TWO

Method in Interpretation

Interpretation theory, or hermeneutics, is a subject of much current debate. One reason for this is, no doubt, the wide variety of specialized topics which fall under its scope and in regard to which it promises to open productive avenues of approach—topics such as the relation between the natural sciences and the humanities, the methodological status of the social sciences, the nature of language and its role in human understanding, the structures of understanding itself, in its various modes, the relation between language and reality, and so on. Hermeneutical theory is a veritable crossroads where tendencies as diverse as phenomenology and linguistic analysis, semantics and the critique of ideologies, structuralism and conceptual analysis, Marxism and Freudianism come together. Hermeneutics is a subject as central to theology as it is to jurisprudence, as central to philosophy as it is to literary criticism.

Indeed, one of the major ongoing debates in hermeneutics is one in which one of the major protagonists is an American professor of English, and the other, a German philosopher. I refer to E. D. Hirsch and H.-G. Gadamer, respectively. Both Hirsch and Gadamer have their antecedents in the nineteenth-century hermeneutical tradition of Schleiermacher and Dilthey, and both appeal, to a greater or a lesser degree, to Husserlian phenomenology. Despite this, Hirsch and Gadamer represent two divergent and irreconcilable tendencies in contemporary hermeneutics. Hirsch’s Validity in Interpretation is, in effect, a systematic attack on the version of hermeneutics worked out and defended by Gadamer in his major treatise, Truth and Method.¹ The principal difference between these two leading theorists is that whereas Gadamer seeks to defend what is proper to the humanities against encroachment by the ideal of “scientific” knowledge and to this end attacks the concept of “method,” arguing that method, as it it is understood in the positive sciences, has no role whatsoever to play in the humanities, Hirsch, inspired by logical positivism, argues that there is or should be no significant difference between the empirical sciences and the humanities and that the hypothetical-deductive method as advocated by positivist-style philosophers of science is as applicable in the matter of literary textual interpretation as it is in the physical sciences.² The main thrust of Hirsch’s criticism of Gadamer is that his position opens the door to arbitrariness and cannot therefore serve to make the métier of the interpreter a serious, respectable business.

This conflict is an extremely fundamental one, in that it involves two irreconcilably different theories of understanding and interpretation. It is a conflict between what could be called positivistic hermeneutics and phenomenological hermeneutics. While I do not believe that Hirsch’s criticism of phenomenological hermeneutics is itself “valid” or that his version of hermeneutics can be defended and have myself taken a place in this ongoing debate by arguing against it elsewhere,³ I do believe that Hirsch’s criticism of the Gadamerian, phenomenological version of hermeneutics must be taken seriously. For only if it could be shown that phenomenological hermeneutics does not afford a license for arbitrariness and does in fact provide for methodological rigor in interpretation could phenomenological hermeneutics be positively argued for and defended.

The problem is that in arguing against the application of scientific method to interpretation, Gadamer has extremely little to say about what the methodological criteria of interpretation should be. I do not think that Hirsch is entirely without justification when he says that “Gadamer’s most precise statements are those which declare what the norm is not” (VI, p. 251). In deliberately opposing truth to method and in saying that “a philosophical theory of hermeneutics is not a methodology” (TM, p. 466), Gadamer tends to give the impression that method has no place in interpretive understanding.⁴ I do not feelthat this is the case, i.e., that there is no place for method in interpretation, and even a phenomenological thinker as close in so many ways to Gadamer as Paul Ricoeur has expressed reservations as to the satisfactoriness of Gadamer’s stance.⁵

The point I wish to argue here is that while Gadamer does indeed tend to ignore the subject of what the criteria of interpretation should be, phenomenological hermeneutics, as represented by him, can provide for norms or criteria for assessing interpretations, and I wish to list of few of the more important of them. Before doing so, I should indicate why I think that the question of method is so important. There are, I think, two major reasons why an acceptable hermeneutical theory must provide for methodological criteria.

In the first place, while I agree with Gadamer that the scientific and Hirschian preoccupation with “objectivity” is itself a supreme instance of modern subjectivism, I feel that in his own reaction to subjectivism, which is influenced by Heidegger’s rejection of “humanism,” Gadamer does not accord enough importance to the notion of subjectivity, understood in a nonsubjectivistic sense. His key notion of “effective history” tends to privilege “what is said” over “the speaker” and tends to make of subjectivity a mere “expression” of the really active factor (the real “subject”) in interpretive understanding: the Tradition itself.⁶ In other words, Gadamer does not appear to make a sufficient distinction between what, in traditional terms, is referred to as the ordo essendi and the ordo cognoscendi. While, in order to counteract subjectivism, it is necessary to maintain that the Tradition, Being, or What-not has ontological primacy and that “in the last analysis” it is this which “expresses itself” or “comes to expression” in the understanding of the individual, it is nevertheless necessary to recognize that it is the subjectivity of the interpreter himself which has methodological primacy. This is necessary because only the individual, human, conscious, reflecting subject can be held responsible for what he or she says or does (Being, the Tradition, and so on, cannot be held accountable for what it “does”). To be responsible means to be able to respond, i.e., to be able to make an attempt at defending or justifiying one’s own words and deeds, which is to say, at providing arguments for them. Now to argue is always to appeal to certain principles, the ensemble of which can be said to constitute a set of criteria. In short, correct method is necessary if interpretation is to be a responsible business.

In the second place, a viable hermeneutics must allow for method if it is to be in a position to grapple with what is after all the most urgent problem arising out of the actual work of interpretation. This problem is pinponted by Hirsch when he remarks: “Suppose, as it often happens, two readers disagree about the meaning of a text at exactly the same moment of time. What principle would they have for determining who is more nearly right?” (VI, p. 249). In other words, methodological criteria are needed in order to arbitrate what Ricoeur would call “the conflict of interpretations.” How is one to decide which of two or more conflicting interpretations is the better, and to do so impartially, nonarbitrarily, if there are no general, recognized criteria one can appeal to? It does not suffice to say (as one sometimes gets the impression Gadamer is saying) that the criterion is “die Sache selbst.”⁷ This, at the very least, amounts to a petitio principii. It also does not enable one to come to grips with the problem of choosing between interpretations (anyone and everyone can claim that their interpretation is the meaning of the thing itself, and thus such a claim gets us nowhere). At its worst, it can provide a license for irresponsibility (why should one have to defend one’s views in the forum of public opinion if they are sanctioned by the thing itself?).

A satisfactory philosophical theory of hermeneutics must therefore include, in addition to a basic theory of just what understanding is (of the sort provided by Gadamer)—and on the basis of this—an organon or set of criteria to be adhered to in the actual work of interpretation. Theory is always the theory of a certain praxis, which means that theory should be able to serve as a guide in certain concrete activities, in this case, interpreting texts. There is always a back-and-forth between theory and practice, in the sense that a theory can always be translated into a concrete procedure to be followed in practice, and also in the sense that the satisfactoriness (or lack thereof) of the practice is a measure of the truth (or falsity) of the theory. The question then is: What would be the appropriate methodological criteria for a phenomenological hermeneutics? Or again: Is phenomenological theory such as to allow for practical criteria to be derived from it?

To respond to this question, it is first necessary to indicate in general what sort of method is compatible with a phenomenological hermeneutics, for it is most certainly the case that method, as Hirsch understands it, the hypotheticaldeductive method of science, is not compatible with it. It seems to me that there are two general ways of viewing method and that the second of these has its place in a phenomenological hermeneneutics.

(1) METHOD IN AN ABSTRACT AND FORMAL SENSE. This is method in the modern sense of the term. Method in this sense dispenses with personal, subjective judgment (this indeed is one of its raisons d’être). One has only to learn the method itself, in and for itself; it is an intellectual technique (like the “scientific method”). Having done so, one has only to apply it to whatever subject matter one chooses; the only criterion in applying the method is correctness of application (not appropriateness in choosing to apply it in the first place); one’s guide is the method itself, not the subject matter (such as human beings) to which it is applied. This is assuredly a rational undertaking, where rationality refers to what people like Habermas would call “instrumental rationality.” The purpose of method in this first sense is to make possible exact knowledge.

(2) METHOD IN A NORMATIVE SENSE. Method in this sense, far from supplanting personal, subjective judgment or eliminating the need for it, is meant as an aid to good judgment (I say “good,” not “correct”). It is what ensures that judgments or conclusions arrived at are not gratuitous or the result of subjective whim. It therefore makes for rational judgments—not in the aforementioned sense, but in the sense that one can give reasons (persuasive arguments) for the judgments one makes: One can defend one’s judgments or interpretations by arguing that they embody or conform to certain generally accepted criteria, norms, or principles. It therefore makes for responsible judgments. The norms here in question could be compared to ethical norms. As moral theology teaches us, such norms can never be simply “applied” to concrete situations where an ethical decision is called for in such a way as automatically and unequivocally to tell us what exactly we should do. They (much to the chagrin of medical students and nurses, for example) do not tell us what decisions we must make in a given case; they serve only as guiding principles in choosing among courses of action. “Applying” them cannot, therefore, be a science but is always only an art. Strictly speaking, they are more properly referred to as principles than as rules (they are more like laws than like orders). Rationality here does not consist merely in subsuming an individual under a general rule, requiring nothing more than the recognition of the necessary and sufficient conditions for application of the general rule. Rationality here demands interpretation, as when a judge interprets the law in “applying” it. Unlike the methodological norms Hirsch would like to see in place, these have no “capacity to enforce practical decisions” (VI, p. 206). Thus the implication here is, again, that there can be no science of interpretation, i.e., there can be no “ruthlessly critical process of validation” (VI, p. 206). One cannot become a good interpreter simply by mastering a certain method. One therefore does not “test” interpretations, one evaluates them. The set of interpretive principles can be called a method, if by method we mean a system whose purpose it is to orient action. Method in this second sense can be defined as a rational discipline which formulates the norms for a certain procedure. Or, again, it is a norm-governed way of doing something (in distinction from arbitrary, whimsical behavior).

Understanding method in this second sense, we can ask: What are the methodological principles appropriate to, or derivable from, a phenomenological hermeneutics? I advance the following as some of the more likely candidates (wherever possible I have illustrated these “rules” with Gadamerian texts).

(a) COHERENCE. The interpretation of an author’s work must be coherent in itself; it must present a unified picture and not contradict itself at points. (This rule holds true even if the work being interpreted has contradictions of its own; the interpreter must then attempt to make coherent sense of these contradictions.)—“The harmony of all the details with the whole is the criterion of correct understanding. The failure to achieve this harmony means that understanding has failed” (TM, p. 259).

(b) COMPREHENSIVENESS. Unlike(a), which concerns the interpretation as such, this concerns the relation of the interpretation to the work itself which is interpreted. In interpreting an author’s thought, one must take account of his thought as a whole and not ignore works of his which bear on the issue. Whether or not a philosopher seeks to express, à la Hegel, a “system,” the thought of any author worth interpreting—even an “unsystematic” one such as Kierkegaard—forms something like a “system,” i.e., a unified whole.—“ . . . the ‘fore-conception of completion’ . . . is obviously a formal condition of all understanding. It states that only what really constitutes a unity of meaning is intelligible. So when we read a text we always follow this complete presupposition of completion, and only when it proves inadequate, i.e., the text is not intelligible, do we start to doubt the transmitted text and seek to discover in what way it can be remedied” (TM, p. 261).

(c) PENETRATION. A good interpretation should be “penetrating” in that it brings out a guiding and underlying intention in the work, in this way making an author’s various works or statements intelligible by seeing them as attempts to resolve a central problematic. (The need for this rule is especially apparent in the case of authors such as Heraclitus, of whom we possess only fragments).

(d) THOROUGHNESS. A good interpretation must attempt to answer or deal with all the questions it poses to the interpreted text, or which the text poses to one’s understanding of it.

(e) APPROPRIATENESS. To be considered a valid interpretation of a text, the questions the interpretation deals with must be ones which the text itself raises; if one claims to be interpreting, one must not simply use the text as an occasion for dealing with questions of one’s own having nothing to do with the questions the original author was concerned with.—“The real power of hermeneutical consciousness is our ability to see what is questionable” (PH, p. 13).

(f) CONTEXTUALITY. This principle is related to the preceding one. An author’s work must not be read out of context, i.e., without due regard to its historical and cultural context. (Herein lies the usefulness for interpretation of philology, historiography, and other such disciplines.)

(g) AGREEMENT (1). An interpretation must agree with what the author actually says, that is, one must not, or normally not, say that the “real” meaning of what an author says is something quite other than what he actually does say and intends to say. (This is a principle rejected by reductive interpretations such as those of a Marxist or a Freudian variety, what Ricoeur calls the “hermeneutics of suspicion.” It should be noted, however that to the degree that the suspension of this criterion enables us to see in the text things we might not otherwise have seen, and to the degree it can be buttressed by the other criteria, it is a worthwhile countertactic of reading.)

(h) AGREEMENT (2). A given interpretation should normally be in agreement with the traditional and accredited interpretations of an author. This principle must not be blindly adhered to, however, for often a good interpretation will be precisely one which breaks with traditional readings, in that it opens up new perspectives on the work. (In this case the interpretation must still take account of previous interpretations, by showing how they are deficient.) This is related to the following principle.

(i) SUGGESTIVENESS. A good understanding will be “suggestive” or fertile in that it raises questions that stimulate further research and interpretation. This is where originality finds its place in interpretation. —“It is imagination that is the decisive function of the scholar. Imagination naturally has a hermeneutical function and serves for what is questionable. It serves the ability to expose real, productive questions, something in which, generally speaking, only he who masters all the methods of his sciences succeeds” (PH, p. 12).

(j) POTENTIAL. The ultimate “validation” of an interpretation lies in the future. A given interpretation can be judged to be “true” if, in addition to meeting the above requirements, it (like a good metaphor or model) is capable of being extended and if in the process the implications it contains unfold themselves harmoniously (cf. in this regard what Husserl and Merleau-Ponty have to say about perception and its veracity). —“The only ‘objectivity’ here is the confirmation of a fore-meaning in its being worked out. The only thing that characterizes the arbitrariness of inappropriate fore-meanings is that they come to nothing in the working-out” (TM, p. 237).

In laying out these criteria I have not attempted, à la Hirsch or anyone else, to lay down new rules for interpretation; I have simply attempted to articulate, theoretically, the practice which is followed, I think, by most interpreters. I have simply attempted to raise to the level of theory a certain widespread practice. Nonetheless, I can anticipate a possible objection from Hirsch and others like him. The norms I have set out are so general, they will say, as to be practically useless when it comes to, as they say, sponsoring validity claims.”⁸ As I shall argue in a moment, this is not at all the case. I am afraid, however, that if they do not work, interpretation can never be a “serious business, ” for the kind of norms Hirsch advocates—ones which have “the capacity to enforce practical decisions” (VI, p. 203)—simply do not exist. Just as in science (as the postpositivists have come to realize) there can be no experimentum crucis which decisively verifies (or even falsifies) a theory, so there can be no “ruthlessly critical process of validation” (VI, p. 206) in interpretation. The notion that one can “test” interpretations and subject them “to scrutiny in the light of the relevant evidence” such that “objective conclusions can be reached” (VI, p. 206) is a purely utopian notion. There can be no science of interpretation. This, however, does not mean that interpretation cannot be a rigorous (if not an exact) discipline, an art in the proper sense of the term, and that one cannot rationally evaluate interpretations.

The problem with Hirsch is that he appears not to see that there is a difference—a most important one—between demonstrative or theoretical reasoning and persuasive or pratical reasoning, and that it is the rules of the latter, not of the former, which can and ought to serve as the method of interpretation. Hirsch is in a sense trying to base a methodology of rational behavior (textual interpretation, in this case) on a calculus of probabilities.⁹ For the true humanist, however, this is unacceptable because it amounts to an uncritical acceptance of a scientific prejudice which is itself unacceptable, the prejudice, namely, that there are two and only two forms of knowledge which can lay claim to truth: absolutely certain, apodictic, fully demonstrated knowledge (of the sort found in mathematics) and, for the rest, knowledge based on the “objective” measurement of probability (conjectural or stochastic knowledge). The positivistic hermeneut fails to realize that besides this kind of decision theory there also exists what could be called a logic of argumentation which is something else altogether, something which is not at all concerned with the quantitative determination of probability.¹⁰

What are some of the principal traits of the theory of argumentation I am alluding to? What is the conception of rationality involved here?

Like demonstrative reasoning, persuasive reasoning is a method for arriving at what people like to call “the truth.” What “the truth” means here is, however, agreement or consensus as to what shall be held to be true. Persuasive reasoning is a means for, as Peirce would say, the “fixation of belief,” for, that is, coming to an agreement with an interlocutor as to the legitimacy of a decision. Persuasive or practical reasoning is what justifies or legitimates decisions.

In arguing persuasively for a given interpretation one adduces reasons, i.e., one appeals to certain commonly or widely accepted principles and maintains that interpretation 1, as opposed to interpretation 2, more faithfully embodies such principles. Perceiving this conformity is always a matter of discretion, which is to say that agreement is never coerced (as it is in deductive reasoning). Assent always remains a free choice, but it is not an arbitrary, or irrational, choice, since it is “reasons” which serve to motivate the choice (these reasons forming, as Cardinal Newman might say, the “grammar of assent”).

This allusion to Newman and to the question of the rationality of religious faith is not accidental, since the theory of argumentation has played a key role in this area. The traditional theological problem of faith and reason arises out of the need to argue for religious belief. The typical question theologians have had to deal with is: Can choosing or opting to believe in Christianity be rationally justified, or does it amount merely to an irrational “leap of faith”? The position of Saint Augustine was that, although the truth of the Catholic faith cannot be demonstratively proven, nevertheless reasons can be given for faith after its acceptance and reasons can induce us to believe prior to our believing. In short, in practical reasoning reasons influence but do not determine; they justify one’s decisions but do not demonstrate the truth or validity of them.¹¹

It should be noted as well that the principles one appeals to in practical reasoning need not, like logical princples, be “universally” and “eternally” valid. They need not themselves be demonstratively proven or apodictically selfevident, such as to be binding on all men at all times and in all places. All that is required of them is that they be generally accepted by those with whom one is arguing (by one’s audience, to speak in terms of rhetoric). Practical or persuasive reasoning both aims at agreement on a specific subject and presupposes a prior agreement on certain basic norms. Thus there does exist here a kind of universality, that is to say, a communality. Practical reasoning is as impossible in a normative vacuum as it is incompatible with the supposition of a determinate and eternal system of norms which would be universal in the sense of being always and everywhere the same for everyone. This is an important point, for it allows one, unlike Hirsch, to make sense of Gadamer’s remark: “Being bound by a situation does not mean that the claim to correctness that every interpretation must make is dissolved into the subjective or the occasional” (TΜ, p. 359).

Practical reasoning, therefore, bases itself on recognized, commonly accepted norms and seeks, through argumentation, to legitimate new, concrete decisions. This is not a deductive procedure, nor are the norms in question simply “applied,” for, and this is a most important point, the decisions one makes tend to have a retro-effect on the norms themselves. Like juridical decisions, decisions argumentatively arrived at modify in an ongoing way the norms themselves.¹² Canons simply express what, over the course of time, has become “canonical,” i.e., accepted and authoritative. Canons are simply what have managed to impose themselves as normative and binding for a given, historical community of human beings. Thus temporality is as essential an element in argumentative reasoning as timelessness is in logical proof. Just as the norms or argument need not be “universally” valid, so they need not be “eternally” true (or, as Hirsch would say, “genuinely stable”).

In answer to Hirsch (or his likes), I would say, therefore, that the principles I have listed above are adequate and sufficient for performing what Hirsch himself insists is the most important task in interpretation: arbitrating between conflicting interpretations. When phenomenological hermeneutics rejects the notion that there is an interpretation which is “correct in itself” and when it maintains that hermeneutical understanding is historical through and through, it is not providing a license for subjectivism, arbitrariness, or irrationality. One is still left with a perfectly good, intersubjectively valid basis for arguing for or against various interpretations. To argue, for instance, that interpretation 2 is more coherent, more comprehensive, and so on, than interpretation 1 is ample and sufficient reason for deciding to accept it and to take it as “true.” In answer to Hirsch I would say, with Gadamer: “Thus there is a criterion here too. . . . this places hermeneutical work on a firm basis” (TΜ, p. 238).

It should be noted, moreover, that these norms are as pertinent to assessing the worth of already worked-out interpretations as they are to informing critically one’s own understanding in working out an interpretation. The analogy with ethical norms holds valid here, too: Just as ethical norms enable us to assess rationally the value of accomplished deeds, so also they guide us when we ourselves are faced with the necessity of choosing or acting. For, as the great rhetorician Isocrates pointed out, “the same arguments which we use in persuading others when we speak in public, we employ also when we deliberate in our own thoughts.”¹³ Thus the model of practical reason does away with yet another of Hirsch’s false antinomies, namely, his radical opposition between the forming of an interpretation and the assessment of it, between what he calls the “divinatory” moment in interpretation and the following, “critical” moment. Like other positivists, Hirsch says that arriving at an understanding is “unmethodical,” is a mere “imaginative guess,” whereas assessing interpretations involves a “high intellectual standard” and is a matter of “testing” (VI, p. x), and he contrasts “the whimsical lawlessness of guessing with the ultimately methodical character of testing” (VI, p. 204). This leads him to assert: “There can be no canons of construction, but only canons which help us to choose between alternative meanings that have already been constructed from the text” (VI, p. 204). There is, he says, no “method or model of correct interpretation” but only “a ruthlessly critical process of validation” (VI, p. 206). However, once we abandon the model of demonstrative reasoning as being inappropriate to hermeneutics and adopt in its stead that of persuasive or practical reasoning and view interpretation in the light of the theory of argumentation, it can be seen that arriving at an understanding is, in its own way, as methodical as the assessing of it. For a phenomenological hermeneutics, the working out of interpretations is not an altogether different process from that of evaluating or defending them; both involve methodical reasoning, although in neither case is method reducible to mere technique.

I have obviously gone beyond Gadamer in defending actively methodical reasoning in hermeneutics and in suggesting a number of specific criteria. What I wanted to show in so doing is that phenomenological hermeneutics can allow for method (the fact that I was able to illustrate a number of my “rules” with texts from Gadamer indicates that methodical norms are not ruled out by his hermeneutics) and that it can make of interpretation a serious business.¹⁴ Now if it can do this while at the same time ridding us of the highly objectionable conceptual baggage that Hirsch has imported into the humanities from positivism, if, in short, it can make the humanities rigorous while preserving their integrity vis-à-vis the natural sciences, then this seems to me to be ample reason for accepting it as a more satisfactory methodology and theoretical basis for interpretation.

One of the central theses of phenomenological hermeneutics, derived from its theory of understanding, is that, like understanding, the object of understanding, textual meaning, has a temporal mode of being, which is to say that it is ever in the process of becoming and thus (like Merleau-Ponty’s “Being”) never fully is. It is not something fully determinate, unchanging, timeless, eternally the selfsame. I submit that this is the main reason why the model of interpretation should be sought in practical reason and not in theoretical reason, for the proper object of the former is what is contingent and changing while the proper object of the latter is what is timeless and immutable (the paramount instance of the latter being the objects of mathematics and the superlunary bodies). As Aristotle pointed out, ethics is concerned with things which “admit of much variety and fluctuation of opinion, so that they may be thought to exist only by convention and not by nature”; it is concerned with “things which are only for the most part true and with premises of the same kind to reach conclusions that are no better.” And as he goes on to say, we must not, accordingly, expect from it scientific proofs, as when in mathematics we deal with things that exist of necessity.¹⁵

Moreover, unlike theoretical reason, whose purpose is to lead one to an insight into what simply is and which, in principle, exists as what it is independently of the knowing subject, practical reason is concerned with all those situations where one must make a choice, produce something, or decide on a course of action, the outcome of which is contingent in that it depends, precisely, on the subject oneself. This is another reason why practical reason should be taken as the model for interpretation, for interpretation is always a creative business; it is, as Gadamer says, “not a reproductive but always a productive activity.”

The great advantage of taking practical reason as a model is that when one does so one is no longer plagued with the false antinomies which drive such people as Hirsch into an uncritical objectivism. At the same time, the “inner conflicts and inconsistencies” (VI, p. 247) Hirsch thinks he detects in Gadamer’s theory disappear. It then becomes perfectly reasonable to maintain that, while the meaning of the text cannot be equated with “what the author meant,” the interpreter cannot, for all that, simply project his or her own meaning onto the text. It becomes perfectly reasonable to say that while no interpretation can ever be shown to be the “correct” one, some interpretations are, nonetheless, clearly better than others. With practical reason as our model, it becomes perfectly reasonable to assert that while (to use Hirsch’s words) there is no “genuinely stable norm,” we can nevertheless make “a valid choice between two different interpretations” and are not “left with the consequence that a text means nothing in particular at all” (VI, p. 251).

In conclusion, therefore, I submit that while phenomenological hermeneutics cannot provide a logic of validation and a procedure for arriving at “correct” decisions in interpretation, it nevertheless does not afford a license for arbitrariness, for it can allow for a logic of argumentation in the light of which rational decisions can be made. Interpretation should be viewed as a mode of practical reasoning and of persuasive argumentation. The model for interpretation should therefore be looked for in the theory of argumentation and not in what is called the logic of (scientific) explanation. It is not to science but to rhetoric or the theory of persuasive argumentation that interpretation should look for its theoretical and methodological grounding. For what, throughout its long history, as long as that of science itself, to which it has always opposed an alternative conception of rationality, rhetoric has taught is that while in the realm of human affairs and action we can never be absolutely certain of anything, we can nevertheless have legitimate grounds for believing that some things are clearly better than others.

As I was able to illustrate a number of my “rules” with Gadamerian texts, so, to finish, let me appeal to Gadamer in defense of the central thesis I have here been arguing for:

Notes

This paper was originally presented to the Faculty of Philosophy of the University of Ottawa, February 8, 1980, and was subsequently read at the annual meeting of the Canadian Philosophical Association, Montreal, June 3-6, 1980.

1. E. D. Hirsch, Validity in Interpretation (New Haven, Conn.: Yale University Press, 1967) (henceforth cited as VI); H.-G. Gadamer, Truth and Method (New York: The Seabury Press, 1975) (henceforth cited as TM). A number of subsequent essays on hermeneutics has been published in Gadamer, Philosophical Hermeneutics, trans. D. Linge (Berkeley: University of California Press, 1976) (henceforth cited as PH).

2. “Conflicting interpretations can be subjected to scrutiny in the light of the relevant evidence, and objective conclusions can be reached. . . . Devising subsidiary interpretive hypotheses capable of sponsoring probability decisions is not in principle different from devising experiments which can sponsor decisions between hypotheses in the natural sciences” (VI, p. 206). “ . . . the much-advertised cleavage between thinking in the sciences and the humanities does not exist. The hypothetico-deductive process is fundamental in both of them, as it is in all thinking that aspires to knowledge” (VI, p. 264).

Of Hirsch’s position Paul Ricoeur has written: “Some, opposing to the work of Gadamer that of E. Betti, in whom they see the true heir of Dilthey, develop hermeneutics in the direction of a logic of proof; thus E. Hirsch . . . basing himself on the logic of probability of Keynes, Reichenback and K. Popper . . . develops a theory of ‘validation’ of the probable meaning, corresponding in hermeneutics to the theory of ‘verification’ in the empirical sciences. In this way, contact is reestablished between hermeneutics and logical positivism.” Main Trends in Philosophy (New York: Holmes and Meyer, 1979), P. 269.

3. See my article, “Eine Kritik an Hirschs Begriff der ‘Richtigkeit,’ ” in Seminar: Die Hermeneutik und die Wissenschaften, ed. H.-G. Gadamer and G. Boehm (Frankfurt: Suhrkamp Verlag, 1978), pp. 393-425 (reprinted in this volume as essay 1). Boehm is referring to this ongoing debate when he writes in the introduction (p. 8): “Die Debatte, die schon sehr früh zwischen E. Betti, E. Hirsch, später Thomas M. Seebohm u. a. auf der einen und Hans-Georg Gadamer auf der anderen Seite ausgetragen wurde, hatte die Forderung zum Inhalt, das Verhaltnis zwischen Wahrheit und Methode so zu gestalten, dass, in diesen Fall, der ‘Objektivitätsanspruch der philologisch-historischen Methode’ gewahrt Bleibt. Der bislang angedruckte Beitrag von G.B. Madison nimmt diese Einwände nochmals auf.”

4. He does so when he says, for instance, that the work of hermeneutics “is not to develop a procedure of understanding, but to clarify the conditions in which understanding takes place. But these conditions are not of the nature of a ‘procedure’ or a method, which the interpreter must of himself bring to bear on the text . . . ” (TM, p. 263). And: “Our enquiry started from our dissatisfaction with the modern concept of methodology” (TM, p. 421). Gadamer has all along insisted that his primary concern is not to lay down rules for interpretation but, more basically, to clarify the actual conditions under which understanding occurs. “My revival of the expression ‘hermeneutics,’ with its long tradition, has apparently led to some misunderstandings. I did not intend to produce an art or technique of understanding, in the manner of the earlier hermeneutics. I did not wish to elaborate a system of rules to describe, let alone direct, the methodical procedure of the human sciences. Nor was it my aim to investigate the theoretical foundation of work in these fields in order to put my findings to practical ends. . . . My real concern was and is philosophic: not what we do or what we ought to do, but what happens to us over and above our wanting and doing” (TM, p. xvi). He expressed the same point in a letter to Hirsch’s predecessor, E. Betti: “Fundamentally I am not proposing a method, but I am describing what is the case. . . . I consider the only scientific thing is to recognise what is, instead of starting from what ought to be or could be. Hence I am trying to go beyond the concept of method held by modern science (which retains its limited justification) and to envisage in a fundamentally universal way what always happens” (TM, pp. 465-66).

5. According to Ricoeur, in Gadamer’s work “those conclusions of Heideggerian philosophy that are not only anti-psychological but also anti-methodological are developed. There must be a choice between truth and method: whence the title Wahrheit und Methode. In this work, the author contrasts the power of truth which lies concealed in comprehension with any methodology, any technology which might be ascribed to the sciences of the mind. . . . This orientation of hermeneutics, firstly anti-psychological, secondly anti-methodological, ushered in a crisis within the hermeneutic movement; in correcting the ‘psychologizing’ tendency of Schleiermacher and Dilthey, ontological hermeneutics sacrificed the concern for validation which in the founder’s world made up for the element of divination.” Main Trends in Philosophy, pp. 268-69.

6. See the following texts: “ . . . understanding is never subjective behavior toward a given ‘object’, but towards its effective history—the history of its influence; in other words, understanding belongs to the being of that which is understood” (TM, p. xix); “In fact history does not belong to us, but we belong to it” (TM, p. 245); “Understanding is not to be thought of so much as an action of one’s subjectivity, but as the placing of oneself within a process of tradition, in which past and present are constantly fused. This is what must be expressed in hermeneutical theory, which is far too dominated by the idea of a process, a method” (TM, p. 258); “What we mean by truth here can best be determined again in terms of our concept of play. . . . It is worth recalling here what we said about the nature of play, namely that the attitude of the player should not be seen as an attitude of subjectivity, since it is, rather, the game itself that plays, in that it draws the players into itself and thus itself becomes the actual subjectum of the playing. What corresponds to this in the present case is neither play with language nor with the contents of the experience of the world or of tradition that speak to us, but by the play of language itself, which addresses us, proposes and withdraws, asks and fulfills itself in the answer” (TM, p. 446); “It is not really we ourselves who understand: it is always a past that allows us to say, ‘I have understood’ ” (PH, p. 58).

The net impression one gets from texts such as these is that it is not, properly speaking, we who understand, that it is, rather, history or the tradition which understands itself in and through us. Thus it must, I think, be admitted that Gadamer is not simply undertaking what could be called a fundamental ontology of understanding which would be completely without prejudice to the question of correct understanding, i.e., method. The fact is that Gadamer incorporates into his hermeneutical theory a number of Heideggerian elements and that his concept of “effective history” parallels to a considerable extent Heidegger’s notion of the “essential destiny of Being.” For Heidegger, it is not man, the individual human subject, who thinks or speaks; it is, rather, language which speaks itself in man and Being which thinks itself in man, such that man’s thoughts about Being are the thoughts of Being, thoughts belonging to Being, thoughts that Being has of itself, such that the history of human thought is actually the history of Being’s self-manifestation (or self-concealment). When Gadamer speaks of “the ‘language of things’ ” and writes “Is not language more the language of things than the language of man?” (PH, pp. 76-77), one is reminded of nothing so much as Heidegger’s statement at the end of his Letter on Humanism: “Language is thus the language of Being, as the clouds are the clouds of the sky.”

Gadamer has himself clearly remarked on his indebtedness to Heidegger: “The role that the mystery of language plays in Heidegger’s later thought is sufficient indication that his concentration on the historicity of self-understanding banished not only the concept of consciousness from its central position, but also the concept of selfhood as such. For what is more unconscious and ‘selfless’ than that mysterious realm of language in which we stand and which allows what is to come to expression, so that being ‘temporalizes itself’? But if this is valid for the mystery of language it is also valid for the concept of understanding. Understanding too cannot be grasped as a simple activity of the consciousness that understands, but is itself a mode of the event of being. To put it in purely formal terms, the primacy that language and understanding have in Heidegger’s thought indicates the priority of the ‘relation’ over against its relational members—the I who understands and that which is understood. Nevertheless, it seems to me that it is possible to bring to expression within the hermeneutical consciousness itself Heidegger’s statements concerning ‘being’ and the line of inquiry he developed out of the experience of the ‘turn’. I have carried out this attempt in Truth and Method” (PH, p. 50).

It is one thing to say, as Gadamer does say, that (1) the “prejudices” that shape an individual’s understanding are not “subjective” and “individualist” in that they are not freely adopted by him and are not under his conscious control but proceed from the communality that binds him to a tradition and that they constitute the historical reality of his being; it is quite another thing to say that (2) the historical consciousness of the individual is in fact the consciousness of history, where the “of” means: the consciousness that history has of itself. If Gadamer has little to say about method, it is, I think, no accident but is due to the fact that under the influence of Heidegger (as well as of Hegel) he tends to confound assertion (1) with assertion (2).

The issue at stake here, that, in effect, of the importance of “subjectivity,” could be approached in another way, by a consideration of the phenomenological project. Phenomenology is the description of the various structures of experience as they present themselves to the conscious, reflecting subject. Phenomenology thus entails the methodological primacy of consciousness. This is not to say that phenomenology need necessarily be a “philosophy of consciousness” or that it tends toward subjective idealism. For in this case the method of approach is without prejudice to the object approached. Phenomenology can even allow that there is something like an unconscious (in the Freudian sense), but it would insist that if we are to know anything about this unconscious it can only be by means of consciousness. Husserl’s error was to have “ontologized” this purely methodological primacy of consciousness. Because he was guided by the rationalist ideal of an absolute science and thus needed a fundamentum inconcussum on which to erect it, he was led to assert that consciousness is not only one mode or region of being (Bewusst-sein) but is itself absolute, is that outside of which there is nothing (omnitudo realitatis). Phenomenology need not, however, follow this path, which is indeed that of subjectivism (albeit of a nonpsychologistic, transcendental sort). For it could be perfectly well maintained—and on a strictly phenomenological basis (i.e., while granting methodological primacy to consciousness)—that being, not consciousness, has absolute priority, that being “transcends” irreducibly consciousness and is thus absolutely irreducible to it, that consciousness itself “belongs” (in the Heideggerian sense) to being. None of this is incompatible with holding to the methodological primacy of consciousness. In fact, it is precisely “by means of” consciousness that the ontological primacy of being is discovered and asserted (this was, precisely, the tactic of the later Merleau-Ponty). It could even be said that consciousness is the universal measure of all things (including being itself)—in the sense that it is the only means for sounding out the unfathomable depths of being.

Thus, in regard to hermeneutics, recognizing the ontological primacy of tradition and effective history does not necessarily mean that one must downplay the importance of method and subjectivity (although it does of course mean rejecting the modern, subjectivistic conception of method). Emphasis on method need not give rise to what could be called “methodologism,” i.e., subjectivism. Indeed, if we are to let the thing itself speak (cf. Heidegger’s “letting Being be”) and not simply project onto it our own subjective fantasies and desires, then we need to discipline ourselves and must make a consistent (methodical) and conscientious (which means conscious) effort to control ourselves in our dealings with the thing. The name for this discipline is method.

7. For instance: “Certainly, there is here too for all understanding a criterion by which it is measured and a possible completion. It is the content of the tradition itself that is the sole criterion and expresses itself in language” (TM, p. 430).

8. Hirsch writes: “It may be set down as a general rule of interpretation that there are no interpretive rules which are at once general and practical. A truly general rule will fail . . . to guide us in a specific case, and a practical rule—that is, a specific and concrete one—cannot be truly general: it may or may not lead to the valid conclusion. . . . Every practical rule of interpretation has an implicit ‘unless’ after it, which means, of course, that it is not really a rule” (VI, pp. 202-203). Whence he concludes: “The notion that a reliable methodology of interpretation can be built upon a set of canons is thus a mirage.”

9. See, for instance, VI, p. 236: “ . . . no one can establish another’s meaning with certainty. The interpreter’s goal is simply this—to show that a given reading is more probable than others. In hermeneutics, verification is a process of establishing relative probabilities.”

10. Thus the reason why Hirsch thinks that there can be no methodological canons of interpretation (see note 8 above) is because he confuses the rules of practical reasoning with those of demonstrative reasoning. This is fully apparent in the following text, where he equates “practical interpretive canons” with mere “preliminary probability judgments”: “ . . . since all practical interpretive canons are merely preliminary probability judgments, two consequences follow with regard to their intelligent application. First, the canon is more reliable the narrower its intended range of application. Second, since any interpretive canon can be overturned by subsuming the text under a still narrower class in which the canon fails to hold or holds by such a small majority that it becomes doubtful, it follows that interpretive canons are often relatively useless baggage. When they are general, they cannot compel decisions, and even when they are narrowly practical, they can be overturned” (VI, p. 203).

Hirsch’s trouble is that he wants to reduce methodology to technology and is looking for practical rules which can “compel” decisions. He therefore fails to see that the rules of practical reason have an altogether different function. They are not binding rules capable of compelling a choice between different interpretations but are, instead, norms enabling one to argue for a given interpretation and to justify rationally one’s choice.

11. The history of Catholic thought in the Middle Ages is a veritable but much underexploited mine of information for theorists of argumentation, for it was in fact an immense workshop in which thinkers were continually attempting to arrive at a satisfactory solution to the problem of rationality in belief. What makes it so interesting is that Catholic orthodoxy seemed to demand the impossible: The truths of revelation must be rationally defensible but they must not be claimed to be rationally demonstrable or discoverable by reason alone. The twin error to be avoided was irrationalism and gnosticism. The problem faced by phenomenological hermeneutics today is strictly homologous: Interpretation must be shown to be a rational discipline, but it cannot be allowed to be a science. If, then, it is not a science, but if it nevertheless is a rational discipline, what kind of methodology is appropriate to it?

12. Herein, in the fact that decisions form precedents, lies the rationale for rule (h), which prescribes that a given interpretation should (like a legal judgment) normally be in agreement with traditional interpretations.

13. Isocrates, Nicocles, 8.

14. I should emphasize that I put forward these “rules” as suggestions only. It would be the task of a general, philosophical hermeneutics to work out a proper organon of methodological norms. The procedure to be followed in doing so would be twofold. Similar to the twofold approach of Aristotle in the Nichomachean Ethics, one would, on the one hand, attempt to derive practical norms from the general theory of understanding, seeking to determine what exactly it is one expects from any given interpretive understanding, given the way, according to the theory, understanding actually does work. In other words, by means of a kind of transcendental (or phenomenological) reflection on human understanding one would ascertain what norms or criteria are actually presupposed by any interpretation which seeks to be believed. On the other hand, on would, by way of complement to this approach, ascertain the norms which are in fact generally accepted by the overall community of interpreters.

15. Aristotle, Nicomachean Ethics, 1094 b 15-27 (Ross translation).