Hancher's project in this article is to guide the reader to a particular understanding of the meaning of the text as it relates to intention. That particular understanding has three moments: the text means what the author intended it to mean; the intentions which are relevant to the meaning of the text are completely independent of other aspirations the author may have entertained for the text; the meaning of the text is dependent on the author's intention but (almost) independent of the reader's interpretations and evaluations. To support his case, Hancher engages in substantial definition and illustration. But in some ways, his project has as much to do with foreclosure as definition. In order to focus our attention on the author's intention, Hancher brings up other perspectives we might be tempted to consider in order to rule them out. In particular, he is interested in foreclosing the discussion against our consideration of the importance to meaning of the reader's interpretation and evaluation of the text. In lieu of broader considerations, Hancher wants us to consider the author's intended meaning, what he calls active intention, as the meaning of the work. While this strategy seems reasonable and efficient, we should note comments in the last paragraph of the article in which Hancher refers to the two considerations he has fenced off and suggests without further discussion that they might play an important role in our understanding the meaning of the text.

Hancher opens the article by excluding from his discussion a variety of characterizations of intention. He tells us the three kinds of intention he will discuss do not exhaust the general concept. He will not consider practical, ethical, or legal aspects of intention, nor will he spend much time on any supposed autonomous intention in the literary text, nor the phenomenological intentions of the reader toward the work (827). It is not clear why Hancher should expect us to accept his delimitation unproblematically. The legal aspects of intention may not have a bearing on the discussion but according to Hirsch, for example, ethical concerns are the primary and unique justification for insisting on following authorial intention. It is not clear what the practical implications of intention are or why we should disregard them. Moreover, in a sentence he has dismissed the model that describes meaning as established in the interplay of author and reader without even a brief explanation. Hancher picks up the question of the reader's role later in the essay but we will see that he handles it in a way that make this early dismissal seem a philosophical precommitment rather than a reasoned conclusion.

Hancher takes Wimsatt and Beardley's "The Intentional Fallacy" as a point of entry into his discussion of authorial intention. It created a crisis, he says, in its insistence that the "design or intention of the author is neither available nor desirable as a standard for judging either the meaning or the value of a work of literary art". (cited by Hancher, 828). While Wimsatt urged that meaning and value were intimately related, Hancher sides with those critics who claim that interpretation precedes evaluation (829).

Hancher proposes that the "author's intention" as it figures in literary criticism and theory can be thought of as three separate elements. Programmatic intention is "the author's intention to make something or other". We think of it most commonly, for example, as the author's intention to write in one genre rather than another. Programmatic intention is subject to miscarriage, and is either realized or fails to be realized by the author diachronically in the work. We recognize it as the work succeeding or failing over time. Active intention is "the author's to be (understood as) acting in some way or other". Active intentions are the actions the author understands himself to be performing in the text at the time he finishes it. While the author's active intentions for the parts for the work can be thought of as coming together diachronically as the parts are written, the active intention for the work as a whole, the sum of these active intentions, is synchronic and is established at the time he completes the work (830, fn. 10). Active intention bears immediately on the text, and cannot be deflected from a finished text as programmatic intentions can. This is the form of intention that shapes a work's meanings at the time of completion. Final intention is "the author's intention to cause something or other to happen" as a result of having written the work (829-30).

These three forms of intention are independent of one another and heated arguments in literary criticism can sometimes be traced to confusion of active intention with either programmatic or final intention. Throughout the article, Hancher takes pains to establish and reinforce the independence of active intention from other forms of intention. He will remind us continually, for example, that: challenges to programmatic intention are not challenges to active intention or interpretation (834); evaluations of final intention are independent of the effect of active intention on meaning and value (835); neither failed programmatic intention nor failed projected active intention (a form of programmatic intention) disable active intention in what the author completes (836) and so on.

The meaning of a literary work is always an expression of active intention; the work always means whatever it is the author intended it to mean. In Hancher's view, it is by ignoring the categorical difference between active and programmatic intention that we fail to see equivalence of statements of the type 'The poet intended his poem to mean X' and 'The poem means X' (831, fn. 11). Note that Hancher exemplifies the difference between programmatic and active intention with examples from Brown and Warren and T.S. Eliot, examples that seem unclear to me (832-3).

There are three complications with programmatic intention: they involve a kind of active intention, what he calls projected active intention; they are necessarily broad and generic; and there is some question as to whether it has a bearing on interpretation or evaluation. First of all, then, some programmatic intentions are more complicated than others are. The author's intention to create an elegy or epic, for example, carries with it an expectation of the kind of meaning and force that will be transmitted in the active intention, in a way that programmatic intentions to create a sonnet or villanelle do not. These additional expectations Hancher calls projected active intention. The programmatic intention simply to write a poem is less complex than the intention, also programmatic, to write an elegy on the death of Edward King. Hancher is quick to remind us that "though programmatic intention often projects active intention of one sort or another, such projected active intention is not the same as the active intention that ultimately defines the meaning of the completed text" (837).

Second, programmatic intentions are always more or less general. If the author completes the text within the terms of his original programmatic intention, that intention has succeeded. If his work as completed falls outside the limits of his original intention, that intention has failed. Of course the author may alter his intention as he goes (838).

Now we are in a position to see Hancher's subtle handling of programmatic and active intention in an example of his devising. How do we understand the case when I write a comedy and my audience takes my work as a tragedy? In Hancher's view, when my programmatic intention to write a comedy technically succeeds it must be my active intention that fails (837). From Hancher's rebuttal of Roma above we know that my play will always mean whatever I intend it to mean. And from Hancher's explication of his own example we see that my programmatic intention to write a comedy technically succeeds. We can say, therefore, both that I have successfully written a comedy and that the comedy I have written means what I intend it to. Yet, somehow Hancher locates the failure of the comedy with active intention. The failure, moreover, seems to lie in the audience failing to get my meaning. So, in this case at least, the audience's faulty interpretation is my failure of active intent.

We get a clearer understanding of this characterization in Hancher's comparison of his intentional framework with Austin's locutionary moments. Active intention, according to Hancher, encompasses both Austin's illocutionary "force" of a statement (what I mean to do by uttering) as well as the rhetoric component of the locution (the sense and definite reference of my utterance). According to Hancher's understanding of Austin, the performance of the illocutionary act involves the securing of the uptake of the audience. In order for an illocutionary act to be successful, the audience must recognize the meaning and force of the utterance (841). So, in Hancher's scheme, the work always embodies the author's active intention but the active intention may fail if the audience does not recognize the meaning or force of the writing.

In light of this analysis, Hancher's firm claim that the work always means what the author intends it to mean seems equivalent to just the assertion that the author always intends the work to mean what he intends it to mean. Otherwise, we must understand meaning to fail rather than understanding the author to fail in his intention to convey his meaning. We can get at this logic through Hancher's discussion of Grice. Hancher takes Grice to say that in order for an utterance to mean something the utterer must intend the audience to recognize his prior intention in the utterance (843). In other words, the utterance means what the author intends it to mean by virtue of the audience recognizing the author's intentions in the utterance.

It is interesting to read Hancher's rebuttal to Beardsley later in the paper in light of this analysis of Hancher's view of intention. Hancher asks whether the characterizations he has been spelling out apply not only to regular speech but to literary works as well (never mind that virtually every illustration he provides comes from a literary work). He anticipates Beardsley's objection that it is the publicly ascertainable meaning of the text that will most reward our attention, and not any meaning the author meant the text to bear (p. 848). Hancher gives two weak rebuttals. But what is more interesting is that Hancher still sees the need to constrain the concept of meaning after a long discussion that seemed pointed in the direction of recognizing a definition of the meaning of the text that includes two moments: the author's intention and the audience's recognition. He has already said in essence that a text means what the author intends by virtue of the audience recognizing his intentions. It seems a simple step to say that the phenomenology of meaning is itself bipolar: that it has an intentional component, and a receptional component. Remember that Hancher has already said in so many words that even when the text means what the author intends, that active intention fails if the audience does not recognize the meaning.

Finally, by arguing that an utterance reference the token of a concept and not the type, Hancher refutes even the objection that a given utterance must conform to conventional rules of usage in order to have the meaning intended for it (pp. 843-4).

Third, programmatic intention has no bearing on the interpretation of a literary work. Meaning is a matter of active intention and not programmatic intention by definition. Judgments of a work on programmatic grounds are actually judgments of the author and they don't stand up to our judgments based on other standards we may apply. We would not, in the example above, call the author of humorless comedy successful simply because he produced a comedy. Value turns on active intention (pp. 838-9).

While Hancher insists that the text means what the author intends it to mean, he does not claim that active intentions can simply be read off the face of the ambiguous text. They often need to be estimated from external evidence (p. 847).

Hancher's last discussion covers final intention. Final intention, what the author intends to cause by writing the text, is not in the text itself but exists as criticism. It bears on the work only in so far as it informs active intention. In fact, the Melville example that Hancher provides seems to indicate that final intent bears on a text only when it is overtly laid out as part of the text. Final intentions, in so far as they actually affect meaning, inevitably reduce to specific details of active intention. Whether such relevant final intentions succeed or fail have no effect on the author's intentions and hence have no meaning (p. 849).

Hancher's parting issue opens up the whole question of intention again in a new light without providing any substantive discussion or resolution. Are deviant intentions and meanings, read into the work by the audience, to be dismissed out of hand? "I think not", replies Hancher, "particularly when they make the work better" (p. 851). By so saying, Hancher has reopened the whole can of beans as he leaves it. He undermines his strongest thesis that meaning is a function of author's intentions only and not of the reader's reception. And he turns on its head his framework of taking interpretation as prior to evaluation. Interpretation is prior, he seems to say, except when our evaluation tells us that a different interpretation would be better.

His first move and his last have profound consequences but they bear almost no discussion whatsoever. In the first, we reduce intention to one pole and rule out evaluation. This is the same reductionism that McLaverty picks up even when McLaverty talks about phenomenological intention. In the last, we introduce the reader back into the determination of meaning and allow evaluation in the door. We were kidding that intention should really be reduced to one pole.