The American reading debate has acquired a familiar rhythm. A school district announces a new literacy curriculum. Advocates celebrate a victory for science. Critics warn that reading is being reduced to drills. Parents wonder whether their children will finally learn to read. And somewhere inside the classroom, a student successfully sounds out a sentence without understanding why it matters.
That student is the missing person in the argument.
For years, the public dispute has been presented as a contest between phonics and meaning: between teaching children the code of written language and inviting them into the world of books. It is a false choice. Children need to recognize words accurately, and they need to understand them. They need explicit teaching, sustained practice, language, knowledge, and reasons to read. A reform that restores decoding but neglects comprehension is incomplete. A curriculum that celebrates rich literature while leaving children to guess unfamiliar words is no better.
The science of reading is neither a trademark nor a single teaching program. It is a developing body of research about how readers learn, why some struggle, and what instruction can do about it. The question is not whether phonics belongs in literacy education. It does. The harder question is what a school must build around phonics to produce confident, independent, thoughtful readers.
A reading crisis cannot be solved by changing the slogan
The 2024 National Assessment of Educational Progress found that the average reading scores of American fourth- and eighth-graders had fallen by two points compared with 2022. At fourth grade, the average was five points lower than in 2019. These figures describe a serious national problem; they do not, by themselves, identify a single cause or prove that any particular curriculum reform works. NAEP measures reading comprehension across a complex educational system, not the effectiveness of one phonics lesson. Source: National Center for Education Statistics.
Behind those aggregate numbers are different children with different obstacles. One struggles to connect letters to sounds. Another can decode slowly but cannot maintain fluency. A third reads words smoothly and cannot explain the paragraph. A fourth has strong oral understanding but is still learning the vocabulary of English in school. A fifth has had inconsistent instruction or too few opportunities to read demanding texts.
Treating all five as instances of the same problem is not scientific. Nor is offering all five the same intervention.
The American policy conversation is slowly making room for this more complicated reality. A June 2026 review by the National Council on Teacher Quality reported improvements in the proportion of teacher-preparation programs meeting its criteria for evidence-aligned reading instruction, while finding that many programs still fell short on one or more components. That is an evaluation conducted using NCTQ's own rubric, not a direct measurement of children's learning. It is nevertheless a reminder of how difficult it is to translate findings into teacher preparation and everyday practice. Source: NCTQ, 2026.
First, the code: why phonics matters
Imagine asking a child to read the invented word mip. There is no picture to recognize and no familiar story to guess from. The reader must understand that written letters represent speech sounds and that those sounds can be blended into a pronunciation. That is the principle at the center of alphabetic reading.
Phonemic awareness concerns the sounds in spoken language; phonics concerns how those sounds relate to written symbols. They are connected but not identical. An instructional sequence can explicitly introduce sound–letter relationships, help a child blend and segment sounds, and give the child opportunities to decode unfamiliar words. This reduces dependence on memorizing isolated word shapes or using context to guess.
The influential 2018 review Ending the Reading Wars: Reading Acquisition From Novice to Expert, by Anne Castles, Kathleen Rastle and Kate Nation, argues that alphabetic knowledge is foundational for English reading. Its significance lies in what follows that claim: learning to read also involves developing efficient word recognition and text comprehension. Source: Castles, Rastle & Nation.
U.S. federal instructional guidance makes a similarly broad point. The What Works Clearinghouse recommends teaching children in kindergarten through third grade to notice sounds and letters, decode and analyze words, develop academic language, and read connected text every day. Word reading is a foundation for understanding, not a replacement for it. Source: IES Foundational Skills Guide.
The case for explicit attention to the alphabetic code, however, should not be mistaken for proof that any product advertised as “science of reading” is effective, that all programs labeled systematic phonics perform equally well, or that phonics deserves every minute of a literacy block.
The danger of winning the wrong argument
It is possible to win a dispute about instructional materials while losing sight of the learner.
A child may pronounce each word correctly and still be unable to say what a sentence means. A student may perform well on short decoding exercises yet struggle to follow an argument across three pages. An older reader may identify familiar words automatically but stumble over multisyllabic academic language.
These are not objections to phonics; they are reasons to take the wider science seriously.
The classic Simple View of Reading describes reading comprehension as depending on both decoding and language comprehension. Its apparent simplicity is its strength. If a student cannot identify the words on the page, understanding suffers. If the student identifies every word but does not understand the language, comprehension also suffers. The model is a powerful reminder that neither part can substitute for the other.
But no concise formula describes every process involved in reading. In 2021, literacy researchers Nell K. Duke and Kelly B. Cartwright proposed an active view of reading that highlights overlapping abilities and processes such as self-regulation, attention to meaning and flexible strategy use. Their argument was not that decoding had ceased to matter; it was that research had moved forward and teacher education should move with it. Source: Duke & Cartwright, 2021.
Consider the word migration. A reader may decode the printed word accurately. Understanding an article about bird migration, however, requires knowledge of movement, seasons, geography, causes, and perhaps climate. A classroom that teaches the pronunciation but never builds these concepts has not finished teaching reading.
Vocabulary is not a decorative addition
Students cannot comprehend concepts they have never encountered merely because they can pronounce the relevant terms. Vocabulary grows through deliberate instruction, conversation, reading, and exposure to rich ideas. It also grows in subject lessons. Science, history, literature and the arts are not competitors with literacy; they are among its sources.
This is a particular challenge for a narrowly organized reading curriculum. If comprehension lessons consist mostly of isolated passages and repeated generic exercises, students may practice answering questions without accumulating the knowledge that makes later texts more accessible.
Good instruction allows students to examine demanding ideas through read-alouds and discussion even before their independent decoding catches up. It does not require children to wait until they can read every word themselves before encountering intellectually ambitious content.
Fluency is a bridge, not a stopwatch
A slow reader may devote so much attention to word identification that little capacity remains for following meaning. As recognition becomes more accurate and automatic, a reader can attend more fully to the text.
Yet fluency is not simply speed. It includes accurate, appropriately paced reading and, in oral reading, expression that can reflect meaning. Racing through a passage is not evidence that it was understood. Repeated reading with useful feedback can help, but the purpose is effortless access to meaning rather than a faster performance on a timer.
Comprehension requires intellectual work
To understand a demanding paragraph, readers identify relationships, make inferences, connect new claims with prior knowledge, notice contradictions and sometimes recognize that they have not understood. These are not automatic consequences of sounding out words.
The What Works Clearinghouse guidance for grades four through nine recommends combining decoding support when needed with fluency work, word and world knowledge, questioning, gist-making, comprehension monitoring, and opportunities to tackle challenging texts. Source: IES Grades 4–9 Practice Guide.
The implication is straightforward: students who still need decoding help should receive it. Students who can decode accurately may need something different. Some need both.
The cueing controversy: what exactly is being debated?
Few disputes expose the weaknesses of educational slogans as clearly as the argument about three-cueing.
In one version of the practice, a beginning reader encountering an unfamiliar printed word is encouraged to infer it from the picture, the topic or the first letter rather than examining the complete spelling. This can reward a plausible guess instead of teaching a transferable word-reading skill. For alphabetic word identification, accurate attention to print is indispensable.
But the word context is doing too much work in this dispute. Using a picture to guess an unrecognized word is not the same as using context to infer the meaning of a word that has already been read accurately. Asking whether a sentence makes sense is not necessarily the same as telling a child to skip its letters.
The research discussion itself has not been entirely uniform. A 2020 review by Jeffrey S. Bowers disputed strong claims that systematic phonics is superior to all alternative approaches across the outcomes and comparisons included in the evidence base, while explicitly refusing to treat that as an argument for a return to whole language. A 2026 review by Maria Nicholas and Ondine Bradbury argued for distinguishing three-cueing as comprehension and accuracy monitoring from its use as a method of decoding. These are contested interpretations, not universal endorsements of cueing programs. Source: Bowers, 2020 · Source: Nicholas & Bradbury, 2026.
A serious literacy debate can acknowledge such disagreements without losing a practical principle: children should learn reliable ways to identify the printed words in front of them, while also learning to ask what those words mean within a text.
The objective is not to rescue an old label. It is to stop allowing vague labels to conceal what teachers are actually being asked to do.
The biggest blind spot: reading is also an inequality question
Advice about literacy often assumes that children arrive at school with broadly similar opportunities to develop language. They do not.
Home access to books, the language of instruction, opportunities for conversation, stable attendance, hearing or vision difficulties, instructional quality and the availability of timely support can all shape a child's reading experience. These factors interact; none provides a simple explanation of an individual student's ability.
A strong literacy program therefore cannot rely on the assumption that children will acquire background knowledge somewhere else. The classroom must deliberately build it. It must also identify struggling readers early and provide responsive interventions without treating them as less intelligent or less capable of sophisticated ideas.
This matters especially for students learning English while mastering literacy. A child can have rich knowledge in another language and still lack the English words to demonstrate it on a particular reading test. Teaching the English language of a text and recognizing existing knowledge should go together.
Equity also requires something less glamorous than a curriculum announcement: enough trained adults, suitable materials, protected time and trustworthy assessment to discover whether a child is improving. School systems should be cautious about interpreting a single benchmark score as a complete diagnosis.
What a comprehensive reading lesson might actually look like
Picture a third-grade class studying how rivers reshape land. Before children open the text, the teacher shows how water moves sediment and introduces words such as erosion, bank and deposit. Students talk about the phenomenon and use the vocabulary in full sentences.
Then comes the word-reading work. For readers who need it, the teacher explicitly models a relevant spelling pattern or a multisyllabic word and checks that students can apply it. Children read connected text rather than stopping at a list of words. The teacher listens for accuracy and fluency.
Afterward, students return to the meaning. They explain a process in sequence, justify an inference with evidence from the passage, compare two examples, and ask a question the text has not answered. A student who decodes successfully but cannot explain the central idea receives support with language and concepts. A student who understands the science orally but misreads crucial words receives targeted word-reading instruction.
No single ten-minute activity carries the entire burden. The sequence integrates decoding, language, knowledge, practice and thinking.
This is an illustrative classroom design, not an experimental intervention whose effectiveness has been tested as a whole. Its value is to reveal the decision-making required when evidence is converted into teaching.
Five tests for schools promising evidence-based literacy
Instead of asking whether a school has purchased a “science of reading” curriculum, families and educators can ask more revealing questions.
Can students decode words they have not memorized? Examine whether instruction develops transferable sound–spelling knowledge and whether teachers can identify students who need more help.
Can students explain what they read? Ask for evidence of vocabulary development, discussion, meaningful writing and comprehension across science, history, literature and everyday information.
Does assessment lead to different kinds of support? A reading score is useful only if it helps teachers understand whether the next step should involve word recognition, language comprehension, fluency, or a combination.
Do teachers have the knowledge and time to adapt? A curriculum cannot substitute for professionals who understand reading development, recognize difficulty and respond thoughtfully.
Are students growing into readers who choose, question and investigate? Literacy is not completed when a child passes a benchmark. It is expressed in the ability to learn from texts, weigh claims, enjoy literature and participate in civic and intellectual life.
These questions do not privilege entertainment over discipline. They ask whether discipline has achieved its purpose.
The case against a phonics-only reform—and the case against its opposite
Critics are right to worry that an overly narrow response to the reading crisis could starve children of stories, writing, discussion and general knowledge. A timetable dominated by isolated code exercises may fail to cultivate the broader abilities required for sustained comprehension.
But that critique is not a defense of leaving word recognition to chance. Rich literature does not absolve a school from teaching children how the writing system works. Enjoying stories and learning to decode are compatible aims.
Advocates of systematic decoding instruction are right to insist that children should not be required to infer every word from context. They should also resist treating every question about implementation, comprehension or evidence quality as hostility toward science. A scientific field gains authority through openness to improved models, careful comparisons and honest uncertainty—not by settling permanently on a campaign phrase.
The binary itself is the trap: one side fears a world of mechanical word callers; the other fears a world of children who are taught to guess. Both are describing failures worth preventing.
The reading wars should end where reading begins: with the learner
There is a difference between helping a child say a word and helping that child enter the meaning of a text. The first is an achievement. The second depends on the first, but reaches further.
Real literacy gives children access to stories they have never heard, evidence they can judge, disciplines they can study, and arguments they can challenge. To get there, schools need more than enthusiasm for books and more than a phonics checklist. They need a coherent learning architecture: explicit foundations, language-rich teaching, knowledge across subjects, diagnostic support and increasingly independent reading.
The new reading wars will have accomplished something only if they make themselves unnecessary. The winning side should not be a camp, a company or a slogan. It should be the child who can read a new word accurately, understand a difficult idea and decide what to think next.
