DEEP READ · 102
Edward W. Said · 1978
Europe spent two centuries building an enormous library about "the East" — grammars, histories, travelogues, portfolios of engravings, intelligence briefs — and what Said did was read that library as a specimen rather than use it as knowledge, which turned up something deflating: the "Orient" in these books agrees with itself to a remarkable degree and corresponds to the actual places only loosely; what it describes is not the East but what Europe needed the East to be. And once that description was handed to proconsuls, staff officers and diplomats, it stopped being description at all — it became a map you could govern by.
Said (1935–2003) was born in Jerusalem to a Palestinian Christian family, grew up in Cairo, and became professor of English and comparative literature at Columbia — a literary critic trained on the Western canon, not a Middle East specialist. That is exactly the leverage of the book: he does not attack Western scholarship from outside; he works inside it, using its own craft of close reading on its own texts about other people. Published in 1978, it more or less single-handedly produced the field that came to be called postcolonial studies; and "Orientalism," which had been the respectable name of a scholarly discipline, was converted by this one book into an accusation — a word bent out of shape by a single volume, which is rare in the history of scholarship.
The whole book circles three sentences:
Said opens by splitting the word into three senses stacked on top of each other. They have to be read together; taken separately they produce the standard misreadings of him.
The first is a discipline: anyone who researches, teaches or writes about "the Orient" — philologists (scholars of the history of languages and the transmission of texts), anthropologists, historians, archaeologists — is doing Orientalism. This layer is neutral and respectable: chairs, learned societies, journals. The second is a style of thought: a habit built on an ontological line between East and West. Anyone who starts from the premise that the world divides into those two halves, and that the inhabitants of each differ in their nature, is inside this layer — poets, politicians, and lovers of the East included. The third, and the one he actually means, is an institution: Orientalism is the West's whole way of dealing with the Orient, which it does by continually making statements about it, authorizing certain views of it, describing it, teaching it, settling it, ruling over it. His much-quoted formulation: Orientalism as a Western style for dominating, restructuring and having authority over the Orient.
Stack the three and the deflating conclusion appears: here is an apparently pure discipline whose object of study it manufactured itself. This is not a charge of lying — an Arabic grammar can be correct in every particular — but a claim about the category: "the Orient" (everything from Morocco to Japan in one basket, with the inhabitants of the basket assumed to share a nature) is a product, not a discovery. Said traces the manufacturing a long way back. In Aeschylus's The Persians (472 BCE), Asia is already a defeated, mourning, feminized space whose grief is spoken for it by Greeks; Dante's arrangement is tidier still, with Muhammad consigned to the schismatics in Hell while Avicenna, Averroes — the two great philosophers of the Islamic world — and Saladin are filed in Limbo. The East is sorted, graded, assigned its place — and never permitted to speak.
How it changes what you see: watch any sentence whose subject is a continent, a civilization or a religion. The bigger the subject, the more likely the basket was built rather than found. You cannot say "Europeans are such-and-such" without sounding absurd, yet "Orientals are such-and-such" scans as a normal sentence; that asymmetry is the evidence that the basket is still working — and once a basket is accepted, every subsequent argument about right and wrong is already being fought on its ground.
Said takes "discourse" from Foucault (Michel Foucault, the French philosopher who argued that knowledge and power generate each other). In ordinary usage, "discourse" means what people say; in Foucault's sense a discourse is a set of invisible rules determining which sentences can be uttered in a field, who is licensed to utter them, and what kind of statement will be taken as true. An analogy: discourse is not what the players say, it is the touchlines and the rules — you can play well or badly, but you are always playing the game this pitch allows.
So the point is not that a given individual is a racist; it is that the discourse converts whatever enters it into the same output. A scholar who despises Arabs and a scholar who adores them produce work that is structurally near-identical at the base: both are delivering summary findings on behalf of people who are silent. Said departs from Foucault in one consequential way: Foucault's discourse is nearly authorless and turns over anonymously, while Said insists on keeping the determining imprint of individual writers, and insists on pinning discourse to actual empires and actual armies. That deviation later became his critics' best target.
What keeps the discourse running? Here he offers a concrete mechanism, the textual attitude: faced with a reality that is unfamiliar, disorderly and frightening, people would rather trust a book that has tidied reality up than trust their own eyes. His example: a man who has read a manual on handling fierce beasts and then meets a lion whose behaviour does not match the manual tends not to revise toward what he saw, but to look for a more authoritative book — the book's failure is charged to the reader, so the authority of books increases. This is how the Orientalist library grew: it is citationary, texts citing texts and texts vouching for texts, an edifice that barely needs to touch the ground. How it changes what you see: the confidence you feel about places you have never been and people you have never lived among is almost certainly made of text — and the strength of that confidence has little to do with how much evidence there is, and a great deal to do with how many mutually citing sources you have read.
This is the most usable tool in the book and the most frequently skipped chapter. Said cuts the discourse into two layers. Manifest Orientalism is the stated views, the factual judgements, the policy recommendations — a layer that varies enormously between writers, can be updated, and can be argued down. Latent Orientalism is the floor almost nobody states and almost nobody questions: that the Orient is a single bloc distinct from the West, essentially different, static, sensual, passive, and incapable of setting itself in motion.
His observation: over two centuries, nearly all the violent change happened in the manifest layer, while the latent layer barely moved. From Chateaubriand, the French Romantic who went east looking for a spiritual homeland in the early nineteenth century, to the strategic memoranda of the 1970s, the tone, the loyalties and the sympathies are all different — and the floor ("one bloc, inert, requiring outside force") is still there. This explains something that otherwise puzzles people: the most devoted lover of the East is often the most complete Orientalist, because what he loves is the unchanging, mysterious East that can cure Europe's modern sickness — which makes him the last person who wants it to become an ordinary place with railway budgets and party politics.
| Latent | Manifest | |
| What it is | Unstated floor assumptions | Stated views and policies |
| Content | One bloc, alien, static, sensual, passive | Particular histories, verdicts, prescriptions |
| Stability | Barely moves in two centuries | Swings with era, school, temperament |
| Therefore | Enemies and admirers share one floor | They refute each other here, floor untouched |
Schematic: the argument almost always happens in the manifest layer; the layer doing the work is the one nobody touches.
How it changes what you see: next time you watch an argument about some group, don't pick a side first. Ask what unquestioned premise both sides share — that is usually the thing actually doing work, and the two loudest parties are reinforcing it together.
Orientalist writing has one extremely stable grammatical tic: it hardly uses past or future tenses, only an eternal present. It does not say "a particular person in a particular place did a particular thing"; it says "the Oriental is…". Individuals disappear and specimens remain — every Oriental encountered becomes a sample of the Oriental, and a sample cannot refute a type, only confirm it. This is essentializing: writing a historical situation down as a nature.
Here Said has his hardest evidence, because he is quoting the rulers themselves. Arthur Balfour, defending continued British rule in Egypt before the Commons in 1910, is the specimen: we know Egyptian civilization better than anyone else knows it, and we know it further back (paraphrased) — therefore we are entitled to govern it. Look at the gears of that inference: "we understand it" is doing duty as the premise for "we may dispose of it," and the opinion of the party understood does not appear in the argument at all, because on this body of knowledge's own account they lack the kind of mind self-government requires. Blunter still is Egypt's actual ruler, Lord Cromer: in Modern Egypt (1908) he writes the Oriental as the European's negative — illogical, inaccurate, with an inaccuracy that "easily degenerates into untruthfulness," and that, he says, is the chief characteristic of the Oriental mind (paraphrased). What makes the passage vicious is not its contempt but its structure: it swaps a checkable accusation (this man lied) for an unanswerable identity (these people are untruthful by nature) — and there is no evidence with which you can clear a nature.
How it changes what you see: watch the tense and the subject. When a description slides from "someone did something under some condition" to "that is what they are like," a falsifiable statement has become an unfalsifiable verdict — and that slide tends to occur precisely where a reason is needed for some unequal arrangement.
Said's strongest chapter is not about literature; it is about a fleet. When Napoleon invaded Egypt in 1798 he brought some 160 savants — surveyors, naturalists, physicians, decipherers — whose product was the twenty-three-volume Description de l'Égypte (published 1809–1828). This is the book's paradigm: the scholarly apparatus and the army set out together; Egypt is first transcribed into a consultable text and only then treated as administrable territory. The next two steps fit just as tightly: Silvestre de Sacy in Paris turned Orientalism into a modern discipline with a method, students and a chair; Ernest Renan, in his general history of the Semitic languages (1855), wrote "Semitic" as an inert linguistic-racial category — a philological bracket turned overnight into a verdict on a people's ceiling. Then in 1869 the Suez Canal opened and de Lesseps cut the isthmus: the East was no longer merely rewritten but operated on.
The mechanism is a loop, not a line:
Schematic: scholarship yields authority, authority licenses rule, rule produces fresh archive — so the knowledge gets ever more "accurate": it has produced the very Orient it describes.
The shape matters. It explains how Orientalism can look simultaneously more and more learned and never corrected by reality: its input is wired to its own output. British-run Egypt generated censuses, legal files and cadastral registers organized by British categories; the next generation of scholars wrote books from that material; the books underwrote the next round of administration. How it changes what you see: any body of knowledge that controls the questions, the sampling and the interpretation at once is guaranteed to validate itself — a sentence that holds unchanged today for customer segmentation, credit scoring and risk models, with only the nouns replaced.
Said spends considerable space on Flaubert's notes from his 1850 journey through Egypt. There Flaubert met Kuchuk Hanem, a dancer and courtesan, and the resulting pages became one of the most famous Oriental scenes in nineteenth-century European literature. The point Said makes is not that Flaubert is lewd or contemptuous — the writing is genuinely moving — but a structural fact: she never once speaks of herself. Her feelings, her situation and her history are all uttered on her behalf by Flaubert, and that version was read by thousands in Europe while her own does not exist and nobody wanted it.
The structure extends far past one man and one woman. The Orient as a whole is imagined as feminine: penetrable, silent, charged with sexual promise, available as an escape for anyone bored of European bourgeois life. For the political meaning of that imagining, Said nails Marx's line from The Eighteenth Brumaire to the book as an epigraph: they cannot represent themselves; they must be represented. (Marx was writing about French smallholders; Said transplants the sentence as Orientalism's epitaph — and the transplant is itself the argument: the position of the represented is portable, and anyone can be put in it.) The central power is not saying unpleasant things; it is speaking on behalf of. Once a group can appear only as the object of description, it has lost the contest over what it is.
How it changes what you see: with any reportage, study or documentary about a group of people, run one mechanical check — are these people speakers, or are they the thing spoken about? A quotation marshalled to prove the author's point does not count as speech; a remark that could have changed the author's conclusion does. The check almost never fails, and it regularly exposes the most sympathetic work.
The book's third part follows the discourse as it moves house after the Second World War: the centre shifts from London and Paris to Washington, and the personnel shift from philologists with classical languages to area studies experts — social-science training, no need for the old manuscripts, writing for foundations and government departments. Said's judgement: the methods changed completely and the latent layer was carried across intact, now dressed as "policy relevance." It no longer says what Orientals are by nature; it says "the Arab mind," "patterns of behaviour in the Islamic world" — still one bloc, still summed up by outside experts, still underwriting some arrangement. He names the popular national-character literature of the day (Raphael Patai's The Arab Mind, 1973) as a late specimen of the same bloodline. In the popular register it compresses to a handful of ready images: the oil sheikh, the terrorist, the veil — the subject of his Covering Islam (1981), whose argument is that after the Iranian revolution "Islam" became a master key for Western media, and a word that explains everything explains nothing: it simply files the politics and history that needed analysis under a people's nature.
How it changes what you see: Said's observation is that in American public life the Arab appears mainly when he constitutes a problem — an oil-price variable, a security threat, background crowd on the news. So watch for the groups that only come on stage when they are trouble: the narrower the occasions on which a group is admitted, the harder its stereotype is to shift — not because anyone is falsifying anything, but because the sampling itself is tilted, and tilted sampling yields conclusions that look perfectly honest.
Here is the whole thick book's one line of argument, said once:
Premise: European writing about "the East" is vast, spans more than two centuries, and crosses every discipline and genre. Observation: these texts agree with one another far more closely than they correspond to the regions described; change the country, the century, the political sympathy, and the shape of "the Orient" stays much the same. Diagnosis: therefore it cannot be description; it can only be a discourse — citationary, self-feeding, institutionally licensed as to what counts as true, converting particular encounters into permanent types. The demonstration: Napoleon's embedded savants and the Description de l'Égypte → de Sacy founding the discipline and Renan casting "Semitic" as an inert category → Balfour and Cromer wiring "we understand them" directly to "we are entitled to rule them" → postwar American area studies and think tanks → the media Arab who appears only when something goes wrong. Conclusion: here knowledge and power do not merely influence each other, they are each other's conditions; pure scholarship had political effects and political preconditions alike.
Two denials in that line are the parts most often dropped. First, he does not claim that behind these representations there is a "real Orient" waiting to be restored — he says explicitly that he has no interest in positing one. Second, he does not claim Europeans cannot know the East. He claims one thing: "the Orient" as an object of knowledge was produced inside a grossly asymmetrical situation of power, and that situation has to enter any assessment of the knowledge. Remove the two denials and the book collapses into a nationalist slogan — which is, in many hands since, exactly how it has collapsed.
Misreading 1: "Said says Westerners have no standing to study the East." The reverse. He was himself a critic trained on the Western canon, argued all his life that criticism is a shared cross-cultural enterprise, and publicly opposed turning culture into a question of descent — the identity gatekeeping conducted in his name was one of his standing late complaints.
Misreading 2: "He says the Orientalists' facts were wrong," or "Orientalism is just racism." Neither. His claim concerns structure and position, not truth value: the problem is that accurate facts were loaded into a frame ("one bloc, incapable of moving itself") and used to support an arrangement. Nor is it identical with racism — it also includes admiration and romanticization, and those benign versions are the hardest to detect, because they too require the other side to stay as it is.
Misreading 3: "The book is about Asia." His material is almost entirely British and French writing on the Arab-Islamic Near East; India, China, Japan and Southeast Asia are nearly absent, yet the claims are stated universally. That is not only a misreading but a flaw in the book (below).
The strongest objections, stated honestly:
Read only Orientalism and you get a lopsided Said: a man who only takes things apart. The most important pieces are elsewhere.
Culture and Imperialism (1993): contrapuntal reading. This book answers "where are the colonized?" directly, and supplies his most usable method — contrapuntal reading (from counterpoint in music, where several lines run at once and constrain each other with no single governing melody): read a metropolitan work while also listening for the colonial line that sustains it and has been muted almost to silence. The famous instance is Austen's Mansfield Park — the order and leisure of that English country house are paid for by Sir Thomas Bertram's sugar plantation in Antigua, which Austen mentions in a handful of lines and never revisits. The point is not to convict Austen but this: those few lines are the foundation of the whole house, and criticism's job is to hear the foundation and the drawing room in one field of attention. The book's second half is itself the counterweight to the first: it deals at length with cultures of resistance — Yeats, Fanon, C. L. R. James, Césaire — where the colonized finally speak.
The Question of Palestine (1979) and the permission to narrate. This is where you see why representation mattered to him so much. His well-known 1984 essay in the London Review of Books turns on the permission to narrate: a people without a narrative framework the international public will accept finds that its facts are not believed — lacking narrative permission, it has to win the standing to present evidence before it can present any. He defended Palestinian political rights all his life, and also broke with the Palestinian Authority over his public opposition to the Oslo Accords, his books briefly banned under it — he was no gentler with power on his own side.
Representations of the Intellectual (the 1994 BBC Reith Lectures): exile and amateur. His rulebook has two counterintuitive entries. Be an exile — not necessarily by leaving, but by refusing to belong comfortably to any camp, because belonging quietly buys your judgement. Be an amateur — he treats professionalism as the primary corrosion: the expert's rewards (grants, access, standing) are issued by institutions, and the price of collecting them is to ask only technical questions and never whether the thing should be done at all. Speaking truth to power is the job, and professionals are structurally unable to do it.
Late humanism, and the orchestra. Humanism and Democratic Criticism (2004, posthumous) is his positive programme: close reading and philology are not empire's accomplices but the only antidote, because they oblige you to respect the particularity of a text and refuse to let a type cover a case. In 1999, with the pianist and conductor Daniel Barenboim, he founded the West–Eastern Divan Orchestra (the name taken from Goethe's collection), putting young Israeli and Arab musicians at the same stands. That is his thought made physical: not a "dialogue of civilizations," which still concedes there are two, but one shared contrapuntal space in which nobody has to speak for a bloc.
① The book's object is not the East but the library about the East: a building used as knowledge, re-read as a specimen.
② "The Orient" was almost a European invention (his opening, paraphrased): it was made so that Europe would have a foil against which to confirm itself rational, adult, fit to govern.
③ The problem is not prejudice but discourse — an institution that fixes what can be said and who may say it. Put the hostile and the adoring in; the same thing comes out.
④ It runs on the textual attitude: people trust a book that has tidied reality over their own eyes, and when the book fails, the reader is blamed, so the book gains authority.
⑤ Latent versus manifest is the best tool here: for two centuries all the change was manifest, and the floor — one bloc, unable to move itself — barely shifted.
⑥ Its grammar is the eternal present: from "someone did something" to "that is what they are" — a falsifiable statement becomes an unanswerable verdict.
⑦ Balfour's case for ruling Egypt has exactly two gears: we understand it, therefore we may dispose of it. The view of the party understood never appears.
⑧ The mechanism is a loop: scholarship yields authority, authority licenses rule, rule produces archive, archive becomes scholarship — knowledge that controls the questions, the sampling and the interpretation is bound to validate itself.
⑨ The central power is not insult but speaking-on-behalf-of. The epigraph borrows Marx — they cannot represent themselves, they must be represented — and a group that can appear only as the object of description has lost the contest over what it is.
⑩ The hardest objection comes from a sympathizer: tracing Orientalism to Aeschylus makes it a trans-historical European essence (al-'Azm), which is the essentializing he prosecutes, and it did breed Orientalism in reverse. His own late answer was not to abandon humanism: there is no way past these texts except through them.