a novel are not four franchises. They are four ways of making. A sonnet holds an argument in a small room; a play holds it in bodies and speech; a novel can take a year of a life and still be made of scenes. Mix the genres across the year. AP Literature, if you use it as an optional overlay, recycles short fiction, poetry, and longer fiction or drama rather than marching through a national author list.[^157] CLEP Analyzing and Interpreting Literature does not require specific works.[^158] Your kitchen list can be chronological, mixed, paired with history, or a year of living books. None of those is national law. All of them still require the three questions. Knowledge of the world of the book is not a content-free “reading strategy” hour. Strategy instruction has a place and a ceiling: a handful of sessions may confer the same benefit as fifty, if the mechanism is noticing that one should monitor, question, and connect. Building knowledge does not have the same upper limit.[^159] Teach a brief, modeled pass — “what is this sentence doing?” — in the text on the table. Spend the year on the books and the worlds they assume. A packet of “find the main idea” with no book is not this course. Allusion is why the world matters twice. A later work that borrows a flood, a garden, a tragic hero, a founding speech, or a fairy-tale plot is counting on the student to have met the first thing. Florida names a comparative and adaptation strand; you do not need Florida’s samples to see the job.[^160] You need one earlier work still in the house when the later one arrives. That is a reason to read some older books. It is not a reason to print a canon and call it federal. Hallucinated quotations are a new wrong answer with an old name: making it up. Language models invent titles, pages, and critic attributions. Until you have checked the line in the physical book (or a reliable edition you actually opened), treat a model-supplied quotation as untrusted. The student points at the page. The page is the authority. ### Wrong answers you should be able to hear **1. Plot summary offered as the whole of the work.** The student retells the chapter, neatly, and stops. Asked what the chapter is doing, they retell it again with “the author is trying to show” taped on the front. *What it usually means.* Question one was answered. Questions two and three were never asked, or were asked as a theme worksheet (“friendship, courage, love”) that can be filled without the book. *What to say.* “That is what happens. Now point to one sentence that is doing extra work — a comparison, a joke, a repeated word, a lie. What does that sentence do that a plot summary misses?” **2. A symbol hunt with no page.** “The river symbolizes freedom.” “The bird is hope.” No line. No because. *What it usually means.* The student has learned that literature class rewards big nouns. The claim is not yet attached to making. *What to say.* “Show me the sentence. If we cannot find the river doing something in the language — repeating, changing, sitting next to a locked gate — we do not get to upgrade it to Freedom. Say what the sentence does first.” **3. The book was not read.** A fluent paragraph about a novel, with diction that jumped a grade, and the student cannot find the scene. Or the film was the book. Or a summary site was the book. *What it usually means.* The work was outsourced. A model can produce a literature essay that looks finished. It cannot sit the student in front of the page. Performance on a packet is not the same as having read the chapter.[^161] *What to say.* “Open to the scene. Read me the sentence you are using. If it is not on this page, we start the paragraph again from the book.” **4. “What the author meant,” with nothing to point at.** The student announces the author’s intention, or a critic’s name they cannot source, or a historical fact that is not in this year’s notes. *What it usually means.* Guessing has been upgraded to authority. Biography and history can help — they are part of the world of the book — but they do not replace the sentences. *What to say.* “We can ask what the made thing is doing on the page. Intention is a guess unless we have a letter or a preface in our hands. What is this stanza doing, whether or not we know the writer’s mood that morning?” **5. The year is a list, not a course.** Twenty titles named, three finished, no writing about any of them. Or a chronological American-to-British march treated as if a registrar in another state required it. *What it usually means.* Coverage has replaced teaching. Four English credits are typical; a federally named American–British–World sequence is not.[^162] Texas already mixes American, British, and world literature in English I. A short list finished and written about beats a famous list abandoned in October. *What to say.* “We will finish this book. We will write about it. Next book. The transcript will tell the truth about what we read.” ### Five-minute parent warm-up Before the lesson, on paper, no student in the room, book open: 1. In two sentences, say what happens in the stretch you will teach today. 2. Copy one sentence that is doing work. Next to it, write five words naming the work (comparison, delay, joke, reversal, lie). 3. Write one claim about the stretch a reasonable person could disagree with. Star it if you could support it from that same page. If those three are fluent for you, you can hear a summary wearing an analysis hat. That is the job. The student still holds the book. ## How to teach it this week Run the door before you call the year literature. **Can they read the page?** Fluency is accurate, reasonably paced, expressive reading of text the student can already decode — not a race.[^163] If tonight’s chapter is still a decoding fight, this is not a literature problem. Return to the code and to fluency on easier connected text. Listening to you read a harder book aloud can still build the world of the book. That is knowledge. It is not independent reading of that book yet. **Can they retell?** After a scene, with the book closed for a minute, can they say what happened in order, with the people named? If not, do not ask for symbolism. Ask for the story again, maybe with a quick sketch of who wants what. **Can they point?** Finger on a line. “This sentence does X.” If they cannot find a line, they are not yet in close reading. They are in conversation about a memory of the book. When the gates are open, here is the shape of a sitting. Warm-up. Short model. Student attempt. One good question. Mixed practice. Exit ticket. The English hour already taught this shape. This chapter fills it with a book. **Warm-up (3–5 minutes, unaided).** Yesterday’s stretch, not tonight’s new chapter. Two-sentence retell. One “what is this sentence doing?” on a line they have already seen. One fact-or-claim about the book (composition’s ear, now pointed at literature). No summary site. No tool. **Short model (5–8 minutes).** You read a short stretch aloud — a paragraph, a sonnet, a page of dialogue — with the student following. Then you think aloud on one marked sentence: “Here is what happens in this bit. Here is the sentence I marked. It compares X to Y, or it repeats a word, or it lets the speaker lie. That making lets the book do Z. My claim, which you could argue with, is \_\_\_\_. The because-link is this sentence on this page.” Keep it short. You are showing the three questions, not delivering a lecture on the history of the novel. Fade as soon as you can. Second stretch: they explain what your marked sentence is doing. Third: you mark nothing; they mark one sentence and talk. Fourth: they write the paragraph from the book, unaided. Include at least one incorrect example per new move: a theme poster with no line; a symbol upgrade; a fluent paragraph whose quotation is not in the chapter. Ask “what did they think counting as evidence?” “They didn’t read” is not a diagnosis until you have checked. “They treated a topic word as a claim, so there was nothing to prove from the page” is. **Exact wording you can steal.** On retell: “Book closed for one minute. What happens in this scene, in order? Names, please, not ‘the guy.’” On a marked sentence: “Read it out loud. What extra work is it doing — comparison, delay, joke, reversal, lie, a word that keeps coming back? If the answer is ‘describing,’ we have not started yet.” On a claim: “Say a sentence about this scene I could reasonably disagree with. ‘There is a river’ is a fact. ‘The river is the only honest voice in the chapter’ is a claim. Which did you just say?” On knowledge: “What does a reader have to know about the world to get this joke, this insult, this prayer? If we don’t know it, we look it up *from a source we can name*, then we come back to the sentence.” On allusion: “Have we met this story before — a flood, a garden, a test, a trickster? If yes, what is this book doing with the old story: repeating, twisting, mocking, blessing?” On the film or the summary: “Those can come after. Today the page is the book.” On a model-supplied line: “Find it in our edition. If it is not here, it is not evidence.” **First problem for the student, after the model.** Read a scene (or a poem, or a short scene of a play). Mark one sentence that does work. In their own words, say what that sentence does. Book open. Model closed. Then, if the sitting has time, turn that into the paragraph you already know how to teach: claim, bit of the page, because-link. **One good question.** After the attempt: “If we deleted your marked sentence, what would the scene lose?” The student generates the why. You do not narrate it. **Student attempt, then mixed practice.** Two or three marked-sentence jobs in the same book, unaided, after the fade. Then mix: a retell of yesterday, a paragraph of literary argument, a four-line paraphrase of a stanza with no extra opinion, one “what world-knowledge does this line need?,” a short comparison if two works are alive in the house. A purchased anthology that is almost entirely “answer these ten questions” can still be interleaved by you pulling yesterday’s poem onto the same day as today’s chapter. **Exit ticket (5–10 minutes, unaided).** Book open, notes from a tool closed. One marked sentence plus two or three sentences of their own saying what it does, or a short paragraph of literary argument. You grade the unaided page. A study-guide paragraph is performance. The sentence they can find and explain is the test. **How to fade help.** You read and mark → they explain your mark → they mark and talk → they write → they write about a stretch you have not modeled. When they can mark and explain without your sentence, put the heavy scaffold away. Bring it back when a new genre arrives — the first sonnet, the first act of a play, the first unreliable narrator. **When to stop talking.** After one model and one question. Struggle before rescue. Ask, wait, hint, then model a sentence if needed. You do not finish the paragraph. You do not turn the hour into a lecture about the author’s biography. Biography is world-knowledge when you need it for a line. It is not the course. Reading load is a kitchen fact, not a national minute count. A novel wants chunks: a chapter or a scene most days, with one slower close-reading sitting in the same week, and writing about the book before the next book starts. A poem wants rereading more than coverage. A play wants some of it aloud, because the line is speech. If you are teaching two ages, the younger one can listen above decoding level while the older one reads; that is knowledge for the younger and literature for the older. It is not two full courses in one hour unless you pretend. ## Practice that actually builds learning Understanding and practice grow together. Practice the three questions on this week’s stretch. A packet of theme words with no book is not practice. **Blocked, for a new move.** The week a student first writes a literary paragraph from a marked sentence, a short blocked set is honest: several stretches in the same book that all want that move. Same for a first sonnet, a first scene of dialogue, a first comparison of two tellings of the same story. **Mixed, for when to use it.** Later the same week, and next week, the new skill reappears between unlike jobs. A page that is only “identify the theme” teaches a ritual. A page that mixes “retell in five sentences,” “mark the sentence,” “write the claim paragraph,” “what does a reader have to know?,” and “is this narrator safe?” teaches choosing. **Brief retrieval of what the book already did.** Names, the last reversal, a line you marked Tuesday. A short oral at the start of the hour: “Who wanted what yesterday? What stood in the way? What sentence did you mark?” Knowledge that is not retrieved fades, and then the next chapter has no world to sit in. **One incorrect example to diagnose.** Show a paragraph that never quotes or paraphrases a line, or that quotes a line not in this chapter, or that upgrades a bird to Hope in the first sentence. Ask what they thought counted as evidence. Then an isomorphic sitting: same job, their book, unaided. **Compare after execution exists.** After they can do the three questions on a short story, put a poem next to it, then a page of a play. “What changes because this is speech meant to be heard?” Flexibility is a third outcome, not a luxury. Comparing two works is a later move; it fails if neither work was actually read. Kitchen and life can motivate. A family story next to a folk tale is a real comparison. Travel next to a setting is a real world-knowledge boost. They do not replace the unaided page. Visiting a river will not teach a metaphor by itself. Read some of it aloud, even in high school. Guided oral reading with feedback is how fluency was shown to move; silent independent reading as the only method did not have the same causal evidence in that literature.[^164] At this age the point is not a fluency program. The point is the ear: verse, dialogue, a narrator’s joke. The student should hear the line they are about to mark. Write about the book before you start the next one. Composition’s product goal still applies: a claim someone could argue with, support from a page actually read, a because-link. Extra writing alone is a small lever; a paragraph with a strategy is the literature work.[^165] Journals can help memory. They are not the analytical paragraph. Monthly: an unannounced sitting, book open, tool closed. Retell a stretch from two weeks ago. Mark a sentence in a stretch from this week. Write the short paragraph. That is the retrieval the year needs, not a stack of study-guide packets the night before a “test” the packet already answered. ## For the student A literature year is you and a book. Not a film, not a summary, not a paragraph a tool wrote in the voice of a student. You read. You notice how it is made. You say what it is saying, with a finger on the page. Three questions, any story: 1. What happens? 2. How is it made? 3. What is it saying — a claim a reasonable person could disagree with, not a poster word like “friendship”? Close reading is slowing down on a stretch until you can answer 2 and 3 without dropping 1. You mark one sentence that does extra work. You say what the work is. You will understand more if you know the world of the book. A joke about a king falls flat if you do not know what a king was for. That is not cheating. That is reading. If you do not know, find out from a source you can name, then come back to the sentence. **Tiny worked example.** Stretch: a narrator says the door slammed “before supper,” and two pages later the same slam is “well after dark.” What happens? A door slams. The teller cannot keep the time straight. How is it made? The book gives two clocks for one sound. That is a making. It is not a typo we politely ignore. What is it saying? We should not take this narrator at his word. A person who cannot keep a small fact straight is not a safe guide to a large one. The claim is arguable. The evidence is on the page. The because-link is the rule about small facts and large ones. You already wrote this kind of paragraph in composition. Literature is that paragraph about a made thing. **Try 1.** Open tonight’s stretch. Mark one sentence that does extra work. In the margin, five words: what is the work? If you write “describing,” mark a different sentence. **Try 2.** Book open, tool closed. Write five sentences: two for what happens, one that copies (or tightly paraphrases) your marked sentence, two for what that sentence does. Then, if you can, add one because-link and you have a paragraph. **Explain it back.** In two sentences, tell someone the difference between what happens in a scene and what a sentence in that scene is *doing*. Use a real line from a book in the house. **Challenge.** Take a line that alludes to something older — a flood, a garden, a famous speech, a fairy tale, a war everyone in the book seems to know. Name the older thing. Say whether this book is repeating it, twisting it, mocking it, or blessing it. Point at the line. If you cannot name the older thing, that is the world-knowledge job for this week: find it, then come back. You are allowed to struggle. A summary is information, not a verdict. A missing line is a reason to open the book, not a reason to open a tool. The page is the test. You take it. ## If it isn’t clicking Three diagnostics, and the next move for each. No shame. Placement is by skill, not by birthday. **Diagnostic 1 — the page is still the gate.** The student cannot decode the chapter, or can decode it so slowly that meaning never arrives, or refuses the book and lives on summaries. *Next move.* Tell the truth about the gate. If decoding or fluency is the hole, this is not a literature-list problem. Return to connected text they can already get right, with oral rehearsal and feedback, and read harder books *aloud* so knowledge does not wait. A calendar-grade 10 student who cannot read tonight’s novel independently is not “behind in British Literature.” They are at the door. Independent literature waits until the page can be read. Listening can still be the year’s knowledge. The transcript should describe the work that was actually done. If the hole is refusal plus fluent summaries, remove the summary route for a stretch. Shorter book. Daily sitting. Retell before anything else. The film after, not instead. **Diagnostic 2 — a ritual without a picture.** The student can fill “the theme is courage” and cannot point to a line. Can annotate every noun as a symbol. Can produce a fluent essay whose quotations are not in this edition. *Next move.* Slow down. One stretch. One marked sentence. Talk before writing. Put the book next to the paragraph and match each claim to a page. If they are already lost inside “analysis,” start with a fully worked example on a short stretch, not with “see what you notice.” Close reading is a modeled move. It is not a personality. Invented quotations: open the book. If the line is not there, the paragraph starts over. A tool is not a critic you cite. **Diagnostic 3 — coverage is the hole, not thinking.** The student can do the three questions on a short story you taught slowly and falls apart when the year is a race through a famous list. Or they can talk beautifully at the table and never write the paragraph. *Next move.* Shorten the list. Finish the book you are in. Write about it. Mix a poem into the novel week so genre-choosing gets practice. Keep the writing attached: the literature paragraph is composition pointed at a book. If talk outruns writing, use a shared write for one paragraph, then they do the twin alone. Extra discussion will not produce the unaided page. **When to get a human tutor.** You cannot hear the difference between a summary and a claim, even after the five-minute warm-up. Or the student has the reading gate and weeks of marked-sentence practice, and the unaided ticket is still a blank. A tutor, a co-op seminar, or a later dual-enrollment literature survey is a normal high-school plan. Stay the supervisor who can still ask “where is that on the page?” even if someone else led Tuesday’s discussion. AP Literature, if you want the overlay, is an exam and a course description, not a requirement; credit is by college.[^166] **When to go ahead.** The student can read the year’s books with fluency that leaves room for thought. Retell is accurate. A marked sentence plus a short unaided paragraph is mostly right, including on a stretch you did not pre-teach word by word. One genre is fluent and a second is in progress. Then lengthen: a full essay on a finished book; a comparison once two works are really known; a play read aloud with a second sitting for how the speech is made. Diagnose. Then move. ## Tools, including AI Optional helpers for you. The student reads first. You hold the book. > **AI box — Literature** > > Use a tool to explain the *world* of the book to *you* (what a reader is assumed to know about a war, a custom, a myth), to generate extra “what is this sentence doing?” items from a paragraph *you* typed in, with your answers hidden from the student, to write a hint script that withholds the claim, or to diagnose a paragraph already written. > > The student reads the page and writes from the page. No summary of the unread chapter, no “write my essay on this novel,” no unsupervised chatbot during the attempt, no invented quotation treated as evidence. > > A model-supplied line, title, or critic is untrusted until it matches the book in your hands. Find it or drop it. > > Literary-character chat, if you try it at all, stays on a parent account, time-boxed, and after the chapter is read — a brief interview, not a companion. The book remains the authority. A film, an audiobook, and a well-made lecture can build the world of the book. They are second pictures. They do not replace the unaided sitting with the page. Text-to-speech is a documented help for some students with reading disability when the job is access to content; it is not a literature method that removes the need to read the year’s books.[^167] ## What “done enough” looks like A “grade 11 American Literature” spine is a publisher’s or a local catalog’s scope, not a legal grade and not a federal sequence. Done enough is skill, plus books actually read. **Checklist before moving on.** - Reads the year’s books (or hears them read, if decoding is still the gate) with enough fluency that thought has room. - Retells a stretch in order, with names. - Marks a sentence that does work and says, in their own words, what it does. - Writes an unaided paragraph of literary argument: claim, page, because-link. - Uses knowledge of the world of the book on purpose, and can say when a line requires it. - Treats plot, making, and meaning as three questions, not as a theme poster. - Checks quotations against the physical book. - On a mixed sitting, can move among a poem, a scene of prose, and (if you taught one) a page of a play. What can still be thin: a full research paper on a critic (that is the next chapter’s documented paper); every period in a survey textbook; AP-timed sophistication. Those can travel. The unread famous list cannot count as done. **Transcript.** Write **English 9** or **English 10** or **American Literature** or **British Literature** or **World Literature**, matching what you actually read. If the year mixed regions and centuries, English 10 with a description of the books is more honest than a title you borrowed from a catalog. Texas English I already mixes American, British, and world literature in one legal course; you may too.[^168] Four English credits remain typical.[^169] If the student sits AP English Literature or CLEP Analyzing and Interpreting Literature, put the exam on a separate line. Safer kitchen practice is a readable title plus the score as an artifact, not a brand of seminar as the course name. A parent who felt rusty should be able to say, after this chapter: I know the three questions. I know what a summary wearing analysis clothes usually means. I know what to say tomorrow, with the book open. The student still does the reading. We can do this. # Chapter 6 ## Language A navy dictionary, a handwritten note, a red pencil, and wire-rimmed glasses. ## Why this matters Language, on the maps states actually test, is grammar, usage, mechanics, and vocabulary — in the sentence on the table, not as a worksheet island. It is the chapter that keeps two popular errors from running the year. One error is “grammar does not matter.” The maps say it does, and so do the conventions items on SAT Reading and Writing and on ACT English. The other error is “a year of parts-of-speech worksheets will make them write.” Isolated grammar instruction of that kind, as tested against writing quality, lost. Sentence combining, which is grammar in use, did not. That is worth the struggle because a readable sentence is the gate to every later page. Composition already asked for a claim; the claim still has to be a sentence. Literature already asked for a marked line; the student’s own line still has to hold together. Research will ask for paraphrase and citation; those are language jobs too. A year of circling predicate adjectives, with no change in the draft, is language in name. Name the year by the sentences that improved. The Common Core Language strand, which many families never use and which this book does not treat as a homeschool law, is the shape many other maps rhyme with: L.1, grammar and usage when writing or speaking; L.2, capitalization, punctuation, spelling; L.3, knowledge of language for effect, variety, and consistent style; L.4–6, word meaning, figurative language, academic vocabulary.[^170] Texas puts conventions inside composition — editing drafts — with a grade-by-grade list, and gives vocabulary its own expectation every year, including Greek and Latin in the elementary years and foreign phrases in high school (English I examples: *bona fide*, *caveat*, *carte blanche*; English III: *ad hoc*, *faux pas*, *non sequitur*, *modus operandi*).[^171] Virginia keeps Language Usage as a K–12 strand beside Writing, and spelling beside handwriting in the early grades.[^172] Florida tests Following Conventions and keeps Vocabulary as its own strand K–12.[^173] Steal the cluster. You do not have to steal a workbook. Tests of this strand are not the strand. On the digital SAT, Standard English Conventions is about 26 percent of Reading and Writing, Expression of Ideas about 20 percent, Craft and Structure — much of it words in context — about 28 percent.[^174] ACT English (enhanced format) reports Production of Writing, Knowledge of Language, and Conventions of Standard English.[^175] Those are signals. They are not a reason to turn the kitchen into a test-prep island. A student who can edit *this* sentence, combine two kernels, and learn words from the book in front of them is already in the work those items sample. Steve Graham and Dolores Perin, in the adolescent writing meta-analysis you already met, found strategy instruction 0.82, sentence combining 0.50, and grammar instruction −0.32 on writing quality.[^176] Grammar in that analysis was explicit teaching of parts of speech and sentence types, often against a treatment that itself taught writing. The minus sign means those grammar conditions lost to the comparison, not that no one should ever learn a participle. In the elementary meta-analysis, grammar instruction did not produce a statistically significant positive effect; teaching transcription skills — handwriting, spelling, sentence construction — averaged 0.55.[^177] A federal writing guide says the same thing in kitchen language: writing “places multiple simultaneous demands on the writer,” and fluency with those production skills “allows students to devote more of their attention to composing.”[^178] This is a useful pattern, not a promise that every home will see the same number. Then teach. English spelling is not a chaotic list and not a moral failing. Richard Venezky described the system as mapping first onto a morphophonemic level — units of meaning and allowable letter patterns — and then into sound.[^179] Louisa Moats, writing for teachers, calls it a morphophonological alphabetic system. About half of English words, in the Hanna spelling survey she cites, can be spelled from sound-symbol correspondences alone; another 34 percent would have only one error on that basis (84 percent mostly predictable); about four percent are truly irregular.[^180] Spelling and reading “build and rely on the same mental representation of a word.”[^181] Invented spelling is a useful stage in the years when the code is new. It is not a destination. A high-schooler still inventing *they* and *said* is not being creative. They are missing a taught word. Vocabulary is morphology plus knowledge-rich reading, not a weekly list of ten rare words as the whole program. Direct teaching of some words helps, especially words about to appear in a text. Copying definitions is the comparison condition that loses. Successful programs, as that literature is summarized for teachers, teach a few hundred words a year thoroughly, not a thousand shallowly.[^182] The rest of the year is books, history, science, and talk in full sentences about things in the world. Complex oral language — grammar, defining words, listening comprehension — predicted later reading comprehension much more strongly than simple vocabulary tests did.[^183] A quiz of ten SAT words, disconnected from anything being studied, is not that mechanism. Transcript the language work *inside* English 9, Composition, or Language Arts. There is no separate national credit called Grammar. Four English credits remain typical.[^184] Conventions belong in the description: sentences edited, patterns taught, words from the year’s books. Never write a brand in the title column. The door is the sentence in front of you. Can you hear a run-on, a dangling modifier, a subject and verb that do not match? Can the student combine two kernels on purpose? Can they spell the patterns already taught, and learn a new word from a page they are actually reading? If those are fragile, stay there. The rest of this chapter teaches the strand. This page names why it is worth doing in the draft, and why a worksheet pile is not the same job. ## For the parent: understand it yourself You do not need to be a linguist. You do need to understand today’s idea well enough to hear a run-on, a modifier attached to the wrong noun, and a weekly list offered as a vocabulary program, not as cute slips. **Everyday picture.** Think of a tool in a drawer versus a tool in the hand while you build a shelf. A labeled saw in a catalog is not carpentry. A saw used on this board is. Grammar terms are labels. They earn their keep when they help you and the student fix *this* board — this sentence — and then build the next one on purpose. A year of labeling tools, with no shelf, is the worksheet island. A year of shelves with no names at all is the other ditch: you cannot say “the modifier is attached to the wrong noun,” so you shrug and write “awkward” in the margin. **Precise picture.** A sentence is a complete thought with a subject doing or being something (a predicate). Two complete thoughts glued with a comma and no conjunction make a comma splice; glued with nothing, a run-on. A fragment is a piece waving as if it were a sentence. Subject and verb match in number: *The list of reasons is long*, not *are long*, because the head of the subject is *list*. A modifier sits next to what it modifies; dangling, it attaches to the wrong noun and the sentence accidentally says the sidewalk was walking. Pronouns need a clear noun. Parallel ideas want parallel form: *to write, to revise, and to read it aloud*, not *to write, revising, and a loud voice*. Those are usage and mechanics. You name them in the sentence on the table. You do not need a term quiz first. If a term helps — *clause*, *modifier*, *agreement* — use it here, then ask the student to fix this line. Language Progressive Skills on the Common Core map are skills that keep being expected across years; this chapter will not pretend to retype that entire chart. The teaching move does not depend on a row: edit the draft, then teach the pattern that just failed. Sentence combining is the nearby practice that moved quality (0.50).[^185] Two kernels: *The lantern went out. The path disappeared.* Combined: *When the lantern went out, the path disappeared.* Or: *The lantern went out, so the path disappeared.* Or: *The path, once the lantern went out, disappeared.* The student produces three, picks one, and says why. That is grammar as craft. It is also how SAT Expression of Ideas and ACT Production of Writing sample the same skill without calling it combining. Spelling is the same code as reading, looked at from the other side. Position patterns (*hope / hopping*), inflections (*ed*, *s*, *ing*), syllable types, and then Latin and Greek morphemes (*struct*, *spect*, *photo*) are the sequence Moats has long recommended as professional judgment, not as a new trial.[^186] *Sign* and *signal* share a morpheme; the *g* is not a decoration. *Electric / electricity* keeps the *c*. Students who must think too hard about spelling use up the attention composition needs; poor spellers may restrict what they write to words they can spell.[^187] Spell-check is not spelling instruction. In one study Moats cites, checkers caught 30 to 80 percent of misspellings overall, and identified the intended word from students-with-LD misspellings only 53 percent of the time.[^188] The student still has to produce a spelling the tool can recognize, and still has to know when the tool has guessed wrong (*definitely* / *defiantly*). Vocabulary has three useful piles, as Isabel Beck, Margaret McKeown, and Linda Kucan are summarized for teachers: everyday oral words (little instruction needed); domain-bound technical words (teach them in the subject that owns them); and a middle pile of wide-use words that are not automatic — *merit*, *stride*, *orient*, *concede*.[^189] Teach a few of those thoroughly, in the text where they appear. Morphology is a multiplier: if the student knows *ject* as throw, *reject*, *inject*, and *projection* are not three unrelated spelling demons. The rest of vocabulary is knowledge: the baseball passage is easier if you know baseball. A list of ten rare words, tested on Friday, forgotten on Monday, is not a program. It can be a small supplement. It cannot be the year. Foreign phrases on some high-school maps (*non sequitur*, *ad hoc*) are the same job with borrowed clothes: meaning plus the sentence they appear in, not a decoration quiz. Figurative language is L.5 work and literature work at once: a metaphor is a making you already marked in the last chapter. ### Wrong answers you should be able to hear **1. A clean grammar quiz and an unchanged draft.** The student can circle adverbs and still writes *Walking to school, the backpack bounced* as if the backpack were walking. *What it usually means.* The grammar hour never touched the sentence they were writing. In Graham and Perin, that kind of instruction lost on writing quality (−0.32). The label was learned. The craft was not. *What to say.* “We will name the problem in this sentence. Who is walking, in the words you actually wrote? Fix this one. Then write the next one on purpose.” **2. “Grammar does not matter. They’ll pick it up.”** Or the opposite chant: “Until every part of speech is named, they may not write.” *What it usually means.* The first chant erases L.1, conventions items, and the reader’s ear. The second chant postpones composition until a worksheet island is finished. Neither is what the maps or the writing studies describe. *What to say.* “We need readable sentences. We will teach the pattern that just failed, in this draft. We will not wait a year to write, and we will not pretend a quiz is a paragraph.” **3. The weekly list as the whole vocabulary program.** Ten rare words, a matching quiz, no appearance in anything being read. Next week, ten more. *What it usually means.* Vocabulary has been reduced to a testable pile. Direct teaching of some words helps; copying definitions loses; a few hundred thoroughly beats a thousand shallowly. Knowledge-rich reading is the rest of the engine. *What to say.* “Pick three words from this week’s book that actually do work. We will use each in a new sentence about the book. The Friday list can wait, or shrink.” **4. Invented spelling of a taught word, still welcomed.** *Thay went too the store* in a high-school paragraph, for words taught years ago, with no expectation of correction. *What it usually means.* A useful K–1 stage has been held past its job. Invented spelling should recede as studied patterns are held to standard spelling. *They*, *went*, *who*, and *said* “should not be ‘invented’ or they will be misspelled ad infinitum.”[^190] *What to say.* “That word is a taught word. Fix it. The ideas can stay. The spelling of *they* cannot still be a guess.” **5. Spell-check, or a photograph of the worksheet, as the teacher.** The student never looks at the red underline, or photographs a conventions sheet for an answer key, or lets a rewriter “fix this to an A.” *What it usually means.* Production was outsourced. Spell-check misses a large share of errors and often cannot guess the target from a wild misspelling. A vision-enabled chat will complete a grammar worksheet the way a solver completes a math sheet. The unaided sentence is the test. *What to say.* “Write it first. Then we look at *your* sentence. A tool may explain a pattern to me, or make extra kernels with the answers hidden. It does not take tonight’s edit.” ### Five-minute parent warm-up Before the lesson, on paper, no student in the room: 1. Write one run-on, one dangling modifier, and one subject-verb mismatch. Then fix each. Star the word you would point to first. 2. Combine two kernels three ways. Pick the best. Name the relationship (*when*, *so*, *although*). 3. From a book in the house, copy one sentence and underline a word a student might not own. Write a five-word meaning in context, not a copied dictionary line. If those three are fluent for you, you can hear the wrong turn in tonight’s draft. That is the job. The student still holds the pencil. ## How to teach it this week Run the door before you call the year a language program. **Can they hold a sentence?** If fragments and run-ons are the default, combining and editing *are* the writing hour. Do not assign an essay to a student who cannot yet hold a sentence.[^191] **Can they spell the patterns already taught?** If not, spelling of those patterns is not optional extra. It is transcription skill, and it frees working memory for composing. **Can they learn a word from a page?** If every new word requires a matching quiz and never appears in speech or writing, the list is the whole program. Shrink the list. Grow the book. When the gates are open, here is the shape of a sitting. Warm-up. Short model. Student attempt. One good question. Mixed practice. Exit ticket. Language work is often a short block inside the English hour, attached to the draft or the book, not a second subject with its own 180-day island. **Warm-up (3–5 minutes, unaided).** Yesterday’s pattern, not tonight’s new term. Two kernels to combine. One sentence to edit (run-on, agreement, or modifier — mixed). One taught spelling. No tool. **Short model (5–8 minutes).** One student-like sentence on the board (an illustration, not a portrait of your child). Think aloud: “I hear two complete thoughts and only a comma. That is a splice. I can make two sentences. I can add *and* or *but*. I can make one a *when*-clause. I pick the *when*-clause because the time relationship is the point.” Then combine two kernels the same way. Keep it short. You are showing a repair, not a lecture on the history of the comma. Fade: they name what your sentence was doing; they fix a twin; they combine unaided; they edit a sentence in *their* draft. Include one incorrect example: a “fixed” sentence that only added a comma to a splice, or a combine that smashed two thoughts with no relationship. Ask what they thought the comma was for. **Exact wording you can steal.** On a run-on: “How many complete thoughts? Where does the first one end? You may not glue them with a comma alone.” On agreement: “What is the head of the subject — the one word the verb has to match? Ignore the prepositional phrase in between.” On a dangling modifier: “Who is doing the -ing? Is that who the rest of the sentence is about? If not, the opener is attached to the wrong noun.” On combining: “Two short sentences. Three glues. Pick the one that says the relationship you mean — time, cause, contrast, or extra information.” On spelling: “Show me the morpheme we already know. *Sign* is still in *signal*. The *g* is not a decoration.” On vocabulary: “Read the sentence. What would still make sense in this slot? What would not? Now say the word in a new sentence about this same page.” On the draft: “We will mark three errors of the pattern we are teaching, not thirty of everything. You fix those three. Then you read it aloud.” On fading: “I edited the first sentence out loud. You edit the next two. I am not going to talk during the attempt.” **First problem for the student, after the model.** Two kernels to combine three ways; pick the best; say why. Then, if a draft exists, one pass for the day’s pattern only. You wait. Struggle before rescue. You do not grab the pencil. You do not rewrite the paragraph in the margin in your adult voice. **One good question.** After the attempt: “Why did you pick *although* instead of *so*?” The student generates the why. You do not narrate it. **Student attempt, then mixed practice.** Several combining or editing items of the new type, unaided, after the fade. Then mix: yesterday’s agreement, last month’s comma splice, a spelling of a taught morpheme, a word-in-context from the book, one sentence from their own draft. A purchased grammar book that is almost entirely blocked — thirty identical underlining drills — can still be interleaved by you pulling a sentence from Tuesday’s essay onto the same page. **Exit ticket (5 minutes, unaided).** Three items, mixed, tool closed: one combine, one edit, one word from this week’s reading used in a new sentence. You grade it. Current performance with the rule box at the elbow is often an unreliable index.[^192] The unaided ticket is the index you want. **How to fade help.** Full model of a repair → they explain → they fix a twin → they combine or edit unaided → they apply the pattern in a draft you did not seed. When they can hear the splice without the term on the board, put the term away. Bring it back when the next pattern arrives, or when the splice comes back. **When to stop talking.** After one model and one question. Three marked errors of one pattern beats a page bleeding with every mistake you can name. The second kind of page teaches despair, not craft. Handwriting and typing remain production skills; if the pencil is still the bottleneck, that is this week’s language work, not a character flaw. Moats’s professional-judgment dose for spelling is 15–20 minutes daily or 30 minutes three times a week, applied in writing.[^193] That is a teacher’s recommendation, not a national minute law and not a homeschool trial. Short, frequent, in the words they actually write, beats a long Friday packet. ## Practice that actually builds learning Understanding and practice grow together. Practice the new pattern this week, in sentences, with a why. A stack of circling drills with no draft is the island. **Blocked, for a new move.** The day a student first repairs a dangling modifier, a short blocked set is honest: several sentences that all want that move. Same for a first combine with *although*, a first morpheme family (*struct* / *construction* / *instruct*), a first word-in-context trio from the week’s book. **Mixed, for when to use it.** Later the same week, and next week, the new skill reappears between unlike jobs. A page that is only “underline the subject” teaches a label. A page that mixes “combine,” “fix the agreement,” “spell the morpheme,” “use *concede* about this paragraph,” and “edit this sentence from Monday’s draft” teaches choosing. **Brief retrieval of patterns already taught.** Splices from last month, *they* / *said* / *because*, a morpheme family, three words from last week’s book. A short mixed oral at the start of the hour is enough. Fluency is quick, flexible, accurate control — not a race and not optional. Time only items they can already get right untimed. **One incorrect example to diagnose.** Show a “combine” that is still two sentences with a comma, or a spell-check acceptance of *defiantly* for *definitely*, or a vocabulary sentence that could fit any word in the list (*The king was very \_\_\_\_\_*). Ask what they thought the job was. Then an isomorphic item unaided. **Compare after execution exists.** After they can combine with *because*, put *although* and a relative clause next to it on the same kernels. “Which relationship did you mean?” After they can spell *sign*, put *signal* and *signature* next to it and ask what stayed. Flexibility is a third outcome, not a luxury. Kitchen and life can motivate. A recipe is a sequence of imperatives; a text message is a register. They do not replace the unaided sentence. Shopping will not teach agreement. Apply every pattern in a real draft the same week you teach it. That is the whole argument of this chapter in one sitting. Extra worksheets without that application are the −0.32 condition in spirit, even if you never read the paper. Monthly: an unannounced mixed ticket, tool closed. Combine, edit, spell, word-in-context. That is the retrieval the year needs. If you sample SAT Reading and Writing or ACT English, pull a few Conventions and Expression items *after* the pattern has been taught in the draft, as a check, not as the week’s method. Those items are short; they reward an ear already trained on real sentences. They will not teach a dangling opener if the draft never contained one. Keep the two publisher benchmarks in their lane: SAT Reading and Writing 480 and ACT English 18 are readiness numbers, not the meaning of the Language strand, and they are not equated to each other.[^194] A personal dictionary of frequent misses — taught words that still slip — is more honest than a new list of rare words. The student adds the word, a morpheme if it has one, and a sentence from this week’s book. That notebook is retrieval. It is not a vocabulary program by itself. ## For the student Language is how your sentences hold together, how your spellings map to meaning, and how words from books become words you can use. It is not a pile of circling. It is not “grammar does not matter.” You will name a problem in a sentence you actually wrote, fix it, and write the next one on purpose. A sentence is a complete thought. If you have two, you need a legal way to join them or a period to separate them. A comma alone is not a legal way. Combining is a skill: take two short sentences and make one that says how they relate — time, cause, contrast, extra information. You pick the relationship. That is craft. Spelling is not chaos. *Sign* sits inside *signal*. The extra letters often mark a family of meaning, not a trick. Words you have already been taught — *they*, *said*, *because* — are not still for inventing. Vocabulary lives in what you read. A Friday list can help a little. It cannot replace the book. When you meet a word doing work on the page, use it in a new sentence about that same page. **Tiny worked example.** Kernels: *The lantern went out. The path disappeared.* Three combines: - *When the lantern went out, the path disappeared.* (time) - *The lantern went out, so the path disappeared.* (cause) - *The path disappeared although we still had our feet on it.* (wait — that third one smuggled in extra meaning we did not have. Throw it out. Third legal try: *The path that we needed disappeared when the lantern went out.*) Pick the first if the moment of going-out is the point. Pick the second if cause is the point. The “best” one is the one that says what you mean, not the longest one. Edit example: *Walking to school, the backpack bounced.* Who is walking, in the words on the page? The backpack. That is the dangling modifier. Fix: *Walking to school, I felt the backpack bounce.* Or: *The backpack bounced as I walked to school.* **Try 1.** Combine these kernels three ways: *The river froze. The town lost its ferry.* Pick the best. Say why in five words. **Try 2.** In a sentence you wrote this week, find one problem — a splice, an agreement miss, a modifier, a taught spelling. Name it. Fix it. Then use one word from this week’s reading in a new sentence about that reading. **Explain it back.** In two sentences, tell someone why a comma alone cannot join two complete thoughts, and what you would try instead. **Challenge.** Take a morpheme you know (*struct*, *spect*, *ject*, *photo*, or another from your book). Write three real words that carry it. For each, a sentence that uses the word about something you actually studied this month. If you can only write a fill-in-the-blank (*The \_\_\_\_ was good*), you do not own it yet. Change the sentence until the word is doing work. You are allowed to struggle. A splice is information, not a verdict. Fix this sentence. Write the next one on purpose. The tool does not take this test. You do. ## If it isn’t clicking Three diagnostics, and the next move for each. No shame. Placement is by skill, not by birthday. **Diagnostic 1 — the sentence is still the gate.** Fragments and run-ons as the default. Combining produces a smash with no relationship. Handwriting or typing so slow that editing never happens. Taught words still invented. *Next move.* Pause the worksheet island if you are on one. Return to kernels, combining, boundaries, and spelling of taught patterns, applied in short daily writing. Transcription skill is the composition work this week. A calendar-grade 10 student with sentences still fragile is not “behind in Language.” They are at the door. Stay there until an unaided paragraph holds as sentences and taught words are spelled. Then more sophisticated patterns. **Diagnostic 2 — labels without repair.** The student can name a participle on a quiz and cannot find the dangling opener in their own first sentence. Can define *concede* and cannot use it about this week’s book. Can run spell-check and still miss *defiantly*. *Next move.* Teach in the draft. One pattern. Three marks. They fix. Then a handful of isomorphic sentences. Then back to the draft. If they are lost, start with a fully worked repair, not with “see if you can find the errors.” Missing fluency with a pattern is not a chance to discover. It is a modeled move plus practice. Shrink the vocabulary list to words in this week’s reading. Morphology over trivia. **Diagnostic 3 — choosing is the hole, not executing.** The student can fix agreement when the page says “agreement,” combine when the page says “combine,” and freezes on a mixed ticket or in a real paragraph where no one has labeled the error. *Next move.* Keep new patterns in short blocked bursts, and change the practice. Mix yesterday’s type with last month’s. Pull a sentence from their literature paragraph onto the language ticket. An unannounced mixed sitting once a month. Choosing is a skill. Blocked pages will not teach it. **When to get a human tutor.** You cannot hear the splice or the dangling opener even after the five-minute warm-up. Or the student has weeks of faded models and mixed practice, and the unaided ticket is still a smash. A tutor, a co-op teacher, or extra time on sentence construction inside a writing method you already trust is a normal plan. Stay the supervisor who can still hear the wrong turn on the exit ticket. If difficulty with accurate spelling and decoding is unexpected, persistent after systematic teaching, and out of line with the student’s oral language, a specialist evaluation is a next move; this book does not diagnose anyone.[^195] **When to go ahead.** Sentences hold unaided. Combining is a choice, not a smash. Taught patterns are spelled in drafts. Words from the year’s books show up in speech and writing. One new pattern is fluent and a second is in progress. The mixed ticket is mostly right. Then add the next pattern — a more careful modifier, a more honest parallel, a new morpheme family — without dropping retrieval of the last. Diagnose. Then move. ## Tools, including AI Optional helpers for you. The student attempts first. You hold the key. > **AI box — Language** > > Use a tool to explain today’s pattern to *you* (why *list of reasons is*, not *are*; why the opener attaches to the next noun), to make extra kernels and edit-items with answers hidden from the student, to write a hint script that withholds the fixed sentence, or to diagnose a sentence already written. > > The student writes and edits unaided. No photograph of the worksheet for a key, no “fix this to an A,” no unsupervised chatbot during the attempt. > > Spell-check is a pass after the unaided draft, not the lesson. It will miss errors and sometimes guess the wrong word. > > A detector is not a verdict on a child’s writing. Read the sentence instead. A grammar-practice site can sequence items. It does not replace you hearing the splice in the draft. Facts of usage are not a chat job in the student’s hands. Generate a short oral script if you want. Then you and the student, out loud, no screen between the sentence and the ear. ## What “done enough” looks like A “grade 9 grammar” book is a publisher’s scope, not a legal grade. Done enough is skill in the draft. **Checklist before moving on.** - Writes complete sentences on purpose; hears a splice and a fragment. - Combines two kernels three ways and picks one for meaning. - Edits for agreement, modifiers, and pronoun clarity in *their* sentences, not only on a labeled worksheet. - Spells taught patterns and taught high-frequency words in drafts; uses morphemes as families, not as decorations. - Learns words from knowledge-rich reading and can use a few thoroughly in new sentences; a weekly list, if it exists, is not the whole program. - Reads a draft aloud and catches some errors by ear. - On an unaided mixed ticket, chooses a repair without the worksheet title doing the choosing. - Conventions items on a practice SAT or ACT, if you sample them, are a check, not the course. What can still be thin: every row of a progressive-skills chart; every foreign phrase on a high-school list; perfect spelling of rare words. Those can travel. The unaided readable sentence cannot. **Transcript.** A separate course title called Grammar is only honest if you truly ran a separate half-credit and can describe it. Usual practice: **English 9** or **Composition** or **Language Arts**, with conventions and vocabulary in the description. Four English credits remain typical.[^196] NCAA still wants English years, not a brand of workbook.[^197] A parent who felt rusty should be able to say, after this chapter: I can hear a run-on and a dangling opener. I know combining is the nearby practice that helps. I know what to say tomorrow, in this sentence. The student still does the repair. We can do this. # Chapter 7 ## Research and rhetoric A navy reference book, handwritten index cards, a manila folder, and paperclips. ## Why this matters Research and rhetoric, at the kitchen table, are one job with two names. The student makes a claim, finds a source that actually exists, says why that source is allowed to support the claim, and gives credit in a format you can keep consistent. The documented paper is that job at full length. It is not a new subject invented in eleventh grade, and it is not a course titled Critical Thinking 101. That is worth the struggle because college, work, and civic life will assume it. Composition already taught the paragraph with a because-link. Literature already taught pointing at a page. This chapter adds *other people’s pages*, a citation the reader can follow, and — on some maps — a name for the appeals and the cheap tricks a text can use. A year that prints a bibliography a chatbot invented, or that never leaves a single website, is research in name. Name the year by papers whose sources you could open. Research starts early on every map this book checked. The Common Core writing strand, which many families never use and which this book does not treat as a homeschool law, puts short research in grade 1 (build knowledge about a topic) and keeps it through high-school sustained projects; students gather sources and take notes, list sources by grade 3, paraphrase and list by grade 5, and use a standard citation format by the 6–8 and 9–12 bands. Drawing evidence from texts — W.9 — begins in grade 4.[^198] Texas starts inquiry in grade 1, names primary and secondary sources in grade 2, and asks for a bibliography and the difference between paraphrasing and plagiarism in grade 4; high-school English then adds credibility, bias, named fallacies, and academic citations.[^199] Virginia’s research strand runs K–12: a topic and a question with prompting in kindergarten, title-and-author credit by grade 3, relevance, reliability, and citations in 4–5.[^200] Florida’s researching-and-using-information ladder runs from recall to synthesize, with “literary research” using primary and secondary sources in grade 11.[^201] Steal the cluster. You are not starting from zero in ninth grade. You are making the old habit independent. Rhetoric is not a Common Core strand. Argument writing and the reading of arguments are. Florida introduces rhetorical appeals in grade 6 “so that students will have an understanding of the appeals – logos, pathos, and ethos – when starting argumentative writing.”[^202] Texas asks English I students to explain rhetorical devices such as understatement and overstatement and the effect of logical fallacies such as straw man and red herring; English III adds rhetorical analysis as a composition genre and further fallacies.[^203] AP English Language is the college-equivalent home of rhetorical situation, claims and evidence, reasoning and organization, and style — reading and writing, no prerequisite course.[^204] CLEP College Composition puts Ability to Use Source Materials at about 25 percent of the multiple-choice work and Rhetorical Analysis at about 25 percent; the synthesis essay must cite two provided sources. MLA, APA, and Chicago are all named as in-scope on that page.[^205] Those overlays sit on top. They are not the kitchen method, and they are not required for a homeschool transcript. Four English credits remain typical. NCAA’s sample English description includes a short research paper; CLEP is not an NCAA core course.[^206] Transcript the year as **English 9**, **English 10**, **Composition**, or a readable title plus a description that says “documented paper; sources cited.” Never write a brand in the title column. Never write “Critical Thinking 101.” Evaluation lives inside the English work. The door is a claim with a real page behind it. Can the student state a claim someone could disagree with? Can they point to a source they actually opened, paraphrase it, and attach a because-link that is not a restatement? Can they write a title and an author that match the object in their hands? If those are fragile, stay there. The rest of this chapter teaches the paper. This page names why the paper is worth a hard month, and why a logic textbook is not the course. ## For the parent: understand it yourself You do not need to be a debate coach or a librarian. You do need to understand today’s idea well enough to hear a quotation dump offered as an argument, and a bibliography no one in the house could reopen, not as cute slips. **Everyday picture.** Think of a disagreement at the table about whether to take a jacket. “Because I said so” is a claim with no source. “Because the thermometer on the porch reads 42” is a claim with evidence. “Because 42 degrees is cold enough, on this wind, that you will regret short sleeves by the end of the block” is the because-link — the *warrant* in ordinary language. If the thermometer is broken, or is in the sun, the evidence is weak even if the because-link is honest. Research is that jacket argument with pages instead of a porch thermometer. Rhetoric is noticing whether the speaker is offering reasons, tugging feelings, or asking you to trust them — and whether a cheap trick has replaced a reason. **Precise picture.** A *claim* is an arguable sentence. *Evidence* is a bit of the world that could, in principle, be checked: a paragraph, a table, a scene, a measurement, an eyewitness. A *warrant* is the rule that lets this evidence count for this claim. “The article reports two failed replications; a finding that fails twice is not yet a safe guide to policy” is a warrant. Restating the evidence is not. Restating the claim is not. “Because I found it online” is not. A *source* is an object you opened. Primary sources are close to the event or the making: a letter, a speech, a novel, a data table the researchers published, a photograph. Secondary sources talk about those objects: a history, a review, a textbook chapter. Both are usable. Mixing them up is a problem only when the paper pretends a textbook *is* the speech. A *source card* (paper or a simple digital note you can print) holds the bits you will need later: author, title, where it lives, the date you opened it, one paraphrase in the student’s words, one short quotation if needed, and a one-line note on why this source is in the paper. A bibliography copied at the end from a chatbot is not a stack of source cards. It is a list of objects that may not exist. Citation is a trail, not a decoration. Pick one format — MLA, APA, or Chicago — and keep it for the year. The maps say “standard format” or “academic citations,” not a national style. MLA is common in English papers; APA in psychology and education; Chicago in history. Colleges will teach their preference. Consistency and honest page numbers beat a mash of three styles. Paraphrase is putting the idea in words the student would actually say, then giving credit. Plagiarism is presenting someone else’s sentences or structure as the student’s. Changing every third word is not paraphrase. A missing citation on a paraphrase is still a missing trail. Rhetoric, in kitchen language, is how a made argument tries to work. Florida’s three appeals, when you use that map, are ordinary: *logos* as reasons and evidence; *pathos* as feeling; *ethos* as “why trust this speaker.” A piece can use all three without being a trick. A piece that is only feeling, or only “trust me,” with no checkable reason, is thin. A speech that asks you to pity a person *and* shows a number you can check is doing two jobs at once; you can name both. A speech that asks you to pity a person *instead of* showing anything checkable is thin, even if it is moving. The student should be able to say which job is on the table, in ordinary words, before they reach for a Greek label. Named fallacies — straw man, red herring, ad hominem, slippery slope, loaded language, post hoc, circular reasoning — are cheap tricks with names some maps want the student to know. A straw man is answering a weaker version of the other view than the other person actually said. A red herring is changing the subject. Ad hominem is attacking the speaker instead of the claim. Slippery slope is treating a first step as if a disaster were already proved. Loaded language is stacking the feeling in the words so the reason never has to work. Post hoc is treating “after” as “because.” Circular reasoning is using the claim as its own because. Teach them when they appear in the reading, or as a short named set if you are on Texas’s English I or III list. A year of fallacy worksheets with no paper is a logic island. This book will not become that island. One clause in the student’s paragraph — “this advertisement changes the subject rather than answering the cost” — is enough. Then they still have to make their own case. Evaluating a website is a leave-the-page move. Fact checkers, in the civic-online-reasoning work associated with Sam Wineburg and Joel Breakstone, open new tabs to see who else is talking about an unfamiliar site rather than lingering on the site’s own About page; that habit is called lateral reading.[^207] One sentence is enough: leave the page, search the name of the site or the author, and come back. This chapter is not a sequel to a critical-thinking book. The move lives inside the source card: “I left the page. Here is who publishes this. Here is why I am still using it, or why I am not.” Hallucinated critics, titles, and page numbers are a new form of an old failure. Language models invent sources. Until the student (or you) has opened the object, it is not a source. The documented paper is a set of objects that exist. ### Wrong answers you should be able to hear **1. “We never did research until the junior paper.”** The family waits for a famous assignment, then panics. *What it usually means.* The K–2 question, the grade-4 bibliography, and W.9 were never treated as the same ladder. The paper looks like a new species. It is a longer version of a habit that should already exist. *What to say.* “We have been doing small versions of this. Tonight is one claim, one real page, one because-sentence, one citation. The long paper is several of those, in an order a stranger can follow.” **2. A quotation dump with no warrant.** Three long quotations, no because-link, a claim hiding in the last line — or not appearing at all. *What it usually means.* Evidence has been confused with argument. The student collected. They did not yet say why the collection counts. *What to say.* “Each quotation needs a sentence of yours before or after it that is not a restatement. That sentence is the because. If you cannot write it, the quotation does not belong yet.” **3. A bibliography no one can reopen.** Links that 404, books not in the house, page numbers that do not match, a critic the library has never heard of. *What it usually means.* The list was generated, not gathered. The trail is fake even if the paper sounds fluent. *What to say.* “Show me the object. If we cannot open it, it comes off the list. We will build a source card from the thing in our hands.” **4. “I found it online” as the whole evaluation.** Or a checklist of the site’s own About page, with no second tab. *What it usually means.* The student stayed on the page that wanted to be believed. Lateral reading is the leave-the-page move, then back. *What to say.* “New tab. Who is this publisher? Who else mentions this author? Then we decide whether the source stays.” **5. Fallacy-spotting as the course.** Every advertisement is a “slippery slope”; every disagreement is an “ad hominem”; no paper gets written. *What it usually means.* Names have replaced making and judging an argument. Texas wants some names. The names are tools for reading. They are not the documented paper. *What to say.* “If a cheap trick is in tonight’s article, name it in a clause. Then write *your* claim with evidence and a because-link. Spotting a trick is not the same as making a case.” ### Five-minute parent warm-up Before the lesson, on paper, no student in the room: 1. Write a claim about something you actually know. Under it, one piece of evidence from a page in the house. Under that, one because-sentence that is not a restatement. 2. Copy the title and author of that page as they appear on the object. Add the date you opened it. 3. Open one unfamiliar website related to the same topic. Leave the page: new tab, search the site’s name. Write one sentence on who publishes it. If those three are fluent for you, you can hear a fake trail and a missing warrant. That is the job. The student still finds the source. ## How to teach it this week Run the door before you call the month “the research paper.” **Is there a claim?** If the student still offers facts as theses, return to composition’s sticky note. Research will not supply a claim the writer does not have. **Is there a real page?** If every “source” is a search snippet or a chatbot list, stop. One object in the hands. Then another. **Is there a because-link?** If the paragraph is a summary of a source with no warrant, you have a book report. Teach the because-sentence on that one source before you add a second. When the gates are open, here is the shape of a sitting. Warm-up. Short model. Student attempt. One good question. Mixed practice. Exit ticket. A documented paper is a *project* across sittings, not one heroic night. Daily sittings still use the English-hour shape. **Warm-up (3–5 minutes, unaided).** Yesterday’s because-sentence, rewritten without looking. One “claim or fact?” item. One citation to punctuate from a card already made. No tool generating sources. **Short model (5–8 minutes).** You build one source card out loud from an object on the table. “Here is the title as printed. Here is the author. Here is where it lives. Here is my paraphrase in words I would say. Here is a short quotation, page numbered, that I might need. Here is why I am using it: it is close to the event, or it is a careful second look, or it disagrees with my first source and I have to say so. I left the page (or the database blurb) and checked who publishes this.” Then you write the miniature paragraph: claim, paraphrase, because-link, citation. Fade: they explain your card; they fill a card on a second object while you watch; they write the miniature paragraph unaided. Include one incorrect example: a citation whose page does not exist; a paraphrase that is still the author’s sentence; a warrant that only restates. Ask what they thought a source was for. **Exact wording you can steal.** On a claim: “What do you want a reasonable person to believe? Write that sentence first. We will not hunt sources until it exists.” On a source: “Open it. Finger on the bit you need. Now close it and say the idea in your words. Open it again and check you did not steal the sentence.” On a warrant: “Because *why*? If your because is the evidence in new clothes, you have not yet said the rule.” On credit: “Title, author, where it lives, page if there is one. Enough that I could find it tonight without you.” On leaving the page: “New tab. Search the name of the site or the author. Come back and tell me who they are. Then we decide.” On two sources: “Do they agree? If they disagree, your paper has to say so. Hiding the second source is not research.” On fallacies, when one appears: “What cheap trick is this? Name it in a clause. Then go back to your own because-link.” On the chatbot list: “If we cannot open it, it is not a source. We will start from the object in the house or the library.” **First problem for the student, after the model.** On one claim they wrote, add: the evidence in their words, one because-sentence that is not a restatement, and the title and author of the real page it came from. Book or article open. Model closed. You wait. Struggle before rescue. You do not find the quotation for them. You do not finish the paragraph. **One good question.** After the attempt: “If I hid your source, would your because-sentence still make sense as a rule?” The student generates the why. **Student attempt, then mixed practice.** Two or three miniature documented paragraphs, unaided, after the fade — one source each. Then mix: yesterday’s literature paragraph (still a source: the novel), an informative explanation with no outside source, a because-link repair, one leave-the-page check, one citation to format from a card. When two sources are honest, put them in one paragraph and ask the student to say how they relate. **The paper, as a project.** Product goal on a sticky note for the whole job, not only the sitting: a claim; at least two sources the family can reopen; paraphrase with credit; at least one because-link per source; a counterview in a clause or a paragraph; a consistent citation format; an unaided draft the model did not open. Length follows skill. A strong three-page paper with real sources beats a twelve-page paper with a fake trail. NCAA’s sample description of a short research paper is a description, not a page quota.[^208] Schedule the project as several English hours: question and claim; hunt and cards; leave-the-page checks; outline; unaided draft; revise from *your* questions, not from a rewriter. Keyboarding a paper is normal in this band. Keyboarding is not thinking. Cards and outline still come first. A question worth a paper is small enough to answer from sources you can actually get. “What is truth?” is a lifetime. “Does this town’s Saturday ferry report support restoring the Saturday run?” is a paper. History, science, literature, and a local issue are all fair game; English still owns the trail and the because-link even when the content lives in another subject. If the only sources in the house are the novel and one encyclopedia article, the paper is about those two objects. Waiting for a university library is how the month evaporates. Use what you can reopen. When the draft exists, revise like composition: read it aloud; circle the claim; match each because to a card; cut a quotation that has no sentence of the student’s own. A second source that only repeats the first is not two sources. A counterview may be a clause (“even if weekend traffic is down for other reasons”) or a short paragraph. It is not a quarrel and it is not optional once the maps ask for it. Florida’s writing ladder adds acknowledging a counterclaim in grade 7 and rebutting it in grade 8; high school keeps that control. You do not need Florida’s numbering to steal the move. **Exit ticket (5–10 minutes, unaided).** One claim, one real source, one because-sentence, title and author. Tool closed. You grade the unaided miniature. A fluent paper with unopenable sources is performance. The card plus the because-sentence is the test. **How to fade help.** You build the card → they explain → they build a card → they write the miniature paragraph → they join two sources → they outline a short paper → they draft unaided. When they can make a card without your headings, put the template away. Bring it back when a new kind of source arrives — a poem, a data table, a speech, a website that needs the leave-the-page move. **When to stop talking.** After one model and one question. You do not become the researcher. You do not turn the month into a lecture on every named fallacy. You hold the key: could we open this? Is the because a rule? ## Practice that actually builds learning Understanding and practice grow together. Practice the miniature documented paragraph this week. A stack of citation worksheets with no claim is not research. **Blocked, for a new move.** The day a student first writes a because-link that is not a restatement, several miniatures that all want that move are honest. Same for a first source card, a first leave-the-page check, a first pair of disagreeing sources. **Mixed, for when to use it.** Later the same week, mix: literature paragraph from the novel (the book is a source), informative paragraph with no outside source, one miniature with a citation, one fallacy-in-a-clause if it appeared, one “can we reopen this?” Choosing when a paper needs a source, and when the book on the table is enough, is part of the skill. **Brief retrieval.** Yesterday’s claim; the author of Tuesday’s article; the because-rule in five words. A short oral at the start of the hour. Trails that are not retrieved turn into “I think I read it somewhere.” **One incorrect example to diagnose.** Show a paragraph with a fluent critic citation whose book is not in the house, or a paraphrase that is still the source’s sentence, or a warrant that only restates. Ask what they thought counted as a source. Then an isomorphic miniature, unaided, from an object on the table. **Compare after execution exists.** After they can document one source, put a primary and a secondary on the same claim. “What can the letter do that the textbook cannot? What can the textbook do that the letter cannot?” After they can name one cheap trick, put it next to an honest appeal to feeling. “Feeling is allowed. Feeling instead of a checkable reason is thin.” Kitchen and civic life can motivate. A local issue, a family history, a science question already on the table are real inquiries. They do not replace the unaided miniature. A field trip will not teach a warrant by itself. Practice leaving the page on purpose, not only when a site “feels off.” Take a familiar publisher and a no-name blog on the same claim. New tabs for both. Who funds them? Who quotes them? The familiar name can still be wrong; the unknown name can still be a primary document. The move is the check, not a prejudice about .edu. Then back to the card. If you sample SAT Reading and Writing or ACT English, treat Information and Ideas and Production of Writing items as a check on the same skills — a short passage, a claim, a revision — not as the research course. The documented paper is still the kitchen object. CLEP’s synthesis essay, if a student ever sits it, will hand them two sources in the room; your year is how they already know what to do with two sources. Set the product goal every sitting. Read the miniature aloud. Check the trail. Then change one sentence — usually the because-link. Monthly: an unannounced miniature, object in the room, tool closed. A reader helps. You are the reader: “I cannot find this. I cannot hear the because.” You are not the co-author. ## For the student Research is how you back a claim with a page someone else could open. Rhetoric is how you notice what a made argument is trying to do — and how you make your own without cheap tricks. Neither is a separate course with a clever title. Both live in the paper you write. A *claim* is a sentence a reasonable person could disagree with. *Evidence* is a bit of a real source. A *because-sentence* (the warrant) is the rule that lets that evidence count. If your because is just the evidence again, you have not yet said the rule. A source is an object you opened. Write a card: author, title, where it lives, the bit you need in your words, a short quotation if you must, why it is in your paper. If you cannot open it, it is not a source. If a tool invented it, it is not a source. When you meet an unfamiliar website, leave the page. New tab. Search the name. See who else is talking about it. Then come back. **Tiny worked example.** Claim: *This town should restore the Saturday ferry.* Evidence (paraphrase of a real local report on the table): *The report says weekend crossings fell by half after the Saturday run was cut, while weekday crossings barely changed.* Because: *A drop that hits only the cut day is a reason to think the cut, not a general decline in travel, did the damage.* Citation, in whatever format your house is using: the report’s author, title, date, and where it lives — copied from the object. The because-sentence is not “because crossings fell.” That restates the evidence. The because-sentence is the rule about a drop that hits only the cut day. **Try 1.** On one claim you actually hold, write: the evidence in your words, one because-sentence that is not a restatement, and the title and author of the real page it came from. Open the page. Close the tool. **Try 2.** Take a website you have not used before. Leave the page. In three sentences, say who publishes it, who else mentions them, and whether it stays in your paper. Then make a source card only if it stays. **Explain it back.** In two sentences, tell someone the difference between evidence and a because-link, and how you would check a bibliography line tonight without a chatbot. **Challenge.** Two sources that disagree on the same claim — both real, both openable. Write one paragraph that lets the reader hear both, then your because-rule for why one carries more weight *for this claim*. If you cannot find a real disagreement, you are still in one-source country. Stay there until the second object is in your hands. You are allowed to struggle. A missing trail is information, not a verdict. Open the object. Write the because. The tool does not take this test. You do. ## If it isn’t clicking Three diagnostics, and the next move for each. No shame. Placement is by skill, not by birthday. **Diagnostic 1 — there is no claim, or no sentence.** The student cannot yet write a thesis, or cannot hold a paragraph, and the “research paper” is a stack of copied printouts. *Next move.* Return to composition’s paragraph: claim, support, because-link, with the book on the table as the only source. W.9 began as drawing evidence from texts in grade 4; that is still the move.[^209] A calendar-grade 11 student who cannot write that paragraph is not “behind in the junior paper.” They are at the door. The documented paper waits. Small inquiry can continue: one question, one source, oral credit (“I got this from…”). **Diagnostic 2 — a trail without a mind, or a mind without a trail.** The student can format a citation and cannot say why the source counts. Or they can talk a beautiful argument and cannot point to a page. Or the fluent paper arrived with sources no one can open. *Next move.* One object. One card. One miniature paragraph, unaided. Put the object next to the paragraph and match. If the fluent paper cannot be reopened, it is not a paper yet. Start from the objects in the house. Leave-the-page on every unfamiliar site. If they are lost inside “research,” start with your fully worked card, not with “go find five sources.” Missing skill is not a chance to discover. It is a modeled move. **Diagnostic 3 — naming is the hole, or coverage is.** The student can chant ethos-pathos-logos or a fallacy list and cannot write the because-sentence. Or they can do one miniature and collapse when the assignment is “ten sources, twelve pages.” *Next move.* Keep the names as a clause when they help. The product is the paper. Shorten the project: two sources, three pages, real trail. Mix miniatures with literature paragraphs so choosing when to go outside the book is practiced. An unannounced miniature once a month. A longer paper later, when two sources are honest. **When to get a human tutor.** You cannot hear a missing warrant or a fake citation even after the five-minute warm-up. Or the student has weeks of cards and miniatures, and the unaided ticket is still a dump or a blank. A librarian (public library staff are teachers of this skill), a tutor, a co-op, or a later dual-enrollment composition course is a normal plan. Stay the supervisor who can still ask “can we open this?” AP Language and CLEP College Composition are overlays, not requirements; credit is by college, and CLEP is not NCAA core.[^210] **When to go ahead.** Claims are arguable. Miniatures are accurate unaided. Two sources can be joined, including a disagreement. Leave-the-page is a habit. Citations match objects. Then lengthen the paper, add a named appeal or fallacy when the reading needs it, try a literary-research sitting if your map wants one. Diagnose. Then move. ## Tools, including AI Optional helpers for you. The student finds the source. You hold the object. > **AI box — Research and rhetoric** > > Use a tool to explain a citation pattern to *you*, to generate a search-query list or a blank source-card template, to write a hint script that withholds the because-sentence, or to diagnose a miniature already written. > > The student finds and cites sources. The model does not write the annotated bibliography, invent a critic, or “fill in the citations.” No pasting the prompt for the paper, no unsupervised chatbot during the attempt. > > A model-supplied title, page, or scholar is untrusted until the object is in your hands. Open it or drop it. > > After the unaided draft, you may paste *their* paragraph into a parent account and ask for questions (*Where is the because?*), not a rewrite. A library catalog and a real database are research tools. A chatbot list of “sources” is not. Detectors are not a verdict on the paper. The reopenable trail is. ## What “done enough” looks like A “grade 11 research paper” packet is a publisher’s scope, not a legal grade. Done enough is skill plus a trail you can reopen. **Checklist before moving on.** - Writes an arguable claim before hunting. - Makes a source card from an object actually opened (author, title, where it lives, paraphrase, optional short quotation, why it is here). - Writes a because-link that is not a restatement of the evidence. - Gives credit in one consistent format; paraphrase is in the student’s words. - Leaves the page to check an unfamiliar site or author, then comes back. - Can join two real sources, including a disagreement, in one short paper. - Treats named appeals and fallacies as tools in a clause, not as the course. - Produces an unaided documented draft the model did not open; every bibliography line can be reopened in the house or library. What can still be thin: a capstone-length paper; every fallacy on a Texas list; AP-timed synthesis sophistication; college-style lit reviews. Those can travel. The fake trail cannot count as done. **Transcript.** Write **English 9** or **English 10** or **Composition**, with “documented paper; sources cited” in the description. Four English credits remain typical.[^211] If the student sits AP English Language or CLEP College Composition, put the exam on a separate line. Dual-enrollment composition, where a family can arrange it, is a college transcript object as well as a homeschool line.[^212] A parent who felt rusty should be able to say, after this chapter: I know a warrant from a restatement. I know a source from a list. I know what to say tomorrow — open it, or it comes off. The student still finds the page. We can do this. # Chapter 8 ## Speaking, listening, and media A glass of water and a blank folded index card standing on a wooden table. ## Why this matters Speaking and listening are a real strand, not an assembly extra. Media that is not print — a speech heard, a news clip, a podcast, a still image used as a text — belongs in the same family. A family that never presents anything has not covered this work even if they have four literature credits. Transcript title, when you run it as its own line: **Speech**. Not a brand. Not “AI English.” That is worth the struggle because English is not only a silent page. Discussion is how a claim gets tested in real time. A presentation is composition the audience can hear. Evaluating a speaker is the same job as evaluating a paragraph, now with volume, pace, and the possibility of a cheap trick in the room. College, work, and civic life will assume the student can talk for a few minutes from notes, listen long enough to build on someone else’s idea, and say what a clip is doing. A year of four English credits with no talk is a hole. Name the year honestly. The Common Core Speaking and Listening map, which many families never use and which this book does not treat as a homeschool law, is the shape many other maps rhyme with. The college-and-career anchors stay the same across grades: collaborative discussion; integrating and evaluating information from diverse media; evaluating a speaker’s point of view, reasoning, use of evidence, and rhetoric; presenting information with supporting evidence; using digital media in presentations; adapting speech to context and task.[^213] Grade-specific standards translate those anchors. Steal the cluster. You do not have to steal a speech team. Texas folds oral language into every English course — English I already asks for presentations using informal, formal, and technical language, with eye contact, speaking rate, volume, enunciation, and purposeful gestures — *and* offers separate electives (Public Speaking I–III, Communication Applications, Debate, Journalism) that are additional credits, not replacements for English I–IV.[^214] Virginia keeps Communications and Multimodal Literacies K–12: listening and collaboration; speaking and presentation; integrating multimodal literacies from the early grades; examining media messages from grade 4.[^215] Florida’s oral-presentation ladder runs K–12 (complete sentences in kindergarten; volume; nonverbal cues; rhetorical devices in 11–12) and keeps multimedia as its own communication benchmark.[^216] Four English credits remain typical; Speech may sit inside those descriptions or as a fifth line.[^217] Either is mappable. The electives do not excuse an English year with no talk. NCTE and IRA’s 1996 professional standards, reaffirmed in 2012, still function as a check that English is not phonics-plus-test-prep: students use spoken, written, and visual language for learning, persuasion, and exchange; they read print and non-print texts; they conduct research from print and non-print sources. Those twelve statements are “not prescriptions for particular curriculum or instruction.”[^218] They are not a phonics sequence and not a speech-team syllabus. They are a reminder that the page that is not print still counts. What this strand depends on is something worth saying. Knowledge again. Academic language is not playground talk: inferential language beyond the here-and-now, narrative language with connectives and clear reference, vocabulary from the subject on the table.[^219] Listening comprehension is the ceiling on reading comprehension once decoding is adequate.[^220] A student who cannot follow a spoken argument will not magically follow the printed one. Talk in full sentences. Read aloud above decoding level. Then ask the student to stand up. This chapter will not invent a numbered effect size for “teaching oral presentation.” None was opened in the research behind this book. The sitting is still the English hour: a short model, an attempt, one good question, an unaided ticket. The ticket is a talk with notes, live or recorded, looking at a person, answering one question without reading a script a model wrote. Slides, if you use them, are a prop. They are not the speech. High-school maps already asked for argument and informative writing as the overwhelming focus of the writing year. Speaking the same jobs is not a second curriculum. It is the same claim, now audible. A student who can write the paragraph and cannot say it has finished composition and not SL. A student who can talk beautifully and cannot write the paragraph has the opposite hole; send them back to chapter 4 with the talk as a rehearsal, then the pencil. Both holes are normal. Neither is a personality. The door is a two-minute talk. Can the student say a claim out loud, from notes, with a piece of evidence, looking at someone? Can they listen to a short clip or a family member and say what the speaker wanted and what evidence they used? If those are fragile, stay there. The rest of this chapter teaches the strand. This page names why it is worth a standing-up minute, and why a slide deck is not the same job. ## For the parent: understand it yourself You do not need to be a debate coach. You do need to understand today’s idea well enough to hear a mumbled reading of a paper offered as a speech, and a personality verdict offered as feedback, not as cute slips. **Everyday picture.** Think of giving a neighbor directions at the fence, then giving the same directions to a room of people who cannot ask you to start over. At the fence you can point. In the room you need a beginning, a claim, a reason, a pace the back row can hear, and an ending that is not a shrug. Discussion is the fence: turns, building on someone else’s sentence, disagreeing without a quarrel. Presentation is the room. Media is a third neighbor — a clip, a poster, a podcast — who is not in your kitchen and still wants something from the listener. All three are English. **Precise picture.** Collaborative discussion has rules you actually use: one person at a time; look at the speaker; build on the last idea or say that you are changing it; evidence from the book or the source, not only a feeling. That is SL work from the early grades up. It is not “we chat about the chapter.” A good discussion question is the composition thesis in oral form: someone could disagree. “What happened in chapter four?” is a retell. “Should we trust this narrator after the two clocks?” is a discussion. A presentation is a spoken essay. Claim, support, because-link, a nod to another view if the genre wants it, a finish. Notes on a card — words, not paragraphs. Eye contact means looking at a person often enough that the card is a help, not a script. Volume that the back of the room (or the recording) can hear. A rate that is not a race and not a stall. Enunciation: endings of words, especially in a room that echoes. Gestures that have a purpose — a number held up, a place pointed to — not a sway that is only nerves. Adapt the speech: two minutes at the table is not ten minutes at a co-op; a toast is not a research talk. Digital media in a presentation means a map, a still, a short clip, or a recording that *does a job* the voice cannot do alone. A slide of paragraphs the student then reads is a paper taped to the wall. Evaluating a speaker is literature’s three questions plus the body: what is the claim, how is it made (including voice and image), what is it asking of us, and is the evidence checkable? Rhetoric from the last chapter lives here in the ear. A clip can offer reasons, tug feeling, and ask for trust. A clip can change the subject. The student should be able to say so in a clause, then still make their own two-minute case. Media that is not print is a text. Virginia names media messages from grade 4; Florida names multimedia in communication; the Common Core anchors ask students to integrate and evaluate information from diverse media.[^221] A still photograph used as a document, a public-service clip, a podcast interview, a campaign advertisement, a data dashboard — each has a maker, a claim, and a because (or a missing because). Teach one at a time, the way you teach a poem. A year of “the media is lying” without a clip on the table is a mood. A year of clips with no talk back is a spectator sport. Academic language is the spoken version of the sentence work in chapter 6. Connectives (*because*, *although*, *then*), precise nouns, tense that stays put, pronouns that have a home — all teachable in retells and explanations.[^222] If the student only ever answers in fragments, discussion cannot grow. Model a full-sentence answer. Wait for one. Shyness is not a diagnosis and not a waiver. Many students would rather write. Writing still counts. Standing still counts too, in a dose that matches the student: two minutes at the table to one person, then two minutes recorded, then a co-op or a grandparent on a call. Personality is not the rubric. The rubric is audible claim, evidence, notes not a script, one answered question. ### Wrong answers you should be able to hear **1. “We read so much that speech is covered.”** Four novels, four essays, no talk. *What it usually means.* SL was treated as extra. The maps keep it as a strand. Texas electives are additional credits, not a replacement for talk inside English, and not a requirement that you buy an elective.[^223] *What to say.* “This Friday is a two-minute talk from a card. The novel is the source. The talk is the work.” **2. A slide dump the model wrote.** Fluent bullets the student cannot explain. A recording of a voice that is not theirs. TTS reading the paper as “the speech credit.” *What it usually means.* The talk was outsourced. Text-to-speech is a documented help for some students with reading disability when the job is access to a page; it is not the speech.[^224] A model will happily write slides. The unaided talk is the test. *What to say.* “Card, not a script. Your words. You look at me. Then one question you answer without the card. The slides, if any, wait until that works.” **3. Feedback as a verdict on personality.** “You are so shy.” “You are a natural.” Nothing about volume, evidence, or the missing claim. *What it usually means.* The audience job was skipped. The student heard an identity, not a next move. *What to say.* “One specific note: I could not hear the ending consonants, or I did not hear a because-link. Try that one thing again. Two minutes.” **4. Discussion that is a quiz or a quarrel.** Rapid-fire “what happened,” or a family argument that never returns to the page. *What it usually means.* Retell was substituted for discussion, or heat was substituted for a claim. Both are real human events. Neither is SL practice. *What to say.* “One claim someone could disagree with. One bit of the book. Then you may answer each other — about that claim.” **5. Media critique as a separate clever course.** Every clip is propaganda; no clip is described; no student talk is recorded. *What it usually means.* Evaluation left the English work and became a mood with a fancy title. Keep it inside the talk and the page that is not print. *What to say.* “Play thirty seconds. What is the claim? What is the evidence? What is the feeling doing? Now your two-minute reply, from notes.” ### Five-minute parent warm-up Before the lesson, on paper, no student in the room: 1. Write a two-minute talk as *notes on a card* — not a paragraph — about a page in the house. Claim, one piece of evidence, one because-word. 2. Stand up and say it to an empty chair, once. Mark the place you sped up or dropped volume. 3. Write one audience sentence you could give a student: a specific note on volume, evidence, or the card, not a verdict on personality. If those three are fluent for you, you can hear a script wearing a speech hat. That is the job. The student still stands up. ## How to teach it this week Run the door before you call the year Speech. **Is there something to say?** If the book was not read and the claim does not exist, this is not a speaking problem. Return to the page. Talk cannot replace the unread chapter. **Can they speak a full sentence?** If answers are shrugs and fragments, oral sentence work *is* the sitting: retell in complete sentences, then a one-minute claim. **Can they listen?** If they cannot say what a family member just argued, a clip will not go better. Practice the ear on a two-minute human talk before a produced video. When the gates are open, here is the shape of a sitting. Warm-up. Short model. Student attempt. One good question. Mixed practice. Exit ticket. The English hour already taught this shape. This chapter fills it with a voice. **Warm-up (3–5 minutes).** Yesterday’s claim in two spoken sentences, unaided. Or a thirty-second retell of last night’s chapter. No notes for the warm-up. No tool. **Short model (5–8 minutes).** You give a two-minute talk from a card, on the book or the source already on the table. Think aloud *after*, not during: “Here was my claim in the first two sentences. Here is when I looked at you. Here is the because. Here is the other view in a clause. Here is why I did not read the card as a script.” Then play thirty seconds of a clip, if media is this week’s extra, and model one evaluation sentence: claim, evidence, feeling, checkable or not. Fade: they name what your talk did; they give a one-minute twin from their own card; they take a question. Include one incorrect example: a talk that is only plot summary; a talk read at the floor; a slide paragraph. Ask what they thought a speech was for. **Exact wording you can steal.** On notes: “Words and phrases on the card, not sentences you will read. If you need the full sentence, you do not own it yet. Practice until the card is a map.” On eyes: “Look at a face at the end of every sentence. The card will still be there when you glance down.” On volume: “I need to hear the ends of words from this chair. Try the last sentence again, louder on the last word.” On rate: “Breathe at the period. A race is not fluency. A stall is not thought.” On discussion: “Build on what they said, or say you are turning. ‘I disagree’ needs a because and a bit of the page.” On a question: “You may look at the card after you start the answer, not before. The answer is a sentence, not a shrug.” On slides: “One picture or three bullets that you could say without the screen. If the slide can give the talk without you, the slide is the speaker and you are the mouse.” On a clip: “Thirty seconds. What did they want us to believe? What did they show? What did they skip?” On TTS and tools: “Your voice. If we record, we record you.” **First problem for the student, after the model.** A two-minute talk, notes on a card, looking at a person, then answering one question without reading the card. Book or source allowed as the evidence. Script not allowed. You wait. Struggle before rescue. You do not finish the sentence for them. You do not look at the floor so they can hide. **One good question.** After the attempt: “What was your claim, in one sentence, without the card?” If they cannot say it, the talk was a string of facts. They generate the why. You do not narrate it. **Student attempt, then mixed practice.** Two or three two-minute talks across the week on unlike jobs: a literature claim, an informative how-it-works, a short argument from a source card. Mix with discussion (ten minutes, one question, page open) and one media sitting (thirty-second clip, oral evaluation, then the two-minute reply). A purchased speech book that is only “write a speech” can still be interleaved by you pulling yesterday’s poem onto today’s card. **Exit ticket.** Live at the table or recorded on a parent device. Same job: two minutes, card, eyes, one question. You are the audience. One specific note. The unaided talk is the index. A polished recording after twelve tool takes is performance. The first honest take, plus one retry for the note you gave, is the test. **How to fade help.** You talk from a card → they name the moves → they talk one minute → two minutes → they take a question → they talk to a second listener (sibling, co-op, grandparent on a call) → they add one prop that does a job. When they can do two minutes without a script, put the heavy coaching away. Bring it back when the task changes: a group discussion, a recorded podcast-style explanation, a talk that must include a visual. **When to stop talking.** After one model and one audience note. Struggle before rescue. Ask, wait, hint (“the because is not on the card — can you say it?”), then model a sentence if needed. You do not take over the talk. You do not turn feedback into a verdict on their soul. If two ages share the hour, the younger one can be a real audience for the two-minute talk and can retell in complete sentences; the older one can take the question. That is SL for both. It is not a speech elective unless you make it one. Recording is a gift at home. The student can hear the dropped endings. You can archive one talk per month as part of the English record. A recording is not a reason to chase a studio polish. One camera, one take, one retry. Group discussion at home needs a slightly more formal shape than dinner, or it melts into dinner. Sit. Name the question. Name the page. Each person talks once before anyone talks twice. You are the timer and the page-pointer, not the main speaker. Five to fifteen minutes is enough. A co-op, if you have one, is a real audience of peers. Use the room if you have it. The kitchen still works if you do not. Media sittings stay short. A full feature film is literature-or-leisure, not a media-evaluation hour. Thirty seconds to three minutes of a clip, paused, talked about, answered, is the dose that matches the exit ticket. You can finish the film later. The English job already happened at the pause. ## Practice that actually builds learning Understanding and practice grow together. Practice the two-minute talk this week, with a card and a why. A stack of written “speeches” never spoken is composition in disguise. **Blocked, for a new move.** The week a student first talks from a card, several two-minute talks on the same book are honest. Same for a first discussion with a claimed question, a first clip evaluation, a first visual that does a job. **Mixed, for when to use it.** Later the same week: a literature talk, an informative talk, a discussion, a clip. Choosing register — table, co-op, recording — is part of adapting speech. A page that is only “write your speech” will not teach it. **Brief retrieval.** Yesterday’s claim spoken in one sentence. The evidence word. The name of the speaker in Tuesday’s clip. Oral, at the start of the hour. Listening that is not retrieved becomes “it was a video about stuff.” **One incorrect example to diagnose.** Watch a talk (yours, an illustration, a public clip) that never states a claim, or that reads every word, or that uses a slide as the script. Ask what they thought the speaker was for. Then the student tries an isomorphic two minutes, unaided. **Compare after execution exists.** After they can talk from a card, put discussion next to presentation: “What changes when I can answer you in the middle?” After they can evaluate a speaker, put a print editorial next to a clip on the same claim. Flexibility is a third outcome. Comparing two clips fails if neither was actually heard. Kitchen and life can motivate. A phone call to schedule an appointment, a toast, a how-to for a sibling, a church or community announcement — real audiences. They do not replace the unaided English talk about a book. A busy week of errands will not teach a because-link at the podium. Read some verse and some drama aloud, even if the year is not “Speech.” The line is made for the ear. That sitting is literature and SL at once. Monthly: an unannounced two-minute talk on a finished book or a finished source, card only, one question. That is the retrieval the year needs. Keep the recording if you want a record. Grade the talk, not the camera work. Adapting speech is a taught move, not a personality upgrade. The same claim can be two minutes at the table, five minutes at a co-op, or a recorded explanation a grandparent will play later. Change one variable at a time: length, audience, or presence of a visual. A student who can only do the talk one way has executed. A student who can shrink or expand it on purpose has chosen. That is SL.4 and SL.6 on the Common Core map in ordinary language — present with evidence, adapt to the room — whether or not you use those codes.[^225] If drama is on the shelf, assign a page of a play to be said, not only read. The line is speech. Literature and SL occupy the same ten minutes, and neither strand is cheated. A human audience helps. You are the audience: one specific note. A sibling, a co-op, a grandparent on a call. A chatbot “debate partner,” if you try it at all, stays on a parent account, time-boxed, and after the student has talked to a person. It is not a friend. It is not the credit. If the student will sit SAT Reading and Writing or ACT English, those tests still sample print. They are not a speech exam. The talk remains a kitchen object. AP English Language’s free-response work is written; AP Literature too. CLEP College Composition’s essays are typed. None of that is a reason to skip SL. It is a reason not to confuse a score with a strand. ## For the student This strand is your voice and your ear. You will talk from notes, not from a script a tool wrote. You will look at a person. You will answer a question. You will listen long enough to say what someone else wanted, and whether they showed anything checkable. A presentation is a spoken essay. Claim first. Evidence. Because. A finish that is not “that’s it.” The card holds words, not paragraphs. If you have to read every sentence, practice until you do not. Discussion is not a quiz and not a quarrel. A good question is one a reasonable person could disagree with. You build on the last idea or you say you are turning. You point at the page. A clip is a text. What do they want? What do they show? What do they skip? Then your two minutes. **Tiny worked example.** Card (not a script): - Claim: We should not trust this narrator. - Evidence: door slam — “before supper” / “after dark” - Because: small fact wrong → large guide unsafe - Other view: still interesting - Finish: witness ≠ interesting Talk, about two minutes: you say the claim in a full sentence while looking at a person; you tell the two clocks; you say the because; you admit the narrator can still be fun to read; you stop. Someone asks, “Could the two clocks be a printer’s error?” You answer in a sentence, without diving into the card first: “Maybe, but the book never corrects it, and it sits next to other slips.” **Try 1.** Two-minute talk, card, one person in the room. Topic: a page you have actually read. Then one question. Record if you want to hear yourself. One retry for volume or the because — not twelve retries for polish. **Try 2.** Thirty-second clip (a parent picks it). Say: the claim, the evidence, the feeling, whether anything was checkable. Then your two-minute reply from a new card. **Explain it back.** In two sentences, tell someone the difference between notes and a script, and how you would know a talk had a claim. **Challenge.** Lead a ten-minute discussion with one question someone could disagree with, page open. You may not be the only person who talks. Your job is to ask once, wait, point back at the page when the talk floats away, and close with a sentence of what the group actually said. That is harder than a two-minute speech. It is still English. You are allowed to struggle. Looking at the floor is information, not a verdict. Try the last sentence again, eyes up. The tool does not take this test. You do. ## If it isn’t clicking Three diagnostics, and the next move for each. No shame. Placement is by skill, not by birthday. **Diagnostic 1 — there is nothing to say, or no sentence to say it in.** The student has not read the page, or cannot produce a spoken complete sentence, or cannot retell. *Next move.* Return to the book and to oral sentences: retell in order, full sentences, then a one-minute claim at the table to one person. Academic language is teachable in those retells.[^226] A calendar-grade 10 student who cannot do this is not “behind in Speech.” They are at the door. Listening to you read and talk still builds the ear. Independent presentation waits until there is a claim and a sentence. If anxiety is the hole and the sentence exists, shrink the audience: you only; then a recording with no audience; then one extra person. Time-box it. Personality is not the rubric. Two audible minutes are. **Diagnostic 2 — a script without a talk, or a talk without a claim.** The student can write a fluent speech and cannot look up. Or they can chatter without a because. Or the recording is a model’s sentences in a cloned voice. *Next move.* Card only. Two minutes. One question. No slides until the card works. If they are lost, watch your two-minute model again and name the moves, then they try a one-minute twin. Outsourced audio is not a speech; start over in their voice. TTS can help them *hear a page they must still decode or still say*; it cannot stand in for the credit.[^227] **Diagnostic 3 — choosing is the hole, or coverage is.** The student can talk about a novel and freezes on a clip, or can discuss at dinner and cannot present, or the year is a pile of unwatched “media literacy” worksheets. *Next move.* Mix on purpose: one talk, one discussion, one clip, in the same week. Shorten: thirty-second clip, two-minute reply. Keep evaluation inside the English work. An unannounced two-minute talk once a month. A longer Speech elective later if you want it — additional, not a patch for four silent English credits. **When to get a human tutor.** You cannot give a specific audience note even after the five-minute warm-up. Or the student has weeks of two-minute practice, and the unaided talk is still inaudible or empty. A tutor, a co-op speech class, a debate club, or Texas-style electives (Public Speaking, Debate) as *extra* credits are normal high-school plans.[^228] Stay the supervisor who can still hear a missing claim. Dual enrollment in a communication course, where a family can arrange it, is a college object; it does not erase the kitchen talk. **When to go ahead.** Two minutes from a card, eyes up, claim and because audible, one question answered. Discussion can last ten minutes without becoming a quarrel or a quiz. A clip can be evaluated in a few sentences. Then lengthen: five minutes; a visual that does a job; a recorded explanation; a second listener. Diagnose. Then move. ## Tools, including AI Optional helpers for you. The student talks. You are the audience. > **AI box — Speaking, listening, and media** > > Use a tool to explain a presentation structure to *you*, to generate practice prompts with your answers hidden, to write a hint script that withholds the spoken claim, or to diagnose a talk already given (you can paste *their* outline, not a request for a script). > > The student writes and says the talk. No slide dump the model wrote, no TTS as the speech credit, no unsupervised chatbot during the attempt, no companion “debate partner.” > > A debate-style tool, if used at all, stays on a parent account, time-boxed, after a human talk. The two-minute unaided talk is still the ticket. > > After the talk, you may ask a tool for audience questions *you* could pose. You still pose them. A recorder on a parent phone is a tool. Use it to let the student hear dropped endings. Use it as a record. Do not use it to chase a broadcast polish. Detectors have nothing useful to say about a talk. You do. ## What “done enough” looks like A “grade 10 Speech” book is a publisher’s scope, not a legal grade. Done enough is skill, plus at least one real talk. **Checklist before moving on.** - Gives a two-minute (or longer) talk from notes, not a script, looking at a person, with an audible claim and a because-link. - Answers one question without reading the card first. - Takes part in a discussion that uses the page and a claim someone could disagree with. - Evaluates a short clip or other non-print text: claim, evidence, feeling, checkable or not. - Adapts: table versus a slightly more formal audience; live versus recorded. - Uses a visual only if it does a job the voice cannot do. - Receives one specific audience note and retries that one thing. - Has at least one live or recorded presentation in the English folder a stranger could play. What can still be thin: a full debate season; studio editing; every media form on a state list; a separate Speech elective. Those can travel. Four silent literature credits cannot count as SL done. **Transcript.** Write **Speech** if you ran a real additional credit. Or write **English 9** (or 10, 11, 12) with “presentations and discussion; media evaluation” in the description. Texas electives do not replace English I–IV.[^229] Four English credits remain typical; NCAA wants English years, and a Speech line is readable when the work was done.[^230] Never write a brand in the title column. Never write a clever thinking-course title in place of the talk. A parent who felt rusty should be able to say, after this chapter: I know how to be an audience. I know a script from notes. I know what to say tomorrow — one specific note, then they try again. The student still stands up. We can do this. # Chapter 9 — Records The English on the table can be as serious as any school’s. The *credit* is a family claim until a stranger can use it — a college admissions reader, an NCAA staffer, a public-school counselor placing a transfer. That stranger will not have sat at your kitchen table. They will have a page. This chapter is how you make that page. It is not a law office. It is not a fifty-state encyclopedia. It is the export of the work you already taught: titles a stranger can file, a description that names the books and the writing, a signature that means you checked. This week you can take last year’s English folder and write one course line a counselor would recognize. Today the student can list the books they actually finished and the papers they actually wrote. ## What you are making You are making a record of English, not a national diploma. There is no federal homeschool diploma and no national English credit a registrar already files. The Texas Education Agency says, in so many words, that the State of Texas does not award a diploma to students who are home schooled.[^231] New York’s education department says a high school diploma may only be awarded to a student enrolled in a registered secondary school.[^232] The NCAA Eligibility Center treats proof of graduation as a document the family or an umbrella school issues.[^233] A parent-issued diploma can be a real piece of paper. It is not a registrar’s national English credit. Three other objects get mixed with the transcript and should not be. Dual-enrollment English lives on a *college* transcript. AP and CLEP are exam scores. A state-compliance packet — a Pennsylvania evaluator letter, a Virginia evidence-of-progress file, a North Carolina test score kept on hand, a New York quarterly — is what your statute asked for. A college that wants a transcript, a counselor letter, and scores is asking for a different envelope. Keep the envelopes honest. Each object does its own job. ## Titles a stranger can read Use titles a counselor already has a slot for: **Language Arts, English 9, English 10, American Literature, British Literature, World Literature, Composition, Speech.** Those are the names this book uses. They travel. NCAA’s own sample homeschool transcript uses English 9, English 10, English 11, English 12, and Composition I and II. The Core-Course Worksheet says the course name on the worksheet should be the same as the title on the transcript.[^234] That is good practice even if no one in the house will ever play college sports. The title is the genre. The program you bought is the method. All About Reading, Logic of English, IEW, Well-Trained Mind, Brave Writer, Teaching the Classics, AmblesideOnline, a literature anthology — those may appear in a description or a parent log. They are not the course name. “English 9 — All About Reading” will not be honored as English 9. “Critical Thinking 101” is not an English credit. “AI English” is not a course. Critical thinking lives inside the paper. A tool is not a title. You have two honest shapes, and both map. One is four years of English with literature and composition inside the descriptions: English 9, English 10, English 11, English 12. Texas public-school law names those years English I through IV; Virginia’s Board names English 9 through 12. Local catalogs sometimes subtitle a junior year “Survey of American Literature.” That subtitle is catalog language, not a national sequence.[^235] The other shape splits the work the way this book splits it: Composition; American Literature; British Literature or World Literature; Speech. Either is mappable. Language Arts is the usual elementary and middle-school title. Speech is a real line. A family that never presents anything has not covered speaking and listening, even with four literature credits. ## What belongs on the English line A usable record is made of objects families already produce. For each high-school English course, write: - the **title** from the list above, or a close analogue a stranger can file (English 11, American Literature); - the **year**, and the grade equivalent you actually used; - the **credit** — commonly 1.0 for a year-long course; - the **grade** you assigned, and the scale somewhere in the packet (A = 90–100, or whatever scale you truly used); - a **brief description** of readings and writing — not a brand; - **who taught and who graded**; - your **signature** and a date; - **external artifacts** when they exist: SAT Reading and Writing; ACT English and Reading; an AP score; a CLEP score; a college transcript from dual enrollment; a state-required test or evaluator letter. The description is the part a stranger actually reads. Name the texts. Name the writing load. Name the grading basis. A sentence that says only “IEW” is a brand, not a course. A sentence that says “nine structural models from source texts including *X*, *Y*, and *Z*; four essays; one short research paper; grade from unaided drafts and a conventions checklist” is a course. Here is an illustration, not a reported family: > **English 9** · 2025–26 · 1.0 credit · Grade: A (A = 90–100, B = 80–89) > > Readings: one Shakespeare play; *To Kill a Mockingbird*; a short-story set; weekly informational pieces on U.S. history topics taught the same year. Writing: twelve essays (narrative, argument, informative), one short research paper citing at least three sources, weekly sentence combining. Speaking: one prepared talk with notes. Grading: 50 percent unaided writing, 30 percent reading discussion and annotations, 20 percent language conventions in the draft. Instructor: \[parent name\]. Signature: \_\_\_\_\_\_\_\_ Date: \_\_\_\_\_\_\_\_ Change the books to the books on *your* table. Change the count of papers to the papers in the folder. The shape stays. If you keep a writing folder — unaided drafts, a research paper, a speech outline — you already have the evidence the description claims. The folder is not the transcript. The transcript is the map to the folder. ## Four years is typical Nearly every state in the federal table of high-school graduation credits asks for four English credits. California’s state minimum in that 2018 table is three; a few states leave the number to local decision. Texas, Florida, and Virginia each list 4.00.[^236] Four years is therefore the usual U.S. high-school shape, not a federally named American–British–World sequence and not a Carnegie-unit law hanging over the kitchen. Plan four years of high-school English unless your own statute says otherwise. NCAA’s “year” of English is an athletic-eligibility object, not a federal credit statute. Elementary and middle years still deserve a title: Language Arts. The work is real. The high-school credit clock has not started. ## The law is a pattern, not one form U.S. homeschooling is state law plus compulsory-attendance exceptions. There is no federal English office. Typical moving parts look like this: whether a state agency regulates homeschooling at all; **notice** to a division or a superintendent; whether someone reviews a plan; an annual **test**, **portfolio**, or **evaluator** in some states; and **who issues the diploma** — commonly the parent or an umbrella, not the U.S. Department of Education. That is a pattern of *kinds of objects*. Dates, ages, subject lists, and exemptions change. Read your current department page and the statute it cites. A color-coded chart on a membership site is not your statute.[^237] Four official-page examples show the range. They are examples. They are not your homework unless you live there. **Texas.** The Texas Education Agency does not regulate, index, monitor, approve, register, or accredit home-school programs.[^238] Home schooling has been a legal alternative since 1994. Parents follow a bona fide written curriculum that includes reading, spelling, grammar, mathematics, and a course in good citizenship. Texas does not award a homeschool diploma. A public-school transfer is treated like a transfer from an unaccredited private school; the district may assess for placement or credit. **North Carolina.** Notice of Intent to the Division of Non-Public Education. Operate on a regular schedule at least nine calendar months. Each year, a nationally standardized achievement test that includes English grammar, reading, spelling, and mathematics; keep the scores and show them if asked.[^239] The division does not prescribe your English books. The home school, not the State, issues the diploma. A five-hour school day is a *recommendation* on that page, not the statute. **Pennsylvania.** Before you start, and each year by 1 August, a notarized affidavit or unsworn declaration to the district superintendent, with an outline of objectives by subject. District *approval is not required* once the paperwork is filed. Each year, an evaluator’s report by 30 June. In grades 3, 5, and 8, the portfolio includes a reading/language-arts test score — statewide tests or a nationally normed achievement test.[^240] This is a portfolio-and-evaluator state with a test object in named grades. **Virginia.** Notice of intent to the division by 15 August, a list of subjects, and evidence that the parent meets one of four qualification options. By 1 August after the year of instruction, evidence of progress: a composite score in or above the fourth stanine on a nationally normed test, **or an equivalent score on the ACT, SAT, or PSAT**, or an evaluation or college transcript the superintendent accepts.[^241] Virginia will take those college-admissions tests as *one* evidence-of-progress option. That is Virginia. It is not a national rule that “homeschoolers take the SAT instead of a diploma.” **New York**, as a heavier-process illustration: a letter of intent, then an Individualized Home Instruction Plan the superintendent reviews, quarterly reports, and an annual assessment. Home instruction does not yield a local or Regents diploma.[^242] Instruction for a six-year-old, New York says, should be geared to the student’s needs and previous achievement — placement by skill, not by the birthday that triggered compulsory attendance. If you live in none of those states, you still have a department page. Open it. Your filing date lives there. ## If they transfer Transfer into a public school is a local placement problem. Texas districts commonly assess mastery and treat the student like a transfer from an unaccredited private school. New York leaves grade placement to the principal, as with any transfer, and may award credit from assessment or evidence of completed work. North Carolina’s division *recommends* asking the local school, before you start, how they would place the child on return.[^243] None of those pages promises that a brand in the title column will be read as English 9. Another reason the title should already look like English 9. ## College-ready objects — not the course Scores help a stranger believe the transcript. They do not replace the writing folder. A high SAT Reading and Writing score with empty drafts is a score, not a composition course. **SAT Reading and Writing** is the current section name on the digital SAT — not the older “Evidence-Based Reading and Writing” label. Fifty-four questions in sixty-four minutes, two modules of short passages. College Board’s college-and-career-readiness benchmark for Reading and Writing is **480**: a 75 percent chance of at least a C in first-semester credit-bearing courses in history, literature, social sciences, or writing.[^244] **ACT English** and **ACT Reading** together cover a similar waterfront in a different format. On the enhanced ACT, English is 50 questions in 35 minutes; Reading is 36 questions in 40 minutes. ACT’s English college-readiness benchmark is **18**: a 50 percent chance of a B or higher — about a 75 to 80 percent chance of a C or higher — in English Composition I.[^245] Those two benchmark numbers are not equated to each other. Neither is the meaning of English. Put both ACT section scores in the packet if the student sat the test. **AP English Language and Composition** is not **AP English Literature and Composition**. Language is introductory college composition: rhetorical analysis of nonfiction. Literature is introductory college literary analysis: fiction, poetry, drama. Neither has a prerequisite course. Credit is set by each college.[^246] If the student sits an exam, put the score on its own line. Until you have confirmed how a home course may carry the AP title, the safer transcript line is English 11, or Composition, or American Literature, with the exam score beside it. One exam does not stand in for the other. **CLEP College Composition** is a first-year-composition exam, not a high-school course. Fifty multiple-choice questions in 55 minutes and two typed essays in 70 minutes. ACE recommends **six semester hours at a score of 50**. Each college sets its own policy.[^247] A CLEP score may become college credit. It is **not** an NCAA-approved core course. Analyzing and Interpreting Literature is a different CLEP exam — passage-based, no required reading list — and the same NCAA limit applies. **Dual or concurrent enrollment** is a college course taken during high school, not AP and not CLEP. In 2017–18, 82 percent of U.S. public schools with any of grades 9–12 *offered* it. That is schools offering, not students taking, not English-specific, and not a homeschool rate.[^248] Access at home is local: a North Carolina dual-enrollment page is written in public-school eligibility language; New York generally keeps home-instructed students out of the district’s instructional program; Virginia will accept a college transcript as one form of evidence of progress. Map the course to Composition or English 12, with the college named in the description. The college’s own transcript is the credit-bearing object. None of these overlays is required for a serious English sequence. They sit on top. The kitchen method is still the sentence, the paragraph, the book, the unaided draft. ## NCAA, if athletics are in view If Division I or Division II sports might be in the picture, English planning starts in ninth grade. Division I asks for **four years** of English (or native language) inside sixteen core courses, a 2.3 core GPA, and the 10/7 rule: ten of the sixteen core courses, including seven in English, math, or science, before the seventh semester. Division II asks for **three years** of English inside sixteen core courses and a 2.2 core GPA.[^249] A core English course is college-preparatory, taught at or above regular academic level, and named the same on the worksheet and the transcript. The homeschool packet the Eligibility Center describes includes a homeschool transcript, an administrator statement that instruction followed *state* law, a Core-Course Worksheet for each core course signed by the parent, and proof of graduation. Evaluation begins after those documents are in **and** an NCAA school has placed the student on an Institutional Request List.[^250] CLEP, credit-by-exam, and audited courses do not count as core. A year of English the child did not do cannot be wished onto the worksheet. If athletics are not in view, the same titles and descriptions still serve colleges and transfers. The honesty is the point either way. ## Place by skill, not birthday A publisher’s “Level 4” is a scope, not a legal grade. New York’s home-instruction page already says instruction should match the student’s needs and previous achievement.[^251] Several programs you may already own say some version of the same thing in their own words: sit a placement test; start the writing sequence at the first level that matches the skill; “don’t be governed by age range.” For English, two questions do most of the work. Can the child decode this text without guessing from pictures? If not, they are still in a code program, whatever their age. Can they produce a paragraph that holds a claim without an adult holding the pencil? If not, they are still in taught writing — partnership, imitation, or narration — whatever year the transcript shows. High-school *titles* can still be English 9 while the description tells the truth about supports. A title that matches a fourteenth birthday and not the work helps no college and hurts an NCAA file. ## A template, a signature, and the truth A tool, including an AI helper on your side of the table, may draft a **template**: blank lines for title, year, credit, grade, texts, writing load, grading basis, signature. That is a form. It is not a record until you fill it from the folder. You read every line. You check the books against the shelf. You check the essay count against the writing folder. You check the hours, if your state asks for hours, against the log you actually kept. Then you sign. Hours that were not taught are a false statement, not a kindness to the file. A polished paragraph the child did not write is not the child’s composition. The signature means you stand behind the page. The tool may format. You attest. ## For the student Your transcript is a picture of work you did. You read the books. You wrote the papers. You stood up and spoke. When a parent asks you to list what you finished this year, that list is not busywork. It is how a stranger will understand your English. Name the books. Name the papers. If a line on the page is not true, say so before anyone signs. You do not have to enjoy filing. You do have to own the work the file describes. ## What “done enough” looks like You can hand a stranger one page per high-school English course. The title is from the list a counselor already knows. The description names texts, writing load, and grading basis. Your name is on it, and you would still sign it tomorrow morning. State-compliance objects, if your state asks for them, sit in their own envelope. SAT, ACT, AP, and CLEP scores sit on their own lines — Language and Literature separate, CLEP not labeled as NCAA core. If athletics might be in view, ninth-grade English already has a mappable title and a worksheet behind it. The folder matches the page. That is enough. # Chapter 10 — Resources This chapter names programs families actually use. It does not rank them. It does not sell them. It is a fit list so you can match a stack to the table you already have: how much script you want to read, how the English code is taught, how writing is taught, how a book is taught, and how the program places a student. A “grade 6” on a spine is a publisher’s scope, not a legal grade. Placement is by skill. The transcript, as Chapter 9 said, still says Language Arts or English 9. Combining two honest programs does not authorize a franken-title. This week you can name *why* the books on your shelf fit this child. Today the student can read one page from a book that is actually in the house. ## How to read a program Five questions do the work. **Parent load.** Is the lesson scripted, so you read and go? Or are you the teacher of a method you first have to learn — a writing seminar, a Socratic list, a year’s living-book schedule? **Code path.** Does this program teach the English writing system as a code — phonograms, blending, encoding — or does it assume decoding is already done? Some literature-rich homes historically mixed sight recognition with phonics. That is a fit fact, not a slur. A child who cannot decode is not “behind in literature.” They are still in a code program. **Writing method.** Imitation from a source text? Narration, copywork, and dictation that grow into exposition? A parent as writing partner while voice outruns handwriting? Textbook on-demand essays? Writing is taught, not assigned. The method is how the teaching happens. **Literature method.** Assigned living books and narration? A Socratic list that works on any story? A school anthology with apparatus? Decodable stories inside a reading program? Those are different jobs. **Placement.** A placement test? Start at Level 1 of the writing sequence even if the birthday says otherwise? “Don’t be governed by age range”? The grade printed on the cover? Skill first. Where a company makes a strong claim — twenty minutes a day, ninety-eight percent of English words, Orton-Gillingham, science of reading — that claim is the company’s. This chapter reports what the page said. It does not certify the slogan. ## All About Reading and All About Spelling All About Learning Press sells two related sequences. All About Reading is four levels plus a Pre-reading program: phonics, decoding, vocabulary, fluency, and comprehension, with decodable stories. The company says that after Level 4 the child has the skills to decode just about any word. All About Spelling is seven levels of encoding and spelling rules; the company says that after Level 7 the child will be spelling at the high-school level. Shared letter tiles. The company describes the lessons as scripted, open-and-go, and based on an Orton-Gillingham approach, and it advertises “20 minutes a day.” Placement tests are offered.[^252] **Fit.** High structure, low improvisation: you read the script. Explicit phonics, not a guess-from-pictures path. Writing is not All About Reading’s job; spelling-toward-writing is All About Spelling. Literature inside the program is decodable stories, not a novel seminar. Place with the company’s test, not the birthday. Twenty minutes is a marketing claim, not a research hour. **Transcript.** Language Arts, with the program named in the description. Not “All About Reading Level 3.” ## Logic of English Logic of English sells Foundations (ages 4–7, some eight-year-olds, levels A–D) as a complete early language-arts course: phonics, reading, spelling, handwriting, vocabulary, composition, and grammar, built on 75 basic phonograms. The company says students decode without relying on pictures, guessing strategies, or memorizing sight words. Essentials is for ages 8 to adult — a course in how English spelling works, with those phonograms plus spelling rules and advanced phonograms. Foundations is not a prerequisite for Essentials; age and need decide. The related book *Uncovering the Logic of English* is where the “98 percent of English words” claim lives. That figure is the company’s.[^253] **Fit.** Scripted, denser than a light workbook. You will learn the phonogram-and-rule system too. Code instruction is the point. Writing and grammar show up especially in Essentials. Literature is not the spine. Older beginners can start Essentials; that is closer to skill than to birthday, still marketed in ages. **Transcript.** Language Arts. Essentials is not English 9. ## Institute for Excellence in Writing (Structure and Style) IEW is a writing method, not a phonics program and not a Socratic novel course. Students analyze and imitate existing writing, then move toward original composition. Nine structural models (notes and outlines through formal essay and critique) and a style overlay of dress-ups, sentence openers, and decorations, added at the student’s pace. Units repeat year after year with new source texts. Teacher training is the video *Teaching Writing: Structure and Style*. Student videos exist in grade bands 3–5, 6–8, and 9–12. Fix It! Grammar is a companion, not the same product.[^254] **Fit.** Parent load is high at first — watch the training — then can drop if student videos carry the modeling. Writing by imitation, outline, and checklist: the opposite of “just journal.” Literature appears as source texts, not as a discussion seminar. Phonics is someone else’s job. Placement follows the company’s grade-band packages and checklist mastery, not a decoding diagnostic. **Transcript.** Composition, or English 9–12 with composition described. Never “IEW” in the title column. ## Well-Trained Mind: Writing With Ease to Writing With Skill Susan Wise Bauer and Jessie Wise’s *The Well-Trained Mind* (Essential Edition, 2024) applies a trivium shape: grammar stage roughly grades 1–4, logic grades 5–8, rhetoric grades 9–12, with Great Books and original persuasive composition in the last band. They note that other classical programs place the stages on different grade bands. On the support pages, grammar is *First Language Lessons* (levels 1–4) then *Grammar for the Well-Trained Mind*. Writing is *Writing With Ease* — narration, dictation, copywork — with diagnostic tests to choose a level, then *Writing With Skill* levels 1–3 for grades 5 and up. Start Writing With Skill at Level 1. *The Creative Writer* is optional and, the support page says, not a substitute for expository writing.[^255] **Fit.** Parent load is high: you are the classical teacher, with scripted elementary writing if you buy the workbooks. The sequence assumes a “good foundation of phonics”; it is not itself a beginning-reading program of the All About Reading kind. Writing method: narration, then dictation, then copywork, then outlined exposition. Literature method in the rhetoric years: Great Books paired with history, not a single novel-seminar product. Placement: diagnostics, and the older student who skipped the ladder still starts Writing With Skill at Level 1. **Transcript.** Language Arts in the grammar years; English 9–12, American / British / World Literature, or Composition later. Not “Writing With Skill 2.” ## Brave Writer Julie Bogart’s Brave Writer describes stages of growth in writing. Partnership Writing is the stage the company says families skip: a writing-revising-editing partnership when verbal fluency outruns written fluency, often around nine and ten, “but don’t be governed by age range. Focus instead on the description and match it to your child.” Literature is paired through Dart (and related) novel collections. The parent is the coach in the room.[^256] **Fit.** High presence, lower script density than a Structure-and-Style training video or an All About Reading lesson. Not a decoding program. Writing method: partnership and voice, then ownership — a different theory of help than imitation-from-source or the dictation ladder, though all three put an adult next to the page. Literature method: whole books with a guide. Placement by stage description, not birthday. **Transcript.** Language Arts or Composition. Not “Dart.” Not “Partnership Writing.” ## Teaching the Classics Adam and Missy Andrews’s Teaching the Classics, from Center for Lit, is an eight-hour seminar the company calls “not a curriculum” — a method for any story. Stories share context, structure, and style; identify those to reach theme. The Socratic List is 173 discussion questions, usable as essay prompts. Second Edition 2017; the company dates the method to 2004. Annotated reading lists for all ages. Designed K–12.[^257] **Fit.** One-time training, then reusable questions. Low materials cost after the seminar; high conversational skill. No phonics. Discussion that can become essays, not a full composition sequence. This is the literature-method product: Socratic, any book, no set canon. Questions scale by difficulty, not by a publisher grade on the novel. **Transcript.** The books taught become American Literature, British Literature, World Literature, or English 9–12. Teaching the Classics is the method, not the course name. ## AmblesideOnline AmblesideOnline is a free Charlotte Mason community curriculum, Years 0–12, with weekly schedules and a complete booklist. No cost to use the plans. Literature-rich, living-book spine. Many titles are in the public domain or available in free editions; the site does not claim that every book is. Language arts as the Advisory describes it: in the early years, a reading program the family chooses — “use whatever you like” — plus oral narration as oral composition, and copywork. Written narration begins around age 10–11. Grammar in Year 4. Dictation for spelling and punctuation. In the high-school years, grammar in the student’s writing; narrations become more focused, various formal types; Strunk and White and Zinsser named as books to read and apply. “No formal composition curriculum is recommended here.” A family beginning Mason late may spend as much as two years on oral narration before written narration.[^258] Charlotte Mason’s own early-reading description, on AmblesideOnline’s introduction page, includes sight vocabulary plus phonics. That is a whole-language residue in the historical method. A family using these booklists can still teach the code with an explicit-phonics program. AmblesideOnline already says to use whatever reading instruction meets the child. Short lessons of ten to twenty minutes per subject for younger children are community and historical practice, not a statute. **Fit.** High book-management, lower worksheet load. Narration after every reading is the labor. Code: the family’s choice; this is not a phonics curriculum. Writing method: oral narration, then written narration, then formal types. Literature method: assigned living books, one reading then narrate. Years look like grades; catch-up and compressed paths exist, which is a tacit admission that the Year number is a sequence, not a birthday. **Transcript.** Language Arts; English 9–12; American / British / World Literature matching the year’s books. Not “AO Year 10.” ## A textbook-anthology path Some families want a school-shaped literature book: a used Prentice Hall or Holt McDougal volume, whatever the local district used, or a current basal. One currently sold grades 6–12 example is Houghton Mifflin Harcourt’s *Into Literature*. The company’s pages describe authentic literature, reading, writing, grammar, and vocabulary aligned 6–12, six units per grade, and, at grades 11 and 12, a chronological split — American literature in 11, British literature in 12. Marketing is school-district marketing. Homeschool use is possible the way any basal is possible.[^259] **Fit.** Parent load is high if you run it as designed (teacher edition, unit tasks, assessments), lower if you use it as a reading anthology and skip the apparatus. Not a beginning-reading program. Writing method: textbook prompts and on-demand essays, not Structure-and-Style outlines or Mason narration. Literature method: curated anthology with apparatus, grade-coded. Placement is the grade on the spine — the definition of “a grade 10 reader is a publisher’s scope.” **Transcript.** English 9, English 10, American Literature, British Literature. Still not “Into Literature 11.” Older anthologies occupy the same fit slot. Name the volume in the house. ## Combining honestly A family may combine. Honest combinations follow the products’ own jobs: All About Reading or Logic of English for the code; IEW, Writing With Ease then Writing With Skill, or Brave Writer for writing; Teaching the Classics, AmblesideOnline, or an anthology for literature. The child who cannot yet decode does the code program first, or alongside read-alouds above decoding level. They are not “behind in Year 2.” The sixth-grader who still guesses from pictures is still in a code program, not a literature emergency. Combining does not mint a new course title. The transcript still says English 9. The description may say “phonics sequence X; composition method Y; novels Z.” A stranger can file that. “AAR/IEW/AO hybrid” is a kitchen nickname. ## Where to get the books You do not need a new boxed set for every year of literature. Public-domain novels, essays, poems, plays, and speeches — works whose copyright has expired — are widely available at no cost. Project Gutenberg and the Internet Archive collect plain-text and scanned editions. Standard Ebooks prepares cleaner public-domain texts. The Library of Congress and many university libraries host scans. AmblesideOnline’s booklists often point to free editions of the year’s living books.[^260] Your public library still matters: interlibrary loan, a used-book sale, an anthology already on the shelf. A well-read used copy of a school literature book is the same fit slot as a new basal. Check that the edition in front of you is the work you meant to assign, not an abridgment you did not choose. Public-domain is a copyright status, not a quality stamp. Teach the book, not the brand of the free file. ## When to get a reading specialist This book does not diagnose anyone. It does not promise an IEP at a kitchen table. No named curriculum “cures” dyslexia. The International Dyslexia Association’s (IDA) 2002 definition is the plain picture to hold. Dyslexia is a specific learning disability, neurobiological in origin, marked by difficulties with accurate or fluent word recognition and with poor spelling and decoding. Those difficulties typically come from a phonological-language deficit. They are **unexpected** given other cognitive abilities and given **effective instruction**.[^261] Secondary problems may include comprehension and a thinner reading diet that then slows vocabulary and knowledge. Gough and Tunmer locate the proximal problem, in the simple view of reading, as an inability to decode.[^262] The IES kindergarten-through-grade-3 practice guide notes that students who struggle persistently with phonological awareness often benefit from one-on-one or small-group intervention, and that early intervention can address a phonological core deficit that otherwise may lead to the word-reading difficulties that mark reading disability.[^263] Here is the move at home. Teach the code explicitly. Watch whether accurate decoding is growing. If difficulty is unexpected — the child understands by ear, other abilities look typical — and persistent after that systematic teaching, get a human specialist. A reading specialist, an educational psychologist, or a speech-language pathologist who actually assesses word reading is the next adult in the room. The first move is still teaching. The specialist is for the pattern that does not yield. Text-to-speech can help a documented print disability access a science chapter. That is compensation, not a phonics program.[^264] A chatbot that reads the page so the child skips decoding is not a specialist and not a program. ## What “done enough” looks like You can say, in one sentence, why this year’s English stack fits *this* child: parent load you can actually carry, a code path if decoding is still the job, a writing method that teaches rather than assigns, a literature method that matches the books on the table, and placement by skill. The programs have names. The transcript has titles a stranger can read. The two lists are not the same list. # A Note on Sources Studies named in the chapters are listed in Notes, in one series at the back. That is where the full citations live, so the teaching pages can stay a teaching voice. Some items the research behind this book did not open stay out of the teaching voice. I did not invent a coefficient, a statute, or a product feature to fill a hole. If a study is in the chapter, it is in the notes. If we could not open it, it is not used as a finding here. Access date 31 August 2026. [^1]: . Ellena Sempeles and Jiashan Cui, *Parent and Family Involvement in Education: 2023*, NCES 2024-113 (Washington, DC: National Center for Education Statistics, September 2024), Table A-6: 3.4 percent homeschooled, approximately 1,765,000 students, ages 5–17 with a K–12 grade equivalent, 2022–23. https://nces.ed.gov/pubs2024/2024113.pdf. Access date for URLs in these notes: 31 August 2026. Latest federal count this book uses; not a 2026 national total. A neighboring 5.2 percent received instruction at home (homeschooled or full-time virtual) and is a different bucket. [^2]: . Texas Education Agency, “Home Schooling,” https://tea.texas.gov/families-and-students/finding-school-your-child/home-schooling. New York State Education Department, “Home Instruction Questions and Answers,” https://www.nysed.gov/nonpublic-schools/home-instruction-questions-and-answers. Homeschooling is legal in all fifty states and D.C.; these two pages illustrate a light-process and a heavier-process pattern, not a national template. This book is not legal advice. [^3]: . Texas Education Agency, “Home Schooling”: “The State of Texas does not award a diploma to students that are home schooled.” New York State Education Department, “Home Instruction Questions and Answers”: a high school diploma “may only be awarded to a student enrolled in a registered secondary school.” NCAA Eligibility Center, *Home School Toolkit 2025–26*, http://fs.ncaa.org/Docs/eligibility_center/Student_Resources/Home_School_Toolkit.pdf, treats proof of graduation as a family- or umbrella-issued artifact. [^4]: . National Center for Education Statistics, “Table 5.7. State course credit requirements for high school graduation, by state: 2018,” https://nces.ed.gov/programs/statereform/tab5_7.asp. Nearly every listed state requires 4.00 English credits (California 3.00; a few locally determined). Four credits are typical, not a federally named American–British–World sequence. Transcript titles in this book: Language Arts, English 9, English 10, American Literature, British Literature, World Literature, Composition, Speech. [^5]: . Philip B. Gough and William E. Tunmer, “Decoding, Reading, and Reading Disability,” *Remedial and Special Education* 7, no. 1 (1986): 6–10: reading equals the product of decoding and linguistic comprehension, R = D × C, each variable ranging from 0 to 1. [^6]: . National Center for Education Statistics, *NAEP Report Card: 2024 Reading*, grades 4, 8, and 12. NCES states that “the NAEP Proficient achievement level does not represent grade-level proficiency as determined by other assessment standards.” NCES does not publish a nationally representative homeschool NAEP reading score. [^7]: . Common Core State Standards Initiative, *Common Core State Standards for English Language Arts & Literacy in History/Social Studies, Science, and Technical Subjects* (NGA Center / CCSSO; ADA-compliant posting of the 2 June 2010 standards), https://www.corestandards.org/wp-content/uploads/2023/09/ADA-Compliant-ELA-Standards.pdf: “The Standards define what all students are expected to know and be able to do, not how teachers should teach.” [^8]: . Barbara Foorman et al., *Foundational Skills to Support Reading for Understanding in Kindergarten Through 3rd Grade*, NCEE 2016-4008 (Washington, DC: Institute of Education Sciences, 2016), Recommendation 4: the panel “discourages teachers from allowing students to use guessing strategies to identify unfamiliar words” and from guessing unknown words using beginning letters or pictures. https://ies.ed.gov/ncee/wwc/Docs/PracticeGuide/wwc_foundationalreading_070516.pdf. Access date for URLs in these notes: 31 August 2026. Linnea C. Ehri, “Orthographic Mapping in the Acquisition of Sight Word Reading, Spelling Memory, and Vocabulary Learning,” *Scientific Studies of Reading* 18, no. 1 (2014): 5–21: sight-word reading is reading from memory after letter-sound connections have been formed, not guessing from pictures. [^9]: . New York State Education Department, “Home Instruction Questions and Answers,” https://www.nysed.gov/nonpublic-schools/home-instruction-questions-and-answers: instruction for a six-year-old need not be “first grade”; “instruction should be geared to the level appropriate to the student’s needs and previous level of achievement.” A placement test that ships with a program you already own is a tool for the same job, not a legal grade. [^10]: . Philip B. Gough and William E. Tunmer, “Decoding, Reading, and Reading Disability,” *Remedial and Special Education* 7, no. 1 (1986): 6–10: R = D × C. National Reading Panel, *Teaching Children to Read*, NIH Pub. No. 00-4769 (Washington, DC: NICHD, 2000): systematic phonics is a means, not an end; kindergarten children are not “not ready”; guided oral reading with feedback supports fluency. Access date for URLs in these notes: 31 August 2026. Daniel T. Willingham, “How Knowledge Helps,” *American Educator*, Spring 2006; Daniel T. Willingham and Gail Lovette, “Can Reading Comprehension Be Taught?,” *Teachers College Record*, 2014: strategy instruction has a place and a ceiling; background knowledge does not have the same ceiling. [^11]: . Steve Graham and Dolores Perin, “A Meta-Analysis of Writing Instruction for Adolescent Students,” *Journal of Educational Psychology* 99, no. 3 (2007): 445–76: strategy instruction 0.82, sentence combining 0.50, grammar instruction −0.32 on writing quality. Steve Graham et al., *Teaching Elementary School Students to Be Effective Writers*, IES/WWC practice guide, revised 2018, Recommendations 2–3: teach a writing process for a variety of purposes; teach handwriting, spelling, and sentence construction so working memory is free for composition. [^12]: . Architecture of this book, matching NCAA Eligibility Center, *Home School Toolkit 2025–26*, sample transcript titles (English 9–12, Composition) and the Core-Course Worksheet rule that the course name match the transcript title. http://fs.ncaa.org/Docs/eligibility_center/Student_Resources/Home_School_Toolkit.pdf. Texas Education Agency and New York State Education Department: no state-issued homeschool diploma. [^13]: . National Reading Panel, *Report of the National Reading Panel. Teaching Children to Read*, NIH Pub. No. 00-4769 (Washington, DC: National Institute of Child Health and Human Development, 2000), https://files.eric.ed.gov/fulltext/ED444126.pdf: guided repeated oral reading with feedback had a significant positive impact on word recognition, fluency, and comprehension; the Panel did not find sufficient causal evidence that programs of independent silent reading with minimal guidance, by themselves, reliably improve reading. Timothy Rasinski, “Creating Fluent Readers,” *Educational Leadership* 61, no. 6 (2004): emphasizing speed at the expense of prosody is “a corruption of the concept of reading fluency.” Access date for URLs in these notes: 31 August 2026. This book does not print a single NRP fluency effect size; the summary report does not give one. [^14]: . Barbara Foorman et al., *Foundational Skills to Support Reading for Understanding in Kindergarten Through 3rd Grade*, NCEE 2016-4008 (Washington, DC: Institute of Education Sciences, 2016), Recommendation 4: each student reads connected text every day; the panel discourages guessing unknown words from beginning letters or pictures. https://ies.ed.gov/ncee/wwc/Docs/PracticeGuide/wwc_foundationalreading_070516.pdf. [^15]: . Session shape in this chapter is pedagogical design for the kitchen table, not a national English-minutes study (none was found) and not an English-specific interleaving trial. National Reading Panel (2000) and Foorman et al. (2016) want the new correspondence used in varied connected text, not a page of identical items. [^16]: . Steve Graham and Dolores Perin, “A Meta-Analysis of Writing Instruction for Adolescent Students,” *Journal of Educational Psychology* 99, no. 3 (2007): 445–76: extra writing 0.30; process writing approach 0.32. Strategy instruction and sentence combining moved quality more. Steve Graham et al., *Teaching Elementary School Students to Be Effective Writers*, IES/WWC practice guide, revised 2018, Recommendation 2: writing time includes instruction for a variety of purposes, not only a journal. [^17]: . Nicholas C. Soderstrom and Robert A. Bjork, “Learning Versus Performance: An Integrative Review,” *Perspectives on Psychological Science* 10, no. 2 (2015): 176–99: current performance is “often an unreliable index” of learning. The unaided exit ticket is the check this hour trusts. An AI-polished paragraph is performance. The unaided sentence the next morning is the test. [^18]: . Hamsa Bastani, Osbert Bastani, Alp Sungu, Haosen Ge, Özge Kabakcı, and Rei Mariman, “Generative AI without guardrails can harm learning: Evidence from high school mathematics,” *Proceedings of the National Academy of Sciences* 122, no. 26 (2025): e2422633122. Field experiment, nearly 1,000 Turkish high-school math students, GPT-4-0613: a ChatGPT-like tool raised assisted practice grades 48 percent relative to control, then cut unaided exam grades 17 percent; a hint tutor that withheld full solutions raised practice 127 percent and was statistically indistinguishable from control on the unaided exam (point estimate −0.004). GPT Base was correct on 51 percent of the practice set. The paper notes that math has objective scoring “which is unavailable in creative subjects such as writing.” This is a math RCT, not a writing RCT; the transferable caution is the crutch. Author PDF: https://hamsabastani.github.io/education_llm.pdf. [^19]: . Weixin Liang, Mert Yuksekgonul, Yining Mao, Eric Wu, and James Zou, “GPT detectors are biased against non-native English writers,” *Patterns* 4, no. 7 (2023): 100779: average false-positive rate on human TOEFL essays 61.3 percent (61.22 percent in the preprint, arXiv:2304.02819). Hermes Agent: https://hermes-agent.org/ (open-source agent runtime, February 2026). Hermes 4 is a separate chat/reasoning model family; Nous Portal docs state it is not recommended inside Hermes Agent. Grok Bot: xAI, “Introducing Grok Bot,” 11 August 2026, https://x.ai/news/introducing-grok-bot, and Grok Bot FAQ, https://docs.x.ai/grok-bot/faq — cloud computer; no education product and no FERPA claim on the pages opened 31 August 2026. Grok the assistant, Grok Bot, and the El Salvador school partnership for Grok the assistant are three different objects. [^20]: . Linnea C. Ehri, Simone R. Nunes, Dale M. Willows, Barbara Valeska Schuster, Zohreh Yaghoub-Zadeh, and Timothy Shanahan, “Phonemic Awareness Instruction Helps Children Learn to Read: Evidence From the National Reading Panel’s Meta-Analysis,” *Reading Research Quarterly* 36, no. 3 (2001): 250–87: d = 0.86 on phonemic awareness, d = 0.53 on reading, d = 0.59 on spelling; effects larger with letters, one or two skills, small groups, and 5–18 hours rather than longer. Timothy Shanahan, *The National Reading Panel Report: Practical Advice for Teachers* (Naperville, IL: Learning Point Associates, 2006): about 15 minutes a day for a semester of kindergarten is a research-shaped translation, not a statute. [^21]: . Louisa C. Moats, “How Spelling Supports Reading,” *American Educator* 29, no. 4 (Winter 2005/06): 12–22, 42–43. The 15–20 minutes daily or 30 minutes three times a week figure is Moats’s professional judgment in that essay, not a randomized trial. Use it as a ceiling on the spelling block, not as a federal minute count. [^22]: . Philip B. Gough and William E. Tunmer, “Decoding, Reading, and Reading Disability,” *Remedial and Special Education* 7, no. 1 (1986): 6–10. “Reading equals the product of decoding and comprehension, or (R = D C), where each variable ranges from 0 (nullity) to 1 (perfection).” Comprehension in this formula is linguistic (listening) comprehension. [^23]: . Hollis S. Scarborough, “Connecting Early Language and Literacy to Later Reading (Dis)abilities: Evidence, Theory, and Practice,” in *Handbook of Early Literacy Research*, vol. 1, ed. Susan B. Neuman and David K. Dickinson (New York: Guilford Press, 2001), 97–110. Eight strands; fluency is not a ninth. Scarborough, “The Reading Rope: Key Ideas Behind the Metaphor,” presentation to The Reading League, 2023, is Scarborough explaining her own 2001 metaphor. [^24]: . Common Core State Standards Initiative, *Common Core State Standards for English Language Arts & Literacy in History/Social Studies, Science, and Technical Subjects* (National Governors Association Center for Best Practices and Council of Chief State School Officers, 2010), ADA-compliant posting, https://www.corestandards.org/wp-content/uploads/2023/09/ADA-Compliant-ELA-Standards.pdf, accessed 31 August 2026. Introduction: “The Standards define what all students are expected to know and be able to do, not how teachers should teach.” Cited as a map, not a homeschool mandate. [^25]: . CCSS ELA, Reading Standards: Foundational Skills (K–5). Phonological awareness headings stop after Grade 1; phonics and fluency run through Grade 5. No RF strand in grades 6–12. [^26]: . 19 Tex. Admin. Code § 110.3 (Grade 1 English Language Arts and Reading, adopted 2017, amended 2019). Encoding and decoding named as reciprocal skills. [^27]: . Virginia Board of Education, *English Standards of Learning for Virginia Public Schools* (adopted 28 March 2024). Phonological and phonemic awareness specified through Grade 2; transition to cursive begins in Grade 2 (2.FFW.1). [^28]: . Florida Department of Education, *Benchmarks for Excellent Student Thinking (B.E.S.T.) Standards for English Language Arts* (2020/2021 posting on CPALMS). Phonological awareness in the regular progression through Grade 1; Grade 2 handwriting still print (ELA.2.C.1.1); clarification that many high-frequency words are irregular or temporarily irregular. [^29]: . National Institute of Child Health and Human Development, *Report of the National Reading Panel: Teaching Children to Read*, NIH Pub. No. 00-4769 (Washington, DC: U.S. Government Printing Office, 2000). Linnea C. Ehri, Simone R. Nunes, Steven A. Stahl, and Dale M. Willows, “Systematic Phonics Instruction Helps Students Learn to Read: Evidence from the National Reading Panel’s Meta-Analysis,” ERIC ED479646: overall mean (d = +0.41); kindergarten and first grade (d = 0.55). Phonics is a means to an end. [^30]: . Linnea C. Ehri, Simone R. Nunes, Dale M. Willows, Barbara Valeska Schuster, Zohreh Yaghoub-Zadeh, and Timothy Shanahan, “Phonemic Awareness Instruction Helps Children Learn to Read: Evidence from the National Reading Panel’s Meta-Analysis,” *Reading Research Quarterly* 36, no. 3 (2001): 250–287. Fifty-two studies, ninety-six comparisons; (d = 0.86) on phonemic awareness, (d = 0.53) on reading, (d = 0.59) on spelling. Larger with letters; one or two skills; 5–18 hours. [^31]: . Linnea C. Ehri, “Learning to Read Words: Theory, Findings, and Issues,” *Scientific Studies of Reading* 9, no. 2 (2005): 167–188; Ehri, “Orthographic Mapping in the Acquisition of Sight Word Reading, Spelling Memory, and Vocabulary Learning,” *Scientific Studies of Reading* 18, no. 1 (2014): 5–21. Four overlapping phases. Sight-word reading is reading from memory after letter–sound connections have been formed. Prediction from initial letters plus context must be verified against the spelling. [^32]: . Barbara Foorman, Nicholas Beyler, Kelly Borradaile, Michael Coyne, Carolyn A. Denton, Joseph Dimino, Joshua Furgeson, et al., *Foundational Skills to Support Reading for Understanding in Kindergarten through 3rd Grade*, NCEE 2016-4008 (Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Department of Education, 2016). Recommendation 4: connected text every day; the panel discourages guessing unknown words using beginning letters or pictures. Recommendations 2–3 (strong evidence): segments of sound linked to letters; decode, analyze word parts, write and recognize words. [^33]: . Ohio Rev. Code § 3313.6028 (2023): core curriculum “does not rely on any model of teaching students to read based on meaning, structure and syntax, and visual cues, including a three-cueing approach.” Louisiana Act 517 (2022) prohibits the three-cueing system / MSV technique in curriculum and instructional materials. Pictures, syntax, and meaning remain useful for monitoring meaning after the word is identified. [^34]: . Foorman et al., *Foundational Skills* (2016), Obstacle 3.1: invented spelling useful in kindergarten and first-grade independent writing; it should recede as studied patterns are held to standard spelling. Louisa C. Moats, “How Spelling Supports Reading,” *American Educator* 29, no. 4 (Winter 2005/06): 12–22, 42–43. [^35]: . 19 Tex. Admin. Code § 110.3. [^36]: . Wesley A. Hoover and Philip B. Gough, “The Simple View of Reading,” *Reading and Writing* 2, no. 2 (1990): 127–160. Longitudinal English–Spanish bilingual sample, grades 1–4; the product improved prediction beyond the linear combination. Abstract fetched 31 August 2026. [^37]: . Timothy Shanahan, *The National Reading Panel Report: Practical Advice for Teachers* (Naperville, IL: Learning Point Associates, 2006). Duration associated with the greatest reading gains: 5–18 hours; about 15 minutes a day for a semester of kindergarten, adjusted diagnostically. Segmenting and blending gave the greatest reading advantage. [^38]: . CCSS ELA RF.1.3, RF.1.4, RF.2.3, RF.2.4. [^39]: . 19 Tex. Admin. Code § 110.3(b)(2) (phonics, syllable types). [^40]: . Florida B.E.S.T. ELA, Grade 1–2 foundations clarifications on high-frequency / sight words. [^41]: . Foorman et al., *Foundational Skills* (2016), Recommendation 3: a few essential non-decodable words may be introduced as whole words before a text; high-frequency regular and irregular words are still taught. [^42]: . Steve Graham, Debra McKeown, Sharlene Kiuhara, and Karen R. Harris, “A Meta-Analysis of Writing Instruction for Students in the Elementary Grades,” *Journal of Educational Psychology* 104, no. 4 (2012): 879–896. Teaching transcription skills, effect size 0.55. Elementary grades 1–6; 115 documents. [^43]: . Timothy Shanahan and Christopher J. Lonigan, “The National Early Literacy Panel: A Summary of the Process and the Report,” *Educational Researcher* 39, no. 4 (2010): 279–285. Simple vocabulary average (r .33) with later decoding, reading comprehension, and spelling; complex oral language (grammar, defining words, listening comprehension) (r = .70) with later reading comprehension. [^44]: . CCSS ELA W.1.1–W.1.3; parallel opinion / informative / narrative pieces on the Texas, Virginia, and Florida maps. [^45]: . Steve Graham and Dolores Perin, “A Meta-Analysis of Writing Instruction for Adolescent Students,” *Journal of Educational Psychology* 99, no. 3 (2007): 445–476. Grammar instruction as tested, weighted average effect size (-0.32) on writing quality; sentence combining (0.50). Grades 4–12. Named here as the later finding that already supports teaching conventions inside the sentence, not as a first-grade experiment. [^46]: . Moats, “How Spelling Supports Reading,” citing Paula D. Mehta, Barbara R. Foorman, Lee Branum-Martin, and W. Patrick Taylor, “Literacy as a Unidimensional Construct: Validation, Sources of Influence, and Implications in a Longitudinal Study in Grades 1–4,” *Scientific Studies of Reading* 9, no. 2 (2005): 85–116. Houston sample: passage comprehension stayed near the 50th percentile while spelling percentiles dropped (58th to 20th by Grade 4), as reported by Moats. [^47]: . CCSS ELA RF.1.1: distinguishing features of a sentence (first word, capitalization, ending punctuation). [^48]: . Foorman et al., *Foundational Skills* (2016), Recommendation 4 (moderate evidence). [^49]: . John Sweller and Graham A. Cooper, “The Use of Worked Examples as a Substitute for Problem Solving in Learning Algebra,” *Cognition and Instruction* 2, no. 1 (1985): 59–89; Alexander Renkl and Robert K. Atkinson, “Structuring the Transition from Example Study to Problem Solving in Cognitive Skill Acquisition,” *Educational Psychologist* 38, no. 1 (2003): 15–22. Worked-example and fading method, applied here to a word card; not a new first-grade reading trial. [^50]: . CCSS ELA RF.K.1 and RF.1.1. [^51]: . 19 Tex. Admin. Code § 110.3 (at least 100 high-frequency words from a research-based list in Grade 1); Florida B.E.S.T. Grade 1 fluency clarifications. [^52]: . Mixing types as a practice. Clean classroom randomized trials of interleaved mathematics (Doug Rohrer and colleagues) are grade 7 and are not treated here as an English coefficient. [^53]: . NICHD, *Report of the National Reading Panel* (2000), fluency: guided repeated oral reading with feedback from teachers, peers, or parents had a significant positive impact on word recognition, fluency, and comprehension. The Panel did not find sufficient causal evidence that programs encouraging large amounts of independent silent reading, with minimal guidance, reliably increase how much students read or improve reading skills. [^54]: . Scarborough, “Connecting Early Language and Literacy” (2001); Scarborough 2023 presentation. [^55]: . Ehri et al., “Phonemic Awareness Instruction” (2001); Shanahan, *Practical Advice* (2006). [^56]: . International Dyslexia Association, definition adopted by the IDA Board of Directors, 12 November 2002 (IDA Definition Consensus Project). This chapter does not diagnose any child. [^57]: . Gough and Tunmer, “Decoding, Reading, and Reading Disability” (1986). Within the simple view, the proximal problem of dyslexia is an inability to decode. [^58]: . Foorman et al., *Foundational Skills* (2016), Obstacle 2.2. [^59]: . NICHD, *Report of the National Reading Panel* (2000): the Panel rejected the claim that kindergarten children are “not ready” for phonics; effects in K and Grade 1 were significant and substantial. [^60]: . Sarah G. Wood, Jennifer H. Moxley, Elizabeth L. Tighe, and Richard K. Wagner, “Does Use of Text-to-Speech and Related Read-Aloud Tools Improve Reading Comprehension for Students with Reading Disabilities? A Meta-Analysis,” *Journal of Learning Disabilities* 51, no. 1 (2018): 73–84. Average weighted ({d} = 0.35) on comprehension for students with reading disabilities. Compensation and access, not a phonics program. [^61]: . Nicholas C. Soderstrom and Robert A. Bjork, “Learning Versus Performance: An Integrative Review,” *Perspectives on Psychological Science* 10, no. 2 (2015): 176–199. Current performance is often an unreliable index of learning. [^62]: . Hamsa Bastani, Osbert Bastani, Alp Sungu, Haosen Ge, Özge Kabakcı, and Rei Mariman, “Generative AI without Guardrails Can Harm Learning: Evidence from High School Mathematics,” *Proceedings of the National Academy of Sciences* 122, no. 26 (2025): e2422633122. GPT Base raised assisted practice grades 48 percent relative to control, then cut unaided exam grades 17 percent, in a high-school mathematics RCT. Not a reading or writing trial. The sitting condition — a tool that produces the response — is the mechanism used here. [^63]: . All About Reading / All About Spelling, Logic of English, and similar code-emphasis programs advertise placement tests (vendor descriptions). No homeschool-curriculum randomized trial comparing them was opened. Choose by fit. [^64]: . Philip B. Gough and William E. Tunmer, “Decoding, Reading, and Reading Disability,” *Remedial and Special Education* 7, no. 1 (1986): 6–10. Last chapter introduced the product (R = D C); fluency is the coordination of the two factors in running text. Hollis S. Scarborough treats fluency as an indication of overall skill, not a ninth strand of her rope (“Connecting Early Language and Literacy to Later Reading (Dis)abilities,” in *Handbook of Early Literacy Research*, vol. 1, ed. Susan B. Neuman and David K. Dickinson \[New York: Guilford Press, 2001\], 97–110). [^65]: . National Institute of Child Health and Human Development, *Report of the National Reading Panel: Teaching Children to Read*, NIH Pub. No. 00-4769 (Washington, DC: U.S. Government Printing Office, 2000), fluency findings. Guided repeated oral reading with feedback from teachers, peers, or parents had a significant positive impact on word recognition, fluency, and comprehension. The Panel did not find sufficient causal evidence that programs encouraging large amounts of independent silent reading, with minimal guidance, reliably increase how much students read or improve reading skills. This chapter does not invent a single Panel fluency *d*. [^66]: . Timothy Rasinski, “Creating Fluent Readers,” *Educational Leadership* 61, no. 6 (2004): 46–51. Emphasizing speed at the expense of prosody is “a corruption of the concept of reading fluency.” [^67]: . Common Core State Standards Initiative, *Common Core State Standards for English Language Arts & Literacy*, ADA-compliant posting, https://www.corestandards.org/wp-content/uploads/2023/09/ADA-Compliant-ELA-Standards.pdf, accessed 31 August 2026. RF.3.3, RF.3.4, RF.4.3, RF.4.4, RF.5.3, RF.5.4. Grade 5 is the last grade with an RF strand. The Standards “define what all students are expected to know and be able to do, not how teachers should teach.” Map, not mandate. [^68]: . CCSS ELA W.4.6 (one page in a sitting); W.5.6 (two pages); W.6.6 (three pages). [^69]: . CCSS ELA W.4.9 / W.5.9: draw evidence from literary or informational texts. W.9 begins in Grade 4. [^70]: . Daniel T. Willingham, “How Knowledge Helps: It Speeds and Strengthens Reading Comprehension, Learning — and Thinking,” *American Educator* 30, no. 1 (Spring 2006): 30–37. [^71]: . Donna R. Recht and Lauren Leslie, “Effect of Prior Knowledge on Good and Poor Readers’ Memory of Text,” *Journal of Educational Psychology* 80, no. 1 (1988): 16–20, as described by Willingham, “How Knowledge Helps” (2006). Junior-high students who knew baseball comprehended a baseball passage better than “good readers” who did not, even when the passage was written at an early fifth-grade level. [^72]: . Daniel T. Willingham and Gail Lovette, “Can Reading Comprehension Be Taught?,” *Teachers College Record*, September 26, 2014. Strategy instruction “undeniably works” but functions as “a bag of tricks” that are “quickly learned and don’t require a lot of practice”; a handful of sessions may confer the same benefit as fifty. “When it comes to improving reading comprehension, strategy instruction may have an upper limit, but building background knowledge does not.” [^73]: . Timothy Shanahan, *The National Reading Panel Report: Practical Advice for Teachers* (Naperville, IL: Learning Point Associates, 2006). Seven strategies with sufficient evidence; multiple strategies together produced more learning; studies typically used texts students could already decode; duration in the studies often daily lessons for four or more weeks. Reciprocal teaching (Annemarie Sullivan Palincsar and Ann L. Brown, “Reciprocal Teaching of Comprehension-Fostering and Comprehension-Monitoring Activities,” *Cognition and Instruction* 1, no. 2 \[1984\]: 117–175) as I do / we do / you do. [^74]: . Richard L. Venezky, “English Orthography: Its Graphical Structure and Its Relation to Sound,” *Reading Research Quarterly* 2, no. 3 (1967): 75–105. Spelling maps first onto a morphophonemic level, then into sound. [^75]: . Louisa C. Moats, “How Spelling Supports Reading,” *American Educator* 29, no. 4 (Winter 2005/06): 12–22, 42–43, citing Paul R. Hanna, Jean S. Hanna, Richard E. Hodges, and Edwin H. Rudorf, *Phoneme-Grapheme Correspondences as Cues to Spelling Improvement*, OE-32008 (Washington, DC: U.S. Office of Education, 1966). About half of English words spelled from sound-symbol alone; another 34 percent one error; about four percent truly irregular. [^76]: . CCSS ELA RF.3.3; RF.4.3 / RF.5.3. [^77]: . Florida Department of Education, *B.E.S.T. Standards for English Language Arts* (2020/2021 CPALMS posting). Greek and Latin morphology from Grade 3 in the Vocabulary strand; “Knowledge Matters” among the document’s premises. [^78]: . 19 Tex. Admin. Code § 110.6 (Grade 4): beginning reading and writing, including syllable types, syllable division, prefixes and suffixes, matching spelling; write legibly in cursive to complete assignments; bibliography; paraphrasing versus plagiarism. [^79]: . Virginia Board of Education, *English Standards of Learning for Virginia Public Schools* (adopted 28 March 2024). Phonological and phonemic awareness headings stop after Grade 2; phonics and word analysis continue through Grade 5. [^80]: . NICHD, *Report of the National Reading Panel* (2000), vocabulary; Shanahan, *Practical Advice* (2006). Direct and indirect methods; copying dictionary definitions the weaker comparison. [^81]: . Shanahan, *Practical Advice* (2006), presenting Isabel L. Beck, Margaret G. McKeown, and Linda Kucan’s three-tier cut and the observation that successful programs teach a few hundred words thoroughly. [^82]: . NICHD, *Report of the National Reading Panel* (2000), text-comprehension instruction. Seven strategies: question asking, monitoring, summarization, question answering, story mapping, graphic organizers, cooperative grouping. [^83]: . Palincsar and Brown, “Reciprocal Teaching” (1984); Shanahan, *Practical Advice* (2006). [^84]: . Florida B.E.S.T. ELA, “Knowledge Matters”: “Reading comprehension depends more on relevant background knowledge than on mastery of reading strategies.” State-map premise, not a federal slogan. [^85]: . CCSS ELA W.3.1–W.3.3, W.4.1–W.4.3, W.5.1–W.5.3; W.4.9 / W.5.9. [^86]: . Steve Graham, Debra McKeown, Sharlene Kiuhara, and Karen R. Harris, “A Meta-Analysis of Writing Instruction for Students in the Elementary Grades,” *Journal of Educational Psychology* 104, no. 4 (2012): 879–896. Strategy instruction ES = 1.02; transcription skills 0.55; extra writing 0.30; product goals 0.76; grammar instruction did not produce a statistically significant positive effect. This is a useful study, not a promise that every home will see the same result. [^87]: . Catherine E. Snow, Peg Griffin, and M. Susan Burns, *Knowledge to Support the Teaching of Reading: Preparing Teachers for a Changing World* (San Francisco: Jossey-Bass, 2005), quoted in Moats, “How Spelling Supports Reading,” 86. [^88]: . Moats, “How Spelling Supports Reading” (2005/06). Professional-judgment sequence, not an RCT. [^89]: . Barbara Foorman et al., *Foundational Skills to Support Reading for Understanding in Kindergarten through 3rd Grade*, NCEE 2016-4008 (Washington, DC: Institute of Education Sciences, 2016), Obstacle 3.1 and Recommendation 4 (moderate evidence, 22 studies). [^90]: . Moats, “How Spelling Supports Reading,” citing Paula D. Mehta, Barbara R. Foorman, Lee Branum-Martin, and W. Patrick Taylor, “Literacy as a Unidimensional Construct,” *Scientific Studies of Reading* 9, no. 2 (2005): 85–116. [^91]: . Foorman et al., *Foundational Skills* (2016), Recommendation 4. [^92]: . CCSS ELA, “What is not covered by the Standards” and related notes: several standards can be addressed by a single rich task. Household application, not a new finding. [^93]: . 19 Tex. Admin. Code § 110.6(b)(2)(C); Virginia SOL 2024, 3.FFW.1, 4.FFW.1, 5.FFW.1; Florida B.E.S.T. ELA.3.C.1.1, ELA.4.C.1.1, ELA.5.C.1.1. [^94]: . Steve Graham and Dolores Perin, “A Meta-Analysis of Writing Instruction for Adolescent Students,” *Journal of Educational Psychology* 99, no. 3 (2007): 445–476, product goals 0.70; Graham et al., elementary meta-analysis (2012), product goals 0.76. [^95]: . Hugh W. Catts, Suzanne M. Adlof, and Susan Ellis Weismer, “Language Deficits in Poor Comprehenders: A Case for the Simple View of Reading,” *Journal of Speech, Language, and Hearing Research* 49, no. 2 (2006): 278–293. Eighth-grade poor comprehenders: language-comprehension deficits and normal phonological processing; poor decoders the opposite pattern; profiles visible retrospectively in kindergarten, Grade 2, and Grade 4. Abstract fetched 31 August 2026; full PDF not opened. [^96]: . Sarah G. Wood, Jennifer H. Moxley, Elizabeth L. Tighe, and Richard K. Wagner, “Does Use of Text-to-Speech and Related Read-Aloud Tools Improve Reading Comprehension for Students with Reading Disabilities? A Meta-Analysis,” *Journal of Learning Disabilities* 51, no. 1 (2018): 73–84. Average weighted ({d} = 0.35). Compensation and access, not a fluency or phonics program. [^97]: . Nicholas C. Soderstrom and Robert A. Bjork, “Learning Versus Performance: An Integrative Review,” *Perspectives on Psychological Science* 10, no. 2 (2015): 176–199. [^98]: . Hamsa Bastani, Osbert Bastani, Alp Sungu, Haosen Ge, Özge Kabakcı, and Rei Mariman, “Generative AI without Guardrails Can Harm Learning: Evidence from High School Mathematics,” *Proceedings of the National Academy of Sciences* 122, no. 26 (2025): e2422633122. Mathematics RCT, not a writing RCT. Mechanism: if the tool produces the response, the child is a spectator. [^99]: . All About Reading, literature-rich / living-book years, and textbook-anthology paths named by fit, not ranked. Vendor placement language (All About Reading placement tests; Brave Writer “don’t be governed by age range”; Well-Trained Mind support on starting Writing With Skill at Level 1) is vendor. No homeschool-curriculum RCT was opened for the research behind this book. [^100]: . Common Core State Standards Initiative, *Common Core State Standards for English Language Arts & Literacy*, ADA-compliant posting, https://www.corestandards.org/wp-content/uploads/2023/09/ADA-Compliant-ELA-Standards.pdf, accessed 31 August 2026. Foundational Skills (RF) are K–5 only. “The Standards define what all students are expected to know and be able to do, not how teachers should teach.” Map, not mandate. [^101]: . Florida Department of Education, *B.E.S.T. Standards for English Language Arts* (2020/2021 CPALMS posting). Foundational reading intervention standards for grades 6–12 intensive electives, “not core ELA courses.” [^102]: . CCSS ELA, grades 6–8 Reading Literature and Informational (cite textual evidence; theme/central idea; evaluate arguments); Writing (claims, reasons, evidence; alternate or opposing claims by 7–8); W.6.6 three pages in a sitting; research with standard citation format by 7–8. [^103]: . Florida B.E.S.T. ELA.6.R.3.4, “Identify rhetorical appeals in a text.” The document’s own purpose language: so students will have an understanding of logos, pathos, and ethos when starting argumentative writing. Ordinary-language claims remain the household teaching. [^104]: . CCSS ELA Introduction, footnote 1: “The percentages on the table reflect the sum of student reading, not just reading in ELA settings. Teachers of senior English classes, for example, are not required to devote 70 percent of reading to informational texts.” [^105]: . 19 Tex. Admin. Code § 110.36 (English I): students read and respond to American, British, and world literature inside one course. Middle-school TEKS (19 TAC §§ 110.22–110.24) keep the seven strands; the middle-school pages were not opened in full for the research behind this book. High-school titles remain English I–IV. [^106]: . Virginia Board of Education, *English Standards of Learning for Virginia Public Schools* (adopted 28 March 2024), guiding principle 9: the state does not provide a list of prescribed books; graduates should have been exposed to foundational authors such as Homer and Shakespeare among others. Secondary writing, in the introduction, includes expository, persuasive, argumentative, analytical, reflective, and technical purposes. Grade 6–12 student-expectation pages were truncated in the fetch used for this book; the introduction and K–5 pages were opened in full. [^107]: . Florida B.E.S.T. ELA, sample texts printed by benchmark in the standards document. State exemplars, not a federal canon. [^108]: . Steve Graham and Dolores Perin, “A Meta-Analysis of Writing Instruction for Adolescent Students,” *Journal of Educational Psychology* 99, no. 3 (2007): 445–476. Strategy instruction 0.82; summarization 0.82; peer assistance 0.75; product goals 0.70; word processing 0.55; sentence combining 0.50; process writing 0.32; grammar instruction −0.32. SRSD (Harris and Graham) as a subset of strategy instruction: 1.14 versus 0.62 for non-SRSD strategy instruction. Grammar in this analysis is explicit teaching of parts of speech and sentence types, often versus a treatment that itself taught writing. [^109]: . Hugh W. Catts, Suzanne M. Adlof, and Susan Ellis Weismer, “Language Deficits in Poor Comprehenders: A Case for the Simple View of Reading,” *Journal of Speech, Language, and Hearing Research* 49, no. 2 (2006): 278–293. Abstract fetched 31 August 2026; full PDF not opened. [^110]: . CCSS ELA W.6.1, W.7.1, W.8.1. Alternate or opposing claims enter in grades 7–8. [^111]: . Florida B.E.S.T. ELA.6.C.1.3 (claim, logical reasoning, relevant evidence from multiple sources); Grade 7 adds acknowledging a counterclaim; Grade 8 adds rebutting a counterclaim. [^112]: . CCSS ELA RI.6–8, evaluate arguments. [^113]: . CCSS ELA RL/RI.6.1 (cite textual evidence); Grade 7, several pieces; Grade 8, stronger use of evidence and inferences. Full Grade 8 wording is in the PDF. [^114]: . SAT Reading and Writing and ACT English are college-ready signals, not this band’s course. College Board benchmark for SAT Reading and Writing, 480; ACT English benchmark, 18, as opened on the publishers’ benchmarks pages for the research behind this book (31 August 2026). [^115]: . Graham and Perin, “A Meta-Analysis of Writing Instruction” (2007), sentence combining 0.50; grammar −0.32. [^116]: . CCSS ELA W.7 from Grade 1 through high school; W.8 gathering, notes, and (by 6–8) standard citation format. Texas inquiry from Grade 1; Grade 4 bibliography and paraphrasing versus plagiarism already on the elementary map. [^117]: . Daniel T. Willingham, “How Knowledge Helps,” *American Educator* 30, no. 1 (Spring 2006): 30–37; Donna R. Recht and Lauren Leslie, “Effect of Prior Knowledge on Good and Poor Readers’ Memory of Text,” *Journal of Educational Psychology* 80, no. 1 (1988): 16–20, as described by Willingham. Last chapter’s baseball picture, applied here to the worlds middle-school pages assume. [^118]: . Florida B.E.S.T. 6–12 foundational intervention standards, intensive electives, not core ELA. [^119]: . Graham and Perin (2007), as in note 9. [^120]: . Catts, Adlof, and Weismer (2006), as in note 10. [^121]: . Weixin Liang, Mert Yuksekgonul, Yining Mao, Eric Wu, and James Zou, “GPT Detectors Are Biased against Non-Native English Writers,” *Patterns* 4, no. 7 (2023): 100779. Average false-positive rate on human TOEFL essays 61.3 percent (published). A detector is not a verdict, and it is not part of the English hour. [^122]: . Nicholas C. Soderstrom and Robert A. Bjork, “Learning Versus Performance: An Integrative Review,” *Perspectives on Psychological Science* 10, no. 2 (2015): 176–199. [^123]: . Hamsa Bastani, Osbert Bastani, Alp Sungu, Haosen Ge, Özge Kabakcı, and Rei Mariman, “Generative AI without Guardrails Can Harm Learning: Evidence from High School Mathematics,” *Proceedings of the National Academy of Sciences* 122, no. 26 (2025): e2422633122. Mathematics RCT. The paper notes that math has objective scoring “which is unavailable in creative subjects such as writing.” Mechanism, not a writing trial. [^124]: . Literature-rich / living-book years (including AmblesideOnline as a named fit), textbook-anthology paths (including HMH *Into Literature* as a vendor design), and writing-first programs (IEW, Brave Writer, Well-Trained Mind sequences) named by fit, not ranked. No homeschool-curriculum RCT was opened for the research behind this book. Transcript titles a stranger can read: Language Arts, English 9, English 10, American Literature, British Literature, World Literature, Composition, Speech. Never a brand. Never Critical Thinking 101. [^125]: . National Center for Education Statistics, State Education Practices, Table 5.7, “State high school graduation requirements, 2018,” https://nces.ed.gov/programs/statereform/tab5_7.asp, accessed 31 August 2026. Nearly every listed state requires 4.00 English credits; California 3.00; Arkansas 4.50; Idaho 4.50; a few locally determined. Texas, Florida, and Virginia 4.00. Course titles vary by jurisdiction; four credits are not a federally named American–British–World sequence. [^126]: . Common Core State Standards Initiative, *Common Core State Standards for English Language Arts & Literacy in History/Social Studies, Science, and Technical Subjects* (NGA Center / CCSSO; official ADA-compliant posting of the 2 June 2010 standards), https://www.corestandards.org/wp-content/uploads/2023/09/ADA-Compliant-ELA-Standards.pdf, accessed 31 August 2026. Introduction: high-school writing emphasis consistent with NAEP; “the overwhelming focus of writing throughout high school should be on arguments and informative/explanatory texts.” Design-limit language in the same PDF: “The Standards define what all students are expected to know and be able to do, not how teachers should teach.” Writing types: arguments, informative/explanatory, narratives; research inside the Writing strand. W.9 begins in grade 4. [^127]: . 19 Tex. Admin. Code § 110.36 (English I) and § 110.38 (English III) (Cornell LII copies), accessed 31 August 2026. English I composition genres at §110.36(c)(10): literary, informational (explanatory essays, reports, personal essays), argumentative, correspondence. English I also requires students to “read and respond to American, British, and world literature” inside one course. English III adds literary analysis and rhetorical analysis, §110.38(c)(10)(E)–(F). Legal titles remain English I–IV, one credit each. [^128]: . Florida Department of Education, *Benchmarks for Excellent Student Thinking (B.E.S.T.) Standards for English Language Arts* (2020/2021 CPALMS posting), Communication strand C.1 writing ladder: narrative, argumentative, and expository at every grade, with increasing control of counterclaim, sources, and tone; 11–12 literary analysis. Improving writing (C.1.5) recursive. https://cpalmsmediaprod.blob.core.windows.net/uploads/docs/standards/best/la/elabeststandardsfinal2108.pdf, accessed 31 August 2026. [^129]: . Steve Graham and Dolores Perin, “A Meta-Analysis of Writing Instruction for Adolescent Students,” *Journal of Educational Psychology* 99, no. 3 (2007): 445–476. Grades 4–12; 123 documents; 154 effect sizes for writing quality. Weighted averages: strategy instruction 0.82, summarization 0.82, peer assistance 0.75, product goals 0.70, word processing 0.55, sentence combining 0.50, inquiry 0.32, prewriting 0.32, process writing approach 0.32, study of models 0.25, grammar instruction −0.32. Process writing with professional development 0.46; without 0.03 (nonsignificant). Grammar in this analysis is explicit teaching of parts of speech and sentence types, often versus a treatment that itself taught writing; the negative sign means those conditions lost to the comparison. Averages across heterogeneous studies; not a promise for every home. [^130]: . Graham and Perin, “Meta-Analysis of Writing Instruction” (2007). SRSD (Karen R. Harris and Steve Graham) as a subset of strategy instruction: weighted mean 1.14 (8 studies) versus 0.62 for non-SRSD strategy instruction. Six stages as described there: background knowledge; describe the strategy; model; memorize; support; independent use, with self-regulation (goal setting, self-monitoring, self-instructions, self-reinforcement) taught alongside. Effects concentrated in planning and revising for specific genres, often with struggling writers, often on researcher-made quality measures. Do not read as “SRSD raises NAEP writing by 1.14 standard deviations.” [^131]: . NCAA Eligibility Center, *Home School Toolkit 2025–26*. Sample transcript titles include English 9–12 and Composition I/II. Core-Course Worksheet: course name should match the transcript title. Division I: 4 years English; Division II: 3 years English. NCAA is a private association’s eligibility rules, not a statute. CLEP / credit-by-exam is not an NCAA-approved core course. [^132]: . College Board, AP English Language and Composition course page, Effective Fall 2024 CED description, https://apcentral.collegeboard.org/courses/ap-english-language-and-composition, accessed 31 August 2026. Introductory college-level composition; rhetorical analysis of nonfiction; no prerequisite course. Exam overview: 3 hours 15 minutes, 45 multiple-choice and 3 free-response. Full CED PDF not opened this research pass. Credit is set by each college. [^133]: . College Board, CLEP College Composition, https://clep.collegeboard.org/clep-exams/college-composition, accessed 31 August 2026. 50 multiple-choice in 55 minutes plus two mandatory typed essays in 70 minutes; conventions about 10 percent, revision 40, source materials 25, rhetorical analysis 25. ACE recommends 6 semester hours at score 50; each institution sets policy. [^134]: . College Board SAT Suite College and Career Readiness Benchmarks, accessed 31 August 2026: Reading and Writing benchmark 480, defined as a 75 percent chance of at least a C in first-semester credit-bearing courses in history, literature, social sciences, or writing. Digital SAT Reading and Writing: 54 questions, 64 minutes. ACT College Readiness Benchmarks: English 18, defined as a 50 percent chance of a B or higher (about 75–80 percent chance of a C or higher) in English Composition I. The two benchmarks are not equated to each other by either publisher on the pages opened. [^135]: . Graham and Perin, “Meta-Analysis of Writing Instruction” (2007), product goals 0.70. [^136]: . Graham and Perin, “Meta-Analysis of Writing Instruction” (2007), sentence combining 0.50. Sentence combining is constructing more complex sentences from basic ones. [^137]: . Harris and Graham’s SRSD stages as reported in Graham and Perin (2007). Karen R. Harris and Steve Graham, *Making the Writing Process Work: Strategies for Composition and Self-Regulation* (Brookline, MA: Brookline Books, 1996), the named book behind that subset. [^138]: . Judith C. Hochman and Natalie Wexler, *The Writing Revolution: A Guide to Advancing Thinking Through Writing in All Subjects and Grades* (San Francisco: Jossey-Bass, 2017); and Hochman and Wexler, “One Sentence at a Time,” *American Educator*, Summer 2017. Named method (sentence expansion, combining, topic sentences, outlines), not a randomized trial. No *d* is claimed here. [^139]: . College Board, AP English Literature and Composition course page, https://apcentral.collegeboard.org/courses/ap-english-literature-and-composition, accessed 31 August 2026. Introductory college-level literary analysis of fiction, poetry, and drama; no prerequisite course. Distinct from AP English Language. [^140]: . Graham and Perin (2007), process writing 0.32, 0.03 without professional development. Steve Graham, Debra McKeown, Sharlene Kiuhara, and Karen R. Harris, “A Meta-Analysis of Writing Instruction for Students in the Elementary Grades,” *Journal of Educational Psychology* 104, no. 4 (2012): extra writing ES = 0.30; strategy instruction 1.02; grammar instruction not a statistically significant positive effect; SRSD moderator 1.17. Abstract coefficients; full 2012 PDF not opened. Different grades and inclusion from 2007; keep the two metas as two metas. [^141]: . Hamsa Bastani, Osbert Bastani, Alp Sungu, Haosen Ge, Özge Kabakcı, and Rei Mariman, “Generative AI without Guardrails Can Harm Learning: Evidence from High School Mathematics,” *Proceedings of the National Academy of Sciences* 122, no. 26 (2025): e2422633122. Classroom-level RCT, nearly 1,000 Turkish high-school math students, grades 9–11, GPT-4-0613. GPT Base: assisted practice +48 percent relative to control; unaided exam −17 percent. GPT Tutor (hints, withheld full solution): practice +127 percent; unaided exam statistically indistinguishable from control. GPT Base 51 percent correct on the practice set. The paper notes that math has objective scoring “which is unavailable in creative subjects such as writing.” Mechanism for the English hour, not a writing RCT. [^142]: . Nicholas C. Soderstrom and Robert A. Bjork, “Learning Versus Performance: An Integrative Review,” *Perspectives on Psychological Science* 10, no. 2 (2015): 176–199. Performance during acquisition is often an unreliable index of learning. [^143]: . CCSS ELA (n. 126), W.4.6 one page in a sitting; W.5.6 two pages; W.6.6 three pages. Keyboarding as a writing-production skill, not a homeschool minute-count. [^144]: . Graham and Perin (2007), peer assistance 0.75. [^145]: . Steve Graham et al., *Teaching Elementary School Students to Be Effective Writers*, WWC practice guide (NCEE / Institute of Education Sciences, 2012, revised 2018), Recommendation 3. Writing “places multiple simultaneous demands on the writer”; mastery of handwriting, spelling, sentence construction, and typing “allows students to devote more of their attention to composing.” Graham et al. (2012) elementary meta-analysis, transcription skills ES = 0.55, lines up with that recommendation. Elementary guide used here for the production-skill claim, not as a high-school RCT. [^146]: . NCES, *Dual or Concurrent Enrollment in Public Schools in the United States*, NCES 2020-125 (2017–18 NTPS): 82 percent of public schools with any of grades 9–12 *offered* dual/concurrent enrollment. Schools offering, not students taking, not English-specific, not a homeschool rate. [^147]: . Graham and Perin (2007), word processing 0.55. [^148]: . NCES Table 5.7 (n. 125). [^149]: . Common Core State Standards Initiative, *Common Core State Standards for English Language Arts & Literacy in History/Social Studies, Science, and Technical Subjects* (NGA Center / CCSSO; official ADA-compliant posting of the 2 June 2010 standards), https://www.corestandards.org/wp-content/uploads/2023/09/ADA-Compliant-ELA-Standards.pdf, accessed 31 August 2026. “What is not Covered by the Standards”: the standards “do not—indeed, cannot—enumerate all or even most of the content that students should learn,” while naming mythology, foundational U.S. documents, and Shakespeare as examples. Design-limit language: “The Standards define what all students are expected to know and be able to do, not how teachers should teach.” Appendix B (text exemplars) was not re-opened as a separate PDF this research pass; this chapter does not print an exemplar list as a required canon. Footnote 1 of the ELA PDF: teachers of senior English classes are not required to devote 70 percent of reading to informational texts; the 70 percent figure is across the grade, not inside English class. [^150]: . Virginia Board of Education, *English Standards of Learning for Virginia Public Schools*, adopted 28 March 2024, https://www.doe.virginia.gov/home/showpublisheddocument/53643/638499760936600000, accessed 31 August 2026. Guiding principle 9: the state does not provide a list of prescribed books; graduates should have been exposed to foundational authors such as Homer and Shakespeare as well as works from different genres, cultures, geographies, and time periods. Board course titles: English 9–12; AP Literature may substitute for English 11 or 12 on the Board’s course list. [^151]: . Florida Department of Education, *B.E.S.T. Standards for English Language Arts* (2020/2021 CPALMS posting), accessed 31 August 2026. Sample texts by benchmark; Appendix B literary periods; Civic Literacy Reading List (not transcribed here). ELA.12.R.3.3, influence of classic literature on contemporary world texts; ELA.9.R.1.4, epic poetry. State samples, not a federal canon. [^152]: . 19 Tex. Admin. Code § 110.36 (English I) and § 110.38 (English III) (Cornell LII copies), accessed 31 August 2026. English I: read and respond to American, British, and world literature in a single course. English III: read and analyze American literature across literary periods; the legal credit remains English III. English II and IV literature emphasis not opened in full this research pass. [^153]: . National Center for Education Statistics, State Education Practices, Table 5.7, “State high school graduation requirements, 2018,” https://nces.ed.gov/programs/statereform/tab5_7.asp, accessed 31 August 2026. Four English credits typical; not a federally named American–British–World sequence. [^154]: . College Board, AP English Literature and Composition and AP English Language and Composition course pages, Effective Fall 2024 CED descriptions, accessed 31 August 2026. Literature: fiction, poetry, drama; Language: composition and rhetoric of nonfiction. No prerequisite course for either. Credit by college. Full CED PDFs not opened this pass; genre weightings for Literature multiple-choice as stated on the course page: short fiction 42–49 percent, poetry 36–45 percent, longer fiction or drama 15–18 percent. [^155]: . Philip B. Gough and William E. Tunmer, “Decoding, Reading, and Reading Disability,” *Remedial and Special Education* 7, no. 1 (1986): 6–10. Reading equals the product of decoding and linguistic comprehension, R = D × C, each factor 0 to 1. [^156]: . Daniel T. Willingham, “How Knowledge Helps,” *American Educator*, Spring 2006, describing Donna R. Recht and Lauren Leslie, “Effect of Prior Knowledge on Good and Poor Readers’ Memory of Text,” *Journal of Educational Psychology* 80, no. 1 (1988): 16–20. Junior-high students with baseball knowledge comprehended a baseball passage better than “good readers” without that knowledge, even at an early fifth-grade reading level. [^157]: . College Board, AP English Literature course page (n. 154). Units recycle short fiction, poetry, and longer fiction or drama. No required author list on the course page opened. [^158]: . College Board, CLEP Analyzing and Interpreting Literature, accessed 31 August 2026. About 80 multiple-choice items in 98 minutes; does not require specific works; optional essay scored by the institution. [^159]: . Daniel T. Willingham and Gail Lovette, “Can Reading Comprehension Be Taught?,” *Teachers College Record*, 2014. Strategy instruction “undeniably works” but functions as a “bag of tricks” quickly learned; a handful of sessions may confer the same benefit as fifty. “When it comes to improving reading comprehension, strategy instruction may have an upper limit, but building background knowledge does not.” National Reading Panel, *Teaching Children to Read*, NIH Pub. No. 00-4769 (2000): evidence for named comprehension strategies, stronger when several are taught together. Brief modeled strategy in the text, not a strategies hour. [^160]: . Florida B.E.S.T. (n. 151), comparative / adaptation strand including ELA.12.R.3.3. [^161]: . Nicholas C. Soderstrom and Robert A. Bjork, “Learning Versus Performance: An Integrative Review,” *Perspectives on Psychological Science* 10, no. 2 (2015). Hamsa Bastani et al., “Generative AI without Guardrails Can Harm Learning,” *PNAS* 122, no. 26 (2025): e2422633122 — math RCT; writing is easy to outsource because the model’s native product is a paragraph. Mechanism, not a literature trial. [^162]: . NCES Table 5.7 (n. 153). [^163]: . National Reading Panel, *Teaching Children to Read* (2000): guided repeated oral reading with feedback had a significant positive impact on word recognition, fluency, and comprehension. Timothy Rasinski, “Creating Fluent Readers,” *Educational Leadership* 61, no. 6 (2004): emphasizing speed at the expense of prosody is “a corruption of the concept of reading fluency.” [^164]: . National Reading Panel (2000): the Panel did not find sufficient causal evidence that programs encouraging large amounts of independent silent reading, with minimal guidance, reliably increase how much students read or improve reading skills; that finding does not negate a possible positive influence of wide reading. [^165]: . Steve Graham and Dolores Perin, “A Meta-Analysis of Writing Instruction for Adolescent Students,” *Journal of Educational Psychology* 99, no. 3 (2007): strategy instruction 0.82; extra process-writing package 0.32. The literature paragraph is composition pointed at a book. [^166]: . College Board AP Literature (n. 154); Virginia Board course list (n. 150). Dual enrollment: NCES 2020-125, 82 percent of public schools with grades 9–12 *offered* it (2017–18) — schools offering, not a homeschool take-rate. [^167]: . S. G. Wood, J. H. Moxley, E. L. Tighe, and R. K. Wagner, “Does Use of Text-to-Speech and Related Read-Aloud Tools Improve Reading Comprehension for Students with Reading Disabilities? A Meta-Analysis,” *Journal of Learning Disabilities* 51, no. 1 (2018): average weighted d̄ = 0.35 on comprehension for students with reading disabilities. Compensation and access, not a literature program that removes decoding. [^168]: . 19 Tex. Admin. Code § 110.36 (n. 152). [^169]: . NCES Table 5.7 (n. 153). NCAA Eligibility Center, *Home School Toolkit 2025–26*: sample titles include English 9–12; course name on the worksheet should match the transcript. Division I: 4 years English; Division II: 3 years. [^170]: . Common Core State Standards Initiative, *Common Core State Standards for English Language Arts & Literacy* (NGA Center / CCSSO; official ADA-compliant posting of the 2 June 2010 standards), https://www.corestandards.org/wp-content/uploads/2023/09/ADA-Compliant-ELA-Standards.pdf, accessed 31 August 2026. Language strand L.1–L.6. Language Progressive Skills chart exists; row-by-row grade assignments not re-typed here. Design-limit language: “The Standards define what all students are expected to know and be able to do, not how teachers should teach.” [^171]: . 19 Tex. Admin. Code §§ 110.3, 110.36, 110.38 (Cornell LII copies), accessed 31 August 2026. Conventions inside Composition (editing drafts). Vocabulary as a foundational-language-skills expectation; English I foreign-phrase examples include bona fide, caveat, carte blanche; English III include ad hoc, faux pas, non sequitur, modus operandi. [^172]: . Virginia Board of Education, *English Standards of Learning for Virginia Public Schools*, adopted 28 March 2024. Language Usage K–12; Foundations for Writing spelling K–5. [^173]: . Florida Department of Education, *B.E.S.T. Standards for English Language Arts* (2020/2021 CPALMS posting). Following Conventions (C.3); Vocabulary strand K–12. Conventions Progression by Grade Level (Appendix C) not transcribed as a full table here. [^174]: . College Board, digital SAT Reading and Writing specifications, accessed 31 August 2026. Approximate domain shares: Information and Ideas ~26 percent; Craft and Structure ~28 percent; Expression of Ideas ~20 percent; Standard English Conventions ~26 percent. 54 questions, 64 minutes. College and Career Readiness Benchmark for Reading and Writing: 480. [^175]: . ACT English test description and enhanced-format reporting categories, accessed 31 August 2026. Production of Writing 38–43 percent; Knowledge of Language 18–23 percent; Conventions of Standard English 38–43 percent. English College Readiness Benchmark: 18. The SAT and ACT benchmarks are not equated to each other. [^176]: . Steve Graham and Dolores Perin, “A Meta-Analysis of Writing Instruction for Adolescent Students,” *Journal of Educational Psychology* 99, no. 3 (2007): 445–476. Sentence combining 0.50; grammar instruction −0.32; strategy instruction 0.82. Grammar = explicit teaching of parts of speech and sentence types, often versus a treatment that itself taught writing. Averages across heterogeneous studies; not a promise for every home. [^177]: . Steve Graham, Debra McKeown, Sharlene Kiuhara, and Karen R. Harris, “A Meta-Analysis of Writing Instruction for Students in the Elementary Grades,” *Journal of Educational Psychology* 104, no. 4 (2012). Transcription skills ES = 0.55; grammar instruction not a statistically significant positive effect; strategy instruction 1.02. Abstract coefficients; full PDF not opened. Keep 2007 and 2012 as two metas. [^178]: . Steve Graham et al., *Teaching Elementary School Students to Be Effective Writers*, WWC practice guide (Institute of Education Sciences, 2012, revised 2018), Recommendation 3. [^179]: . Richard L. Venezky, “English Orthography: Its Graphical Structure and Its Relation to Sound,” *Reading Research Quarterly* 2, no. 3 (1967): 75–105. Relational units map first to a morphophonemic level, then to sound. Richard L. Venezky, *The American Way of Spelling* (New York: Guilford, 1999), the book-length statement; not opened page-by-page here. [^180]: . Louisa C. Moats, “How Spelling Supports Reading,” *American Educator*, Winter 2005/06, citing Hanna, Hanna, Hodges, and Rudorf (1966): about half of English words spelled from sound-symbol alone; another 34 percent one error (84 percent mostly predictable); about four percent truly irregular. Linguists, Moats writes, describe English spelling as a morphophonological alphabetic system. [^181]: . Catherine E. Snow, Peg Griffin, and M. Susan Burns, *Knowledge to Support the Teaching of Reading* (2005), quoted by Moats (2005/06). [^182]: . National Reading Panel, *Teaching Children to Read*, NIH Pub. No. 00-4769 (2000), vocabulary findings as summarized in Timothy Shanahan, *The National Reading Panel Report: Practical Advice for Teachers* (Learning Point Associates, 2006): direct and incidental teaching; pre-teaching words about to appear; copying definitions loses; a few hundred words a year thoroughly. Isabel L. Beck, Margaret G. McKeown, and Linda Kucan, *Bringing Words to Life* (2002), three tiers as Shanahan presents them. [^183]: . Timothy Shanahan and Christopher J. Lonigan, “The National Early Literacy Panel: A Summary of the Process and the Report,” *Educational Researcher* 39, no. 4 (2010): simple vocabulary average r ≈ .33 with later literacy; complex oral language (grammar, defining words, listening comprehension) r = .70 with later reading comprehension. National Early Literacy Panel, *Developing Early Literacy* (2008). [^184]: . National Center for Education Statistics, State Education Practices Table 5.7 (2018), accessed 31 August 2026. Four English credits typical. [^185]: . Graham and Perin (2007), sentence combining 0.50. [^186]: . Moats, “How Spelling Supports Reading” (2005/06): professional-judgment sequence — position patterns and inflections, syllables and affixes, then Latin and Greek morphology. Curriculum proposal consistent with the code, not a new RCT. [^187]: . Moats (2005/06), citing Singer and Bashir (2004): writers who must think too hard about spelling use up resources needed for composition. [^188]: . D. J. Montgomery, G. R. Karlan, and M. Coutinho (2001), as cited by Moats (2005/06): spell checkers caught 30–80 percent of misspellings; identified the target from LD misspellings 53 percent of the time. [^189]: . Beck, McKeown, and Kucan (2002), via Shanahan (2006). Tier-one common oral; tier-three technical; tier-two wide-use, not automatic. [^190]: . Moats (2005/06) on invented spelling receding; *they, went, who, said* should not continue to be invented. IES, B. Foorman et al., *Foundational Skills to Support Reading for Understanding in Kindergarten Through 3rd Grade*, NCEE 2016-4008 (2016), Obstacle 3.1: invented spelling useful in K–1 independent writing; use it less as skills develop. [^191]: . Graham et al., IES writing practice guide (2012/2018); Graham and Perin (2007). Model a sentence; write one together; then they write alone. [^192]: . Nicholas C. Soderstrom and Robert A. Bjork, “Learning Versus Performance,” *Perspectives on Psychological Science* 10, no. 2 (2015). [^193]: . Moats (2005/06), spelling dose as professional judgment, not an RCT and not a national minute law. [^194]: . College Board, digital SAT Reading and Writing specifications, accessed 31 August 2026. Approximate domain shares: Information and Ideas ~26 percent; Craft and Structure ~28 percent; Expression of Ideas ~20 percent; Standard English Conventions ~26 percent. 54 questions, 64 minutes. College and Career Readiness Benchmark for Reading and Writing: 480. [^195]: . International Dyslexia Association, 2002 definition (Board of Directors, 12 November 2002): neurobiological in origin; accurate and/or fluent word recognition; poor spelling and decoding; typically a phonological-language deficit; unexpected given other abilities and effective instruction. This chapter does not diagnose anyone. First move is systematic teaching; specialist when difficulty is unexpected and persistent after that teaching. [^196]: . NCES Table 5.7 (n. 184). [^197]: . NCAA Eligibility Center, *Home School Toolkit 2025–26*. Course name on the worksheet should match the transcript title. Division I: 4 years English; Division II: 3 years. [^198]: . Common Core State Standards Initiative, *Common Core State Standards for English Language Arts & Literacy* (NGA Center / CCSSO; official ADA-compliant posting of the 2 June 2010 standards), https://www.corestandards.org/wp-content/uploads/2023/09/ADA-Compliant-ELA-Standards.pdf, accessed 31 August 2026. W.7 short research from grade 1 through high-school sustained projects; W.8 gather, notes, list of sources (grade 3), paraphrase and list (grade 5), standard citation format by 6–8 / 9–12; W.9 begins in grade 4. Introduction: research standards are prominently included in the Writing strand “because of the centrality of writing to most forms of inquiry.” Design-limit language: “The Standards define what all students are expected to know and be able to do, not how teachers should teach.” [^199]: . 19 Tex. Admin. Code §§ 110.3, 110.4, 110.6, 110.36, 110.38 (Cornell LII copies), accessed 31 August 2026. Inquiry from Grade 1; Grade 2 primary/secondary and cite; Grade 4 bibliography and paraphrasing versus plagiarism. English I: credibility and bias; rhetorical devices (understatement, overstatement) and fallacies such as straw man and red herring; English I also names ad hominem, loaded language, and slippery slope. English III: rhetorical analysis as a composition genre; further fallacies (post hoc, circular reasoning, red herring, assumptions); academic citations; ethical use to avoid plagiarism. [^200]: . Virginia Board of Education, *English Standards of Learning for Virginia Public Schools*, adopted 28 March 2024. Research K–12: topic and question with prompting in K; title and author credit in grade 3; relevance, reliability, and citations in grades 4–5. Grade 6–12 research expectation pages were truncated in the fetch used for this book; this chapter does not invent those rows. [^201]: . Florida Department of Education, *B.E.S.T. Standards for English Language Arts* (2020/2021 CPALMS posting). Communication C.4, researching and using information, K recall through synthesize; grade 11 literary research using primary and secondary sources. [^202]: . Florida B.E.S.T. (n. 200), ELA.6.R.3.4 identify rhetorical appeals; introduction: rhetoric in grade 6 so students understand logos, pathos, and ethos when starting argumentative writing. Devices in 7–8; author’s rhetoric in 9–12. [^203]: . 19 Tex. Admin. Code §§ 110.36 and 110.38 (n. 198). [^204]: . College Board, AP English Language and Composition course page, Effective Fall 2024 CED description, accessed 31 August 2026. Introductory college composition; rhetorical analysis of nonfiction; no prerequisite course. Exam overview: 3 hours 15 minutes, 45 multiple-choice and 3 free-response. Credit by college. Distinct from AP English Literature. [^205]: . College Board, CLEP College Composition, https://clep.collegeboard.org/clep-exams/college-composition, accessed 31 August 2026. Ability to Use Source Materials about 25 percent; Rhetorical Analysis about 25 percent; synthesis essay cites two provided sources. MLA, APA, and Chicago named as in-scope. ACE recommends 6 semester hours at score 50; each institution sets policy. [^206]: . NCAA Eligibility Center, *Home School Toolkit 2025–26*. Sample English 1 description includes novel study, essays, short stories, and a short research paper. Course name on the worksheet should match the transcript title. Division I: 4 years English; Division II: 3 years. CLEP / credit-by-exam is not an NCAA-approved core course. [^207]: . Sam Wineburg and Sarah McGrew, “Lateral Reading and the Nature of Expertise: Reading Less and Learning More When Evaluating Digital Information,” *Teachers College Record* 121, no. 11 (2019). Leave-the-page move: open new tabs to see who else is talking about an unfamiliar site rather than lingering on the site’s own presentation. Sam Wineburg, Joel Breakstone, Mark Smith, Sarah McGrew, and Teresa Ortega, “Civic Online Reasoning: Curriculum Evaluation,” working paper 2019-A2, Stanford History Education Group, 2019, https://purl.stanford.edu/xr124mv4805, accessed 31 August 2026. Named here for the move, not as a critical-thinking course and not as a homeschool RCT. [^208]: . NCAA *Home School Toolkit 2025–26* (n. 205), sample description, not a page quota. [^209]: . CCSS ELA (n. 197), W.9 begins in grade 4. [^210]: . College Board AP Language (n. 203); CLEP College Composition (n. 204); NCAA toolkit (n. 205). Dual enrollment: NCES 2020-125, 82 percent of public schools with grades 9–12 *offered* it in 2017–18 — schools offering, not a homeschool take-rate. [^211]: . National Center for Education Statistics, State Education Practices Table 5.7 (2018), accessed 31 August 2026. Four English credits typical; not a federally named sequence. [^212]: . NCES 2020-125 (n. 209). Dual-enrollment English is an accredited college course; the college transcript is the credit-bearing object. Eligibility is local. [^213]: . Common Core State Standards Initiative, *Common Core State Standards for English Language Arts & Literacy* (NGA Center / CCSSO; official ADA-compliant posting of the 2 June 2010 standards), https://www.corestandards.org/wp-content/uploads/2023/09/ADA-Compliant-ELA-Standards.pdf, accessed 31 August 2026. Speaking and Listening CCR anchors: (1) collaborative discussions; (2) integrate/evaluate diverse media; (3) evaluate a speaker’s point of view, reasoning, evidence, and rhetoric; (4) present information with evidence; (5) digital media in presentations; (6) adapt speech. Design-limit language: “The Standards define what all students are expected to know and be able to do, not how teachers should teach.” [^214]: . 19 Tex. Admin. Code § 110.36 (English I) (Cornell LII copy), accessed 31 August 2026. Presentations using informal, formal, and technical language; eye contact, speaking rate, volume, enunciation, purposeful gestures. TAC Subchapter C electives — Public Speaking I–III, Communication Applications, Oral Interpretation, Debate, Journalism — are additional credits, not the four required English I–IV. [^215]: . Virginia Board of Education, *English Standards of Learning for Virginia Public Schools*, adopted 28 March 2024. Communications & Multimodal Literacies K–12: listening and collaboration; speaking and presentation; integrating multimodal literacies (from grade 2/3); examining media messages (from grade 4). Grade 6–12 detail pages were truncated in the fetch used for this book; this chapter does not invent those rows. [^216]: . Florida Department of Education, *B.E.S.T. Standards for English Language Arts* (2020/2021 CPALMS posting). C.2 oral presentation K–12 (complete sentences in K; volume; nonverbal cues; rhetorical devices in 11–12). C.5 multimedia and technology in communication. [^217]: . National Center for Education Statistics, State Education Practices Table 5.7 (2018), accessed 31 August 2026. Four English credits typical. [^218]: . NCTE / IRA, *Standards for the English Language Arts* (1996), reaffirmed by the NCTE Executive Committee November 2012, https://ncte.org/resources/standards/ncte-ira-standards-for-the-english-language-arts/, accessed 31 August 2026. Twelve statements, “not prescriptions for particular curriculum or instruction,” “interrelated and should be considered as a whole.” Full 1996 book PDF not opened this research pass; wording from the official webpage. Used here as a professional check, not as a phonics sequence or a speech syllabus. [^219]: . Barbara Foorman et al., *Foundational Skills to Support Reading for Understanding in Kindergarten Through 3rd Grade*, NCEE 2016-4008 (Institute of Education Sciences, 2016), Recommendation 1: academic language (inferential language, narrative language, academic vocabulary). Minimal WWC evidence in the post-2000 set; still the panel’s recommendation. Not a reason to skip vocabulary; not a numbered oral-presentation *d*. [^220]: . Philip B. Gough and William E. Tunmer, “Decoding, Reading, and Reading Disability,” *Remedial and Special Education* 7, no. 1 (1986): 6–10. R = D × C; listening (linguistic) comprehension is C. Hyperlexia in that paper is strong decoding with weak comprehension — the ceiling image for this chapter’s ear. [^221]: . CCSS SL anchors (n. 212); Virginia 2024 SOL (n. 214); Florida B.E.S.T. C.5 (n. 215). [^222]: . Foorman et al. (2016), Recommendation 1, narrative language: connectives, noun and verb phrases, pronoun reference. Shanahan and Lonigan, “The National Early Literacy Panel,” *Educational Researcher* 39, no. 4 (2010): complex oral language r = .70 with later reading comprehension, versus simple vocabulary r ≈ .33. [^223]: . 19 Tex. Admin. Code § 110.36 and Subchapter C (n. 213). [^224]: . S. G. Wood, J. H. Moxley, E. L. Tighe, and R. K. Wagner, “Does Use of Text-to-Speech and Related Read-Aloud Tools Improve Reading Comprehension for Students with Reading Disabilities? A Meta-Analysis,” *Journal of Learning Disabilities* 51, no. 1 (2018): average weighted d̄ = 0.35 on comprehension for students with reading disabilities. Compensation and access, not a speech credit. [^225]: . Common Core State Standards Initiative, *Common Core State Standards for English Language Arts & Literacy* (NGA Center / CCSSO; official ADA-compliant posting of the 2 June 2010 standards), https://www.corestandards.org/wp-content/uploads/2023/09/ADA-Compliant-ELA-Standards.pdf, accessed 31 August 2026. Speaking and Listening CCR anchors: (1) collaborative discussions; (2) integrate/evaluate diverse media; (3) evaluate a speaker’s point of view, reasoning, evidence, and rhetoric; (4) present information with evidence; (5) digital media in presentations; (6) adapt speech. Design-limit language: “The Standards define what all students are expected to know and be able to do, not how teachers should teach.” [^226]: . Foorman et al. (2016), Recommendation 1; Shanahan and Lonigan (2010) (n. 221). [^227]: . Wood et al. (2018) (n. 223). Hamsa Bastani et al., “Generative AI without Guardrails Can Harm Learning,” *PNAS* 122, no. 26 (2025): e2422633122 — math RCT; writing and speech are easy to outsource because the model’s native product is fluent language. Mechanism, not a speech trial. [^228]: . Texas electives (n. 213). Dual enrollment: NCES 2020-125, 82 percent of public schools with grades 9–12 *offered* it in 2017–18 — schools offering, not a homeschool take-rate. [^229]: . 19 Tex. Admin. Code (n. 213). NCAA Eligibility Center, *Home School Toolkit 2025–26*: course name on the worksheet should match the transcript title. Sample titles include English 9–12. Division I: 4 years English; Division II: 3 years English. [^230]: . NCES Table 5.7 (n. 216); NCAA toolkit (n. 227). [^231]: . Texas Education Agency, “Home Schooling,” https://tea.texas.gov/families-and-students/finding-school-your-child/home-schooling. “The State of Texas does not award a diploma to students that are home schooled.” [^232]: . New York State Education Department, “Home Instruction Questions and Answers,” https://www.nysed.gov/nonpublic-schools/home-instruction-questions-and-answers. A high school diploma “may only be awarded to a student enrolled in a registered secondary school.” [^233]: . NCAA Eligibility Center, *Home School Toolkit 2025–26*, http://fs.ncaa.org/Docs/eligibility_center/Student_Resources/Home_School_Toolkit.pdf. Proof of graduation is a family- or umbrella-issued artifact. [^234]: . NCAA Eligibility Center, “Core-Course Worksheet Instructions,” http://fs.ncaa.org/Docs/eligibility_center/Student_Resources/Core_Course_Worksheet_Instructions.pdf. Sample titles: English 9–12, Composition I/II. Course name matches the transcript title. [^235]: . National Center for Education Statistics, Table 5.7 (2018), https://nces.ed.gov/programs/statereform/tab5_7.asp; 19 Tex. Admin. Code §§ 110.36–110.39; Virginia Board of Education course list (English 9–12; AP Language / AP Literature may substitute for 11 or 12). [^236]: . NCES, Table 5.7 (2018). Typical requirement 4.00 English credits (California 3.00; Arkansas 4.50; Idaho 4.50). Texas, Florida, Virginia: 4.00. [^237]: . Official department pages illustrate a pattern, not a fifty-state code. This chapter is not legal advice. A membership organization’s chart, including HSLDA, is not a statute. [^238]: . TEA, “Home Schooling.” The Agency “does not regulate, index, monitor, approve, register, or accredit” home-school programs. TEA cites *Leeper et al. v. Arlington ISD et al.* [^239]: . North Carolina Division of Non-Public Education, “Starting a Home School,” https://www.doa.nc.gov/divisions/non-public-education/home-schools/requirements-recommendations (last modified 3 August 2026 when fetched). Annual test must include English grammar, reading, spelling, and mathematics. Five hours a day is a recommendation, not law. [^240]: . Pennsylvania Department of Education, Home Education Program BEC, 24 P.S. § 13-1327.1, https://www.pa.gov/agencies/education/resources/policies-acts-and-laws/basic-education-circulars-becs/purdons-statutes/home-education-program. Affidavit by 1 August; evaluator by 30 June; tests in grades 3, 5, and 8. [^241]: . Virginia Department of Education, “Home Instruction,” https://www.doe.virginia.gov/parents-students/for-parents/home-instruction. Notice by 15 August; evidence of progress by 1 August; SAT/ACT/PSAT as one test-option equivalent. [^242]: . NYSED, “Home Instruction Questions and Answers.” IHIP, quarterlies, annual assessment; no local or Regents diploma from home instruction. [^243]: . TEA, “Home Schooling”; NYSED Q&A; North Carolina DNPE, “Starting a Home School.” [^244]: . College Board, SAT structure and Reading and Writing specifications; “SAT Suite College and Career Readiness Benchmarks,” https://satsuite.collegeboard.org/k12-educators/about/understand-scores-benchmarks/benchmarks. Benchmark 480. [^245]: . ACT, exam sections and structure; “ACT College Readiness Benchmarks,” https://www.act.org/content/act/en/college-and-career-readiness/benchmarks.html. Enhanced English 50 questions / 35 minutes. Benchmark 18. [^246]: . College Board AP Central, AP English Language and AP English Literature course pages. No prerequisite course; credit by college. Homeschool AP Course Audit was not treated as settled; safer practice is a non-AP title plus the exam score. [^247]: . College Board CLEP, “College Composition,” https://clep.collegeboard.org/clep-exams/college-composition. ACE: score 50, 6 semester hours. NCAA *Home School Toolkit 2025–26*: CLEP is not NCAA core. [^248]: . NCES 2020-125: 82 percent of public schools with grades 9–12 *offered* dual enrollment (2017–18). NCES 2019-176 (about a third of HSLS:09 *took* such courses) is a different object, not a homeschool rate. [^249]: . NCAA Division I and Division II initial-eligibility pages. [^250]: . NCAA homeschool-student page; *Home School Toolkit 2025–26*. Evaluation after documents are in and an NCAA school has placed the student on an Institutional Request List. [^251]: . NYSED Q&A: instruction “should be geared to the level appropriate to the student’s needs and previous level of achievement.” [^252]: . All About Learning Press, All About Reading and All About Spelling product pages, https://www.allaboutlearningpress.com/all-about-reading/ and https://www.allaboutlearningpress.com/all-about-spelling/. Scripted Orton-Gillingham-approach claim; four reading levels plus Pre-reading; seven spelling levels; “20 minutes a day”; placement tests. Company copy, not an independent efficacy study. [^253]: . Logic of English, “About Foundations,” https://support.logicofenglish.com/hc/en-us/articles/4575255485211-About-Foundations; “About Essentials,” https://support.logicofenglish.com/hc/en-us/articles/4600606428699-About-Essentials; “Foundations and Essentials Comparison.” 75 phonograms; Foundations not a prerequisite for Essentials; “98 percent of English words” is the company’s claim (also the related book *Uncovering the Logic of English*). [^254]: . Institute for Excellence in Writing, “Our Methods,” https://iew.com/about/our-methods. Nine structural models; style dress-ups; *Teaching Writing: Structure and Style*. [^255]: . Susan Wise Bauer and Jessie Wise, *The Well-Trained Mind: Essential Edition* (New York: W.W. Norton, 2024), product page https://welltrainedmind.com/p/wtm-ee/; Well-Trained Mind support, “Where to Start in All Language Arts,” https://support.welltrainedmind.com/hc/en-us/articles/360011380674-Where-to-Start-in-All-Language-Arts. Writing With Skill starts at Level 1. [^256]: . Brave Writer, “Original Thought: Partnership Writing,” https://help.bravewriter.com/knowledge/original-thought-partnership-writing. “Don’t be governed by age range” is the company’s wording. [^257]: . Adam and Missy Andrews, Teaching the Classics Basic Seminar, Center for Lit, https://www.centerforlit.com/product/teaching-the-classics-basic-seminar/. Eight-hour seminar; Socratic List of 173 questions; Second Edition 2017. [^258]: . AmblesideOnline Advisory, “Curriculum Overview,” https://amblesideonline.org/curriculum; “An Introduction to Charlotte Mason,” https://www.amblesideonline.org/cm-intro; “Language Arts,” https://amblesideonline.org/language-arts. Free community curriculum; family chooses early reading instruction; short lessons 10–20 minutes for younger children are community/historical, not a statute. [^259]: . Houghton Mifflin Harcourt, *Into Literature* program page, https://www.hmhco.com/programs/into-literature; “Into Literature: Research Evidence Base,” https://www.hmhco.com/research/into-literature-research-evidence-base. Grades 6–12; chronological American literature at 11 and British literature at 12, per the company. [^260]: . Public-domain collections families actually use include Project Gutenberg (https://www.gutenberg.org), the Internet Archive (https://archive.org), Standard Ebooks (https://standardebooks.org), and Library of Congress digital collections. AmblesideOnline booklists often link free editions. This note names libraries of expired-copyright texts, not a curriculum. [^261]: . International Dyslexia Association, “Definition of Dyslexia” (Board of Directors, 12 November 2002), https://dyslexiaida.org/definition-consensus-project/. Unexpected given other abilities and effective instruction. This chapter does not diagnose. [^262]: . Philip B. Gough and William E. Tunmer, “Decoding, Reading, and Reading Disability,” *Remedial and Special Education* 7, no. 1 (1986): 6–10. [^263]: . Barbara Foorman et al., *Foundational Skills to Support Reading for Understanding in Kindergarten Through 3rd Grade*, WWC Practice Guide (Institute of Education Sciences, 2016; revised 2019), Recommendation 4 and Obstacle 2.2, https://ies.ed.gov/ncee/wwc/practiceguide/21. [^264]: . Sarah G. Wood, Jerad H. Moxley, Elizabeth L. Tighe, and Richard K. Wagner, “Does Use of Text-to-Speech and Related Read-Aloud Tools Improve Reading Comprehension for Students with Reading Disabilities? A Meta-Analysis,” *Journal of Learning Disabilities* 51, no. 1 (2018): 73–84, https://pmc.ncbi.nlm.nih.gov/articles/PMC5494021/. Average weighted d̄ = 0.35 on comprehension for students with reading disabilities — access, not decoding instruction.