Materialist criticism is scientific also because it "situates itself outside the space of the text on the alternative terrain of scientific knownledge" and make manifest those conditions of the text's making which the text itself is unaware of.  The text, though "unaware of" its own making, is not a passive "reflection" or "reproduction" of the dominant ideology.  It is an aesthetic production of the ideology that constitutes it.  The ideology that is at work in a text, moreover, consists of not only the General Ideology (GI), but also the Aesthetic Ideology (AI), the ideology of its Literary Mode of Production (LMP), and the Authorial Ideology (AuI).  The task of a materialist critic, then, is to specify the precise relations between a text and GMP (General Mode of Production), GI, AI, LPM, AuI and the ideology of LMP.  It is a science based on the science of ideological formations, but it does not study "the laws of idielogical formations."  Instead, it studies "the laws of the production of ideological discourses as literature" (96-97).

      The major constituents of materialist criticism, GMP, LMP, GI, AI, AuI, and the text, are closely interrelated and interdetermined.  A GMP produces both a (dominant) LMP and a GI, both of which in turn contribute to reproducing the GMP.  The conjuncture or disjuncture of GI and AI also has an effect on the constitution of a LMP and a text, the ideology of which again inserts itself back into AI and GI.  (This insertion can be very powerful in the educational apparatus.)  The ideology of the text is thus the product of the conjunctures between GMP, LMP, GI and AI.  The only level of ideology that "effectively disappears" in the text is AuI, which is either homologous with GI/AI or "canceled" by their disjoint or conjoint effect. 

     In discussing how the text "actively determines the determinants," Eagleton leaves out the specified layers of ideologies altogether and describes the text in terms of its relation to ideology and history.  More interestingly, Eagleton goes "into" the text to talk about the relation between its signifier and signified.  History appears in the text as ideology, and the text, in constituting in itself its relation to ideology, reveals the relation of ideology to history.  In this sense, history is the ultimate signified and signifier of the text.  The text, on the other hand, appears to be free and self-generative in relation to history, since it fore-grounds its signifier and distantiates its signified.  This "disturbance" of the normative relation between the text's signifier and the signified is what makes the text relatively autonomous and "poetic." 

     The relation between the text and ideology that Eagleton describes here is dialectical, if not paradoxical.  On the one hand, he says that ideology "presents," "offers" itself to the text, and that the former's determination of the latter is precise and rigorous (81; 67).  On the other, however, he argues that the text constitutes ideology in "ways unpremediated by ...ideology itself" and that the text can assert its "pure autonomy" to some of its pre-existent materials (80-81).  While he claims that the autonomous formation of the text is based on its "internal bonds" to ideology (98), it is not clear how the text retains its "internal bonds" to ideology after it "selects...displaces, recasts, and mutates" ideology's different modes of significations according to its own relatively autonomous aesthetic laws. 

     In his ambivalent description of the text's reworking of ideology, moreover, Eagleton varies the term from "operate," "refine," "refashion," "reconstitute" to "produce," "determine," "transform," "destructure" and "displace."  This variation of terms reveals, if not Eagleton's shifting views about the text's relative autonomy, the differential relations between the texts and the ideologies, which Eagleton fails to specify in his move "towards a science of the text."  To the different layers of ideology, different texts bear different relations, which, to me, include both those of mere reflection on the one hand and negation on the other.

      Another thing that I find dissatisfactory is Eagleton's omission of both the author and the reader from his materialist criticism.  Eagleton analyzes very well how the conjuncture and disjuncture between AuI and GI can be determined "diachronically" and "synchronically."  But he is wrong in saying that in the text, the level of AuI "effectively disappears."  It is true that in some texts the AuI is either homologous with GI/AI, or cancelled by them.  But in the other texts, the AuI negates both GI and AI, and thus becomes an important determinant of the text.  (For instance, the overdetermination of the authorial-biographical factors in Tristram Shandy can not be ignored.)      Against what he calls Althusser's "consumer-centredness," Eagleton argues for a "return to the productive process of the text" (86).  Moreover, he believes that materialist criticism should be "outside the space of the text on the alternative terrain of scientific knownledge."  Whether in or outside the text, materialist criticism is endowed with a transcendent position not possible to any other critical methods.  Since Eagleton is aware of the fact that "reading is an ideological decipherment of an ideological product" (p. 62), his criticism must be different from "reading."  But what is Eagleton himself if not a reader?  And how can his analysis of, say, D. H. Lawrence be anything but a reading?