Charles Larmore
What can be meant by “the ethics of reading” is not immediately obvious. The phrase has become popular in contemporary literary criticism largely through the influence of J. Hillis Miller.1 What it suggests is that reading is a practice that raises questions of ethical significance. Yet why this should be so can seem puzzling and the writings of literary critics who invoke the phrase have done little to dispel the obscurity. Consider the term “ethics” itself. In the philosophical tradition, ethics as a discipline has been concerned with two distinct though interrelated questions: how we ought to live in order to live well, and how we are to treat one another and perhaps other living beings as well. In recent times, the term has also been used to refer, more specifically, to the first of these subjects in contrast to the second, which is then commonly classified as “morality.” Now in what way does reading constitute an ethical phenomenon in either of these two respects? Certainly, we live richer lives if we read some good books, and there are likewise many cases in which reading various materials (ranging from newspapers to the Bible, say), and reading them well, can deepen our understanding of how we ought to act with regard to others. But talk about an “ethics of reading” has, I believe, something more essential in mind, namely, that our very relation as readers to what we read—to books or more generally to texts—is of ethical significance. How can this be so? This is the question I will pursue, though my answer will be rather different from the sort that many literary critics would be inclined to accept.
The most basic fact about our relation as readers to texts is that this relation is asymmetrical. We can read the text, but the text cannot read us. Sometimes, I know, literary critics talk about “being read by the text,” but this is a metaphorical usage, referring to the experience of being challenged by the text to reflect on some of our central assumptions, an experience we can have only in and through our reading of the text. Reading is an act, initiated and performed by an agent, and texts themselves are not agents: Texts speak to us only when they are read, and read by us. Now if the reading relation is essentially asymmetrical, and if it is moreover a relation not to another person, but to something else, namely, to a text, how can it be a relation that is, in itself, of an ethical character?
You may feel that I have left out another equally essential feature of our relation as readers to what we read: Though the text that we read is not another person, it was written by a person to embody his or her thinking and feeling. I think this point is right and, moreover, absolutely crucial. If a text is not conceived as being the expression of its author’s intention, then the practice of reading cannot itself be of intrinsic ethical significance. We may consider the ways that reading enriches the lives of readers or gives them instruction about how to treat others well, and in this sense we can talk about the ethical importance of reading, that is, about its beneficial consequences. But there can be an “ethics of reading” that regards the practice of reading as itself a matter of ethical moment only if the relation of reader to text is ultimately a relation of the reader to another person. However, I want to postpone a clarification of the problematic notion of “author’s intention” (I shall return to it shortly) in order to focus on the salient fact about the reading relation, which is that it connects us to another person only indirectly. In the act of reading, the author is not there, only the text. For it is this fact—the author’s absence—that constitutes what is special in the ethical character of the practice of reading.
That reading in this regard, given the absence of the author, gives rise to a special sort of ethical problem is glimpsed by Plato in his discussion of writing at the end of his dialogue, the Phaedrus. I say “glimpsed” because Plato’s main concern in that passage lies with the dangers of writing rather than with the responsibilities of reading, although the two are connected and although one of the remarks that Plato has Socrates make points to what must be the fundamental concern of an “ethics of reading.” The trouble with a text, Socrates says, is that “roaming about everywhere” it comes to be read both by those who are intelligent readers and by those who are not, and yet it is by nature unable to defend itself against the abuses of misunderstanding that may be perpetrated by the latter: “When it is wronged (plemmeloumenos) and attacked unfairly (ouk en dikei), it always stands in need of its father’s support, for on its own it can neither defend itself nor come to its own support.” (275e). By “its father” is meant, of course, the author of the text, and Plato is claiming that the text itself, unlike its author, cannot correct the misinterpretations by which readers may treat it unfairly as they fail to listen to what it says or distort its meaning for their own purposes.
Now it is essential to distinguish the two parts of Plato’s claim: that (1) the text cannot talk back and object to an interpretation of its meaning that has been foisted upon it, and that (2) the author, if he or she were present, could authoritatively correct the erroneous interpretation. We need not accept the second part—in fact, we ought not to accept it since it is untrue (I will have more to say about this)—in order to recognize the crucial importance of the first part for the idea of an “ethics of reading.” It is a corollary of the essential asymmetry of the reading relation that I underscored before. When people are wronged, they are in principle (if they have not been killed in the process) able to object to their mistreatment. Texts, by contrast, cannot object to how they are mistreated—that is, to how they are inadvertently, carelessly, or deliberately misread. In this fact lies, as I have suggested, the special ethical significance of the reading relation.
However, the practice of reading can have an ethical character at all and there can be such a thing as an ethics of reading only if, as I have also argued, our relation as readers to texts is ultimately, if not directly, a relation to another person. The text must be understood to be the expression of its author’s intention. Moreover, if misreading the text consists in failing to grasp what the text itself says—and what else could misreading mean, if it is to mean anything at all?—then what the text says, and in this sense its meaning, must be defined by the author’s intention. Since I have rejected the idea that the author is the authoritative interpreter of a text’s meaning, I need to say how I think the notion of author’s intention should be conceived.2
That this notion is indispensable, on at least some construal, ought to be indisputable; unfortunately, among literary and hermeneutic theorists it is not. Texts do not write themselves. They are written by authors and are written to some purpose or variety of purposes. If we did not suppose that the text was the expression of the author’s intention, we would have no basis for thinking, as we do, that to understand the meaning of some problematic passage we should turn, first of all if not exclusively, to other passages of the same text or to other writings by the same author. We consider them to be directly pertinent because we presume that they have been produced by the same cause as the passage in question, namely, the mind of the given author.
Yet the term “the author’s intention” can signify a number of different things. In some of its senses—what the author set out to say, what he said to himself as he composed the text, what he may have said afterward about his aims—the author’s intention need not coincide with the text’s meaning. Contrary to Plato, the author (though usually a commentator of interest) is not the authoritative interpreter of what a text says. That is because what the text means consists in what it itself says, and not in what the author says about it. Yet the author’s intention in still a further sense—namely, what the author intended insofar as he realized this intention in the composition of the text, with whatever degree of self-awareness—must count as determining what the text itself says or means. For again, if it is possible to misunderstand a text, then there must be something that the text actually says, and as texts do not write themselves, what they say can only be what their authors meant in the very process of writing them as they did.
To be sure, we do not read books solely for their textual meaning. We do not aim only, or even principally, to determine what a book’s author meant to say, but rather seek to discover what the book means for us. We may, for instance, be interested in seeing how it forms part of a larger cultural movement or in charting its influence on other works. More commonly, we may want to find out how it speaks to our own interests and preoccupations. If it is a legal text (a statute, judicial opinion, or constitutional provision), we may want to ascertain its import for a particular case or social problem. However, all these endeavors require that we first form an idea of what the text itself says or means. For how else can we gauge its relevance in these various ways? And what it means, as I have argued, consists in the author’s intention as realized in the text itself. The distinction between a text’s meaning in this sense and its significance, its relevance to the reader’s concerns, is indispensable (though I think there are often errors in the way this distinction is drawn, for instance, by E. D. Hirsch).3 True, part of the author’s intention is normally that the work be in some way significant for its readers; yet intended and actual significance are obviously two quite different things. I know that many literary theorists dismiss the notion of authorial intention and even the notion of textual meaning. But without these concepts, we can make no sense of such basic features of the reading process as the possibility of misreading and the perception of relevance.
Having clarified the idea of author’s intention, I return to the ethics of reading. My thesis has been that there can be an ethics of reading only if the reading relation is itself ethical in nature. It is so, I have argued, in that it is ultimately a relation to another person, namely, the author. However, I have also been arguing that the ethical nature of the reading relation is special in character because it relates us to that person only in the form of a text, which, unlike a person, is unable to object to how it may be mistreated—that is, to how it may be misread. Indeed, even if the author were there as we read and objected that we were misinterpreting what he had written, his interpretation of the text’s meaning would not be necessarily correct, nor ours necessarily wrong. No one can speak authoritatively for the text, and the text cannot defend itself.
Now because of this difference between the text and its author—the text is not itself a person, nor what it says or means the property of its author—one might ask whether the reading relation is in fact an ethical relation and whether there can therefore be such a thing as the ethics of reading. Is misreading a text really to wrong the author? I think it is, particularly when the misreading is not simply inadvertent, but the result of carelessness or deliberate distortion. Consider the case of actions, which are not so unlike texts. Suppose that someone regards the actions of a certain person as of so little consequence that he fails to see the other’s gesture for the generous offer it is or finds it useful for his own interests to portray that offer as something altogether different, say, as an attempt at ingratiation. Surely we would think that he has wronged the other person, even though the immediate object of the wrong was the other’s action (which has been thoughtlessly or perfidiously construed) and even though that action is neither itself a person (and so cannot defend itself) nor the property of the agent, whose statements after the fact about his intentions are not authoritative. The agent has been wronged because what he was doing in acting as he did has been slighted or deliberately falsified. It is similar, I believe, with authors and texts.
Texts do differ from actions in an important respect. Agents are generally in the vicinity of their actions, so when their actions are described falsely, or when they believe that they are, they can object. Texts, by contrast, “roam about everywhere” (kulindeitai pantachou), as Plato says, and do not have their authors nearby, or necessarily anyone, to stand up for them when they are misread. To write a text, to express something of oneself in writing, is thus to make oneself especially vulnerable to others. In this fact there appears the ultimate import of the ethics of reading: it highlights the very essence of the ethical outlook. In general, the moral point of view consists in seeing in another’s good a reason for action on our part, apart from all consideration of our own good. As a result, our ethical character shows itself most clearly in how we treat the vulnerable, since they cannot make it in our interest to treat them well. Though I will not go so far as to say that a bad reader cannot be a good person, I doubt that a habitually careless or deliberately manipulative reader will be one.
I close with a similar observation by Karl Kraus, the Viennese satirist and moralist, about the respect for language in all its forms. It is, he said, a “mental discipline that enjoins the highest measure of responsibility toward the sole thing that can be wronged with impunity, namely language, and that serves like no other to teach respect for every other kind of human good.”4