Theory of Literature is a book on literary scholarship by René Wellek, of the structuralist Prague school, and Austin Warren, a self-described "old New Critic".

[1] The two met at the University of Iowa in the late 1930s, and by 1940 had begun writing the book; they wrote collaboratively, in a single voice over a period of three years.

[5] His theoretical training included the phenomenology of Edmund Husserl, as used in Roman Ingarden's work, and the psychologically influenced linguistics of Karl Bühler.

[6] After Nazi Germany occupied Prague in 1939, Wellek fled London – where he had been teaching – for the United States, finding a position at the University of Iowa under Norman Foerster.

[11] Owing to several academic commitments, work on Theory of Literature did not begin until 1945, after Wellek and Warren received a stipend from the Rockefeller Foundation over a period of two summers.

[37] Wellek and Warren note the importance of identifying forgeries, a task which can be completed in numerous ways: paleography, bibliography, linguistics, and history may all be involved.

[37] At the first level one must locate and identify materials to study, be they written, printed, or oral; such a task may be difficult and depend on factors outside literature in its completion.

[39] Meanwhile, the second level may require greater initiative from the one studying a work; it involves, among other things, selection of what should be published, how it is best arranged in a collection, the establishment of chronology and authorship through internal and external evidence, and the provision of proper annotation and commentary.

Although "[n]obody can deny that much light has been thrown on literature by a proper knowledge of the conditions under which it has been produced", such studies "can never dispose of problems of description, analysis, and evaluation of an object such as a work of literary art.

[44] Likewise, an understanding of personal style (what makes a work "Miltonic", "Keatsian", "Shakespearean", or "Virgilian") does not rely on knowledge of the author's life.

Instead, they agree with the German scholar Rudolf Unger that "literature expresses a general attitude toward life, that poets usually answer, unsystematically, questions which are also themes of philosophy", in a manner that differs over time.

[55] They outline attempts at classifying these ideas, including through Weltanschauung ("world view") and Geistesgeschichte ("time spirit"), before showing shortcomings in these systems.

[64] Instead they suggest that literature is a "potential cause of experiences" consisting of a system of stratified norms – implicit in the work – which can only be partially realized by the reader;[65] it is neither purely material, mental, nor ideal, nor is it static or bereft of value.

[66] Wellek and Warren consider patterns of sound as inherent to the text; these must be analyzed while keeping the meaning (or general emotional tone) in mind.

[75] For other understandings of meaning, Wellek and Warren suggest a look at the sequence of image, metaphor, symbol, and myth, which they consider making up the "central poetic structure" of a work.

[76] In turns, they outline various historical definitions of the terms – which at times overlap – before writing that most of these theories have treated the sequence as "detachable parts of the works in which they appear.

[77] They show that the dominant form of figurative language shifts over time[78] before overviewing two diverging typologies of metaphor, that of Henry W. Wells and Hermann Pongs.

[89] They note various criteria used to identify "good" literature, rejecting Russian formalism's criterion of defamiliarization and similar understandings for one based on the diversity of materials amalgamated within a work.

[98] They note that this opportunity may, however, be lost in a conflict between those advocating change and the inertia (including persons defending the status quo) in American literary studies and institutions.

[105] Also unlike their forerunners, Wellek and Warren saw aesthetic value as not the defamiliarization of the mundane, but an interaction among the strata derived from Roman Ingarden's work: the phonological (sound) level at the base, then semantic (meaning), and the "world" created by literature.

[108] Wellek and Warren's concept of aesthetics borrowed from the writings of Immanuel Kant, implying that a specific "aesthetic realm" was autonomous within the work and required a certain perspective to properly understand;[109] they emphasize this with a quote from the neo-Kantian philosopher and literary critic Eliseo Vivas, that beauty is a "character of some things ... present only in the thing for those endowed with the capacity and the training through which alone it can be perceived".

[110] Meanwhile, their depiction of a dynamic scale of values, as opposed to an anarchical one, is a reimagining of perspectivism, which Wellek and Warren define as "recogniz[ing] that there is one poetry, one literature, comparable in all ages, developing, changing, full of possibilities".

[17] By 1976 Wellek was of the opinion that the book required updating, but asked rhetorically "who can master the astonishing and bewildering literature on theory which since [1949] has been produced in many countries?"

"[118] Although Hatzfeld agreed with Wellek and Warren's main points, he thought it lacking in references to theories and literature from the Romance languages[119] and concrete interpretations.

[120] William Troy, writing in The Hudson Review, echoed the sentiment, stating that, although the book was "unusually difficult" to read,[121] he felt "unqualified agreement with the main position".

[122] Seymour Betsky, writing in Scrutiny, praised the book's summary and adjudication; he wrote that it was "in its way impressive", a "tour de force" which would "usher in a new era".

[123] However, Betsky felt that the book lacked a "controlling purpose" and that it neglected to emphasize the need to differentiate between "the cheap commercial appeal and the genuine" literature.

[125] In The Kenyon Review, Vivas wrote that the book's discussion of the relation between literary criticism and scholarship "leaves nothing to be desired", providing a "well balanced" look at the major points;[126] he found that no other such work existed in English at the time.

[128] Newton Arvin, writing in the Partisan Review, found Theory of Literature to excessively indulge in formalism and expressed concern that the idea of literary history may have "gone into the discard once and for all".

[132] Ingarden, who believed his theories the basis of Wellek and Warren's arguments, considered himself inadequately credited and took offense with the attribution of his ideas to "pure phenomenologists".