pdf Download PDF

Ronald Dworkin and the Classics1

How shall I interpret a classical text? However I interpret it, someone else will interpret it differently.

Interpretation of interpretation: the act of interpretation has a way of turning into the subject of interpretation itself.

Ronald Dworkin (1931–2013) is widely accepted as one of the most important Anglo-American legal philosophers of the past forty years. Besides his numerous writings on legal philosophy, he also wrote extensively on interpretation in general: how the concept should be understood and what it should include. Among the different genres of interpretation attached to different fields of knowledge, he gave particular attention to literary interpretation and pointed to the analogies between legal and literary interpretation: he saw both as belonging to the same very broad interpretive genre of “collaborative interpretation.”2 He believed that thanks to these [End Page 93] analogies, legal interpretation could benefit from a study of the principles that guide literary analysis. His long-term preoccupation with the practice of interpretation led him further to develop his own interpretive philosophy, which (as will become evident below) to some extent emerged from his fruitful engagement with literary criticism and hermeneutics. In this paper, his ideas are imported back into a theoretical discussion of how to interpret literature; this might seem an odd enterprise, but it is justified by the fact that Dworkin usefully reshapes and advances some of the concepts of literary theory.

Although Dworkin’s ultimate focus was the exegesis of law, his interpretive philosophy—which he often applied to examples drawn from English literature—constitutes an important contribution to the ongoing debate about literary interpretation and to the so-called “literature and law movement” (which studies the intersections between literary and legal hermeneutics). This movement, and, indeed, Dworkin’s name, are familiar to those working on literary interpretation theory and to some English literature critics,3 but they seem so far to have gone unnoticed by classicists, although classics is no stranger to fierce theoretical debates about interpretation.

My aim is to introduce Dworkin’s philosophy to classicists: I shall present his theoretical thinking on the nature of interpretation (the aesthetic hypothesis, the chain novel concept) and on the place of author and interpreter in the interpretive process, as well as his main ideas concerning what is involved in the act of interpretation (the role of presuppositions and interpretive stages). I shall not engage (let alone take sides) in the philosophical debate regarding interpretation, especially regarding the notions of objectivity and truth—though I shall briefly refer to these in the context of presenting Dworkin’s interpretive concepts. Admittedly, the literature and law movement is marked by controversy and heated debates,4 [End Page 94] and Dworkin’s own conception of interpretation is far from uncontroversial. Despite this, his interpretative philosophy is relevant and potentially rewarding for our reading and understanding of ancient texts—not least because it can provide a useful angle from which to review the diverging interpretative results obtained by contemporary applications of literary theory to ancient texts.5


Dworkin hints at the theoretical problems surrounding the nature of interpretation at various places in his work, but here we shall focus on three book chapters in which he more specifically implicates literary interpretation: 1) “How Law is Like Literature” (1985.146-66), 2) “Interpretive Concepts” (1986.45–86), and 3) “Interpretation in General” (2011.123–56). The issues addressed are familiar to classical scholars who resort to literary theory for the formulation of and solutions to interpretive questions.

Literary theory is, of course, a general term that encompasses a multitude of very different and sometimes contradictory standpoints, but despite their differences, these standpoints have much in common with each other and with Dworkin’s interpretive philosophy: they all deal with fundamental problems concerning the nature and meaning of literature and pose a number of common questions. In his useful introduction to literary theory and the classics, Thomas Schmitz (2007.2) provides a checklist of questions posed by literary theory; among these, we shall see the following also being addressed by Dworkin: How does a literary text convey its meaning? Why is it that literary texts provoke a great number of different, often irreconcilable, interpretations? How can we guarantee that a text does not mean just anything or nothing at all? And what is literature’s relation to the world by which it is surrounded? To these we must add questions [End Page 95] relating to the judging of quality and beauty raised by philosophical aesthetics and literary criticism and often neglected in classics.6

Dworkin (2011.123) takes as the starting point of his interpretive philosophy the dissent among literary scholars and art critics about the essence of interpretation. “Critics in (art and) literature count, as interpretations, such very different claims as that the value of art lies in moral instruction, that Piero della Francesca’s The Risen Christ is a pagan rather than a Christian painting, and that Shylock’s Jessica betrayed her father because she hated being a Jew.” He is acutely aware that even within one and the same literary or artistic genre, interpretive disagreement is the rule (2011.127, on literary interpretation).

Writing more specifically on the interpretation of classical texts, Malcolm Heath (2002.39) calls interpretive disagreement “a regrettable necessity.” Indeed, the nature of the object interpreted seems to explain this. As far as ancient Greek and Latin literature is concerned, we need to keep in mind that it is a diverse field that encompasses a variety of types of texts with an immense geographical and chronological span. The birth and development of diverse interpretive models, often systematized as literary theories, is partly explained by the fact that different types of texts invite different approaches while they discourage or exclude others (Hitchcock 2008.xii and Schmitz 2007.11). This partial applicability implies that each theory has its “blind spot”: focusing on particular aspects of its object of study and neglecting others. At the moment, there is “no grand unifying theory that brings into sharp focus all things classical” (Hitchcock 2008.xii).

Diversity in interpretation, nonetheless, is not wholly explained by the differences between types of texts. Heath remarks (2002.40) that the coexistence of different interpretive practices “does not resolve the issue of principle.” As Dworkin implies, different interpretations are often the result of applying different interpretive principles—different theories of what interpretation is. It is worrying that different theories suggest such different notions of interpretation and meaning: those notions, for example, in structuralist analysis that help us understand the layout and deeper [End Page 96] structure of stories and plots (but do not necessarily lead to the interpretation of meaning per se) and in deconstruction, which goes so far as to almost deny the possibility of interpretation (every reading is potentially a misreading). But does this imply that there will never be a unifying interpretive theory? According to Dworkin, not necessarily. He suggests that what we need is a uniform understanding of interpretation and notes (1986.45ff.) that genuine disagreement among interpreters is possible even when they all agree on what interpretation is. He then declares his interest in “arguments which offer some sort of interpretation of the meaning of a work as a whole” (1985.149). (Admittedly, as the above examples of structuralism and deconstruction show, literary theory does not always facilitate the extraction of this type of meaning.) Based on meaning thus conceived,7 Dworkin goes on to sketch the rough outlines of a unifying theory of interpretation in literature and law.


Dworkin’s earliest essay (1985) begins with the premise that interpretation (described roughly as a “technique of analysis” or a “mode of knowledge”) contains elements of both description and evaluation but is different from both. Interpretation of literature signifies discovering the meaning of a text,8 where, as already mentioned, meaning is conceived of as that of a work as a whole, not that of a word or phrase, although the meaning of a particular word may bear upon “the larger matters.” This definition of interpretation is strongly reminiscent of the conundrum of the “hermeneutic circle”:9 the mutual interpretive relationship between a text as a whole and its individual parts; Dworkin’s debt to hermeneutics is evident here.

Interpretation of a work as a whole is normally based on assertions about characters, about events and theme, or about the work’s point or tone; for example, we may assert that the Iliad is a poem about war, [End Page 97] or about glory, or about love and loss, and interpret it accordingly. These constitute “interpretive claims” and have practical implications, e.g., they can affect the manner and content of teaching classics to undergraduate students, the way an ancient tragedy is staged or made into a film, or more generally, they may guide our understanding of a literary genre or a particular aspect of an ancient civilization (Dworkin 1985.149). Interpreters of ancient texts (like interpreters of all texts) often bring forth conflicting interpretive claims: they sometimes disagree about themes, and often about characterization, underlying meanings, and a work’s point. For Dworkin, the reason why interpreters arrive at such different claims is what needs explaining. He proposes his “aesthetic hypothesis” to identify the cause of interpretive disagreement in literature and art: “An interpretation of a piece of literature attempts to show which way of reading (or speaking or directing or acting) the text reveals itself as the best work of art. Different theories or schools or traditions of interpretation disagree on this hypothesis, because they assume significantly different normative theories about what literature is and what it is for and about what makes one work of literature better than another” (1985.149; my emphasis).

A theory of “good” literature is perhaps the most difficult to agree on, and attempts at providing objective criteria for the aesthetic judgement of literature are far from uncontroversial.10 This is because a text’s reception is uncontestably rooted in the particular time and place, the particular tastes and ideologies of its readership.11 The process of interpretation is largely and inevitably guided by individual preconceptions and tastes (cf. in particular the critique of historicism for neglecting them12)—we shall take a closer look at these preconceptions and tastes in the next section. Whilst admitting that these may differ among interpreters (and even among different interpretive acts of the same interpreter), Dworkin reposes the question, [End Page 98] what is interpretation, and redefines the purpose of the interpretive process in a holistic manner, thus assuming a single abstract goal for all the diverging approaches. The introduction of a single common interpretive purpose, such as the one proposed by the “aesthetic hypothesis” (all interpretation strives to make an object the best it can be), is not without its benefits. It can provide food for thought and guidance to classicists wishing to review and reevaluate the merits and demerits of the theorizing of the last century; it arguably also suggests a new direction in interpretation.


The pursuit of an interpretation that makes a literary work “the best it can be” is limited by certain constraints or parameters that frame the interpretive process (Dworkin 1985.150–51). These will be familiar to classicists applying literary theory to literary texts; the constraints (especially the second and third parameters discussed below and page 100), characteristically correspond to what Schmitz describes (2007.2) as “preliminary answers” to questions posed by literary theory13 or to the conscious or subconscious presuppositions that an interpreter brings to his or her reading of the texts.14

One self-evident parameter of this kind is the interpreter’s view of the identity of a text, meaning his or her theoretical beliefs about the content and the type or nature of a canonical text. This is a troubled territory for many ancient texts, as the long and complex history of textual criticism shows. Sometimes interpreters interpret ancient Greek and Latin texts on the basis of translations, with no or only partial understanding of the original language. Translation is itself a genre of interpretation and depends on the translator’s beliefs about the language and the behaviour of its speakers (Dworkin 2011.147–49).

Another such parameter is the interpreter’s set of opinions about the coherence or integrity of a work of literature: “An interpretation cannot make a work of art more distinguished if it makes a large part of the [End Page 99] text irrelevant” (Dworkin 1985.150). Dworkin warns of the danger of “reinterpreting” texts as something more ambitious than they are if a critic neglects the constraint of integrity. This concern appears relevant to the interpretation of a number of ancient texts whose uniformity and homogeneity has been questioned; for example, Herodotus’s Histories have been interpreted both as a uniform work (with a unifying concern with morality and the meaning of life) and as a collage of different themes. In addition, our reading of Herodotus is affected by the factor of orality (his text has arguably emerged from an oral narrative background)—a factor that is also central in the interpretation of Greek antiquity’s most famous texts, the Homeric poems. These poems not only preserve traces of an originally oral composition, but were clearly also intended for oral episodic performance (their length would make performance of whole poems impractical). These considerations do not invalidate the principle of integrity but dictate that it be adjusted to the special nature of the texts that are being interpreted. Ancient authors show a relative lack of interest in the concept of unity; they are less concerned with thematic integrity and interrelatedness, and more with formal adherence to genre (Heath 1989.9).

But a broader and indeed most crucial parameter is the interpreters’ theories of art as reflected in their stance on questions like: “What makes a poem good?” “What is good literature?” “What counts as good art?” and so on. If, in the interpreter’s view, an instructive function is what defines a literary text of high quality—if, in other words, (s)he expects such a text to teach us about what the world and people are like—then a psychoanalytic reading or a gender-theory oriented reading will show why it is good art. If good literature is regarded as successful communication, then interpretation should focus on authorial intention (and Saussurean/ structuralist readings may also find a place). If a quality literary work is something that affects and stimulates recipients, then the focus should be on the audience, as is the case with reader-response criticism and narratology. (All of the above theories have been applied to classical texts.) The interpreter’s aesthetic theory is found at the roots of interpretation and is the combined product of his or her philosophical, psychological, social, and even religious interests and beliefs. A critic should not necessarily subscribe to one theory to the exclusion of others: there is not one unique function or point of literature. However, interpretation always “relies on beliefs of a theoretical character about identity and other formal properties of art and . . . about what is good in art” (Dworkin 1985.152).

How can the aesthetic hypothesis affect our thinking on a classical text? Let’s use Sappho fragment 1 (“Hymn to Aphrodite”) as an [End Page 100] example.15 Working from Dworkin’s interpretive presuppositions, and given the poem’s textual problems and difficult language, a basic task would be to decide which text or translation we should interpret; in other words, we need to establish the “best” text. This would involve, for example, choosing between thronos and throna for the second component of the first epithet applied to Aphrodite (l.1: poikilothron’, “enthroned” or “adorned”; see, e.g., Powell’s note 2007.51). Such choices are often affected by answers to interpretive questions, as in the old problem of the reading of the word etheloisa, fem.: “willing” (l.24), which was decided upon (among a number of other possibilities) in the nineteenth century on the basis of external evidence (including that from other Sapphic poems) about the poet’s interest in women (Williamson 1995.50–52). In order to appear as the best it can be, the poem would then need to be assigned to a lyric genre, and its setting and purpose would have to be considered.16 Would Sappho’s poem be the best it can be if it is thought to belong to a tradition of hymnic poetry, or if it is seen as a piece composed for public performance, or if it is interpreted as a “personal” love poem? Does the “best” fragment 1 communicate a religious feeling that is intended to be shared by friends or with an audience? Or is it a snapshot of a psychological state: the perhaps unconscious revelation of the depths of the archaic female self—a repository of feminist ideology?

We shall not attempt concrete answers to these questions, which would demand extensive discussion,17 but they help us better understand Dworkin’s aesthetic hypothesis. It may seem to have a banal aspect, especially insofar as it highlights “the connection between what you interpret and what you believe is worth interpreting.”18 More importantly, it is hard to imagine that all interpreters would ever agree on the same definitive answers to all questions. However, as Dworkin notes (1985.152–53), the value of the aesthetic hypothesis lies not in redefining interpretation as a notion—of which different theories may be taken to be competing conceptions—but [End Page 101] in shifting the emphasis from the apparently conceptual question, “what is interpretation?” to the much more clearly substantive question, “which interpretation would show this or that (body of) work(s) of literature as the best it can be?”

Although this will not always lead to an interpretation unequivocally accepted as the single correct interpretation, it is, at the same time, strikingly different from a subjectivism that holds that no analysis or theory is better than any other (because, by hypothesis, the desired interpretation is the one that makes a literary work the best it can be; see also Patrick and Scult 1990.85). So, for instance, on Dworkin’s aesthetic hypothesis, not all possible interpretations of Sappho fragment 1 are “correct”; interpreters need to establish interpretation(s) that suggest the best realisation(s) of the poem’s genre. The aesthetic hypothesis thus provides a theoretical tool for testing the strength of diverging interpretations that are to be regarded as small-scale substantive value judgements and as partial answers to the question of where the value of a literary work lies.

Eventually, the solidity or value of an interpretation cannot be tested except by deploying further value arguments (which is not an indication of some worrying circularity but an indication that there is no Archimedean point, i.e., no point outside considerations of human value, on which to base aesthetic judgments; cf. Dworkin 2011.100). We are simply justified in thinking a particular interpretation of Sappho fragment 1 (e.g., a reader-response reading, a feminist, or a psychoanalytic approach) to be well founded or true when we are justified in thinking that our arguments for holding it to be true are adequate arguments.

The aesthetic hypothesis is also related to Dworkin’s “responsibility theory of interpretation” (2011.142–43). Interpreters suppose that their critical efforts have a point, that they embody some value. For example, lawyers interpreting documents might disagree on specific laws but agree that the general purpose of the law is to serve justice. Similarly, the aim of literary criticism is to identify excellence in writing. Dworkin further stresses the need for each individual critic to recognise the underlying theoretical foundation of his or her interpretations; this foundation is the interpreter’s sense of his or her interpretive responsibility—in other words, his or her understanding of what interpretation should entail. We have seen that different theoretical foundations lie at the root of interpretive disagreement; we might add here that they also deserve a place at the beginning of a discussion for potential interpretative agreement.

The aesthetic hypothesis finally eliminates the sharp distinction between interpretation (“discovering the meaning of art”) and criticism [End Page 102] (“evaluating its success”). It thus highlights the tacit assumptions about quality that clandestinely guide all interpretation. On the whole, it appears to be capable of providing a solid basis for interpretive thinking, while the pursuit of the best interpretation will serve as a source of motivation.19


Dworkin proposed another challenging idea: there is a certain kind of partnership among art critics (and to some lesser degree—and with a slightly different tone—also between art critics and the artists themselves). No matter how we define and practice interpretation, our efforts are bound to be affected by those of previous interpreters. The interdependence of interpretations is expressed by Dworkin (1985.158–62) with the help of the concept of the chain novel: a novel produced by several authors who write seriatim, each picking up where the previous author left off. This process entails that each author in the chain, except for the first, interpret the work of the previous one(s). Much in the spirit of hermeneutics, Dworkin later spoke of the “traditions of literary criticism” (2011.142): all interpretations form part of a tradition (the historical context and storehouses of interpretation built by our predecessors), from which interpreters can never wholly escape. Interpretation is interpretive of the tradition in which it interprets— as Dworkin puts it (2011.131), it is interpretive “all the way down.”20

Chain novels have only a small role in the world of fiction writing,21 but the metaphor of the chain novel is useful in both legal and literary hermeneutics. The chain novel concept may apply to the reading of such ancient texts as the Homeric poems, whose composition involved more than one poet—poets who clearly depended on the work of predecessors. (We shall, however, stop a step short of calling the Iliad and the Odyssey “chain [End Page 103] poems”; ancient epic poets most probably did not depend on fixed poetic texts, and it would be impossible to reconstruct a “chain.”) Chain composition involves value judgements; in other words, it answers the question “which development (e.g., in plot and characters) would make the work the best it can be?” Clearly not all answers are equally good, hence, despite criticism, chain formulation is pertinent to the interpretative process.22

Dworkin uses this metaphor to refer to the role of precedent in judicial decision-making; judges are like literary authors because chain interpretation in law involves the task of (re)creating the law by interpreting it23 (especially for cases judged on indeterminate legal propositions). A classicist may remark that the task of modern judges shares similarities with that of the ancient scholars who were responsible for copying and preserving the ancient texts in the form that we have them today: these scholars occasionally tried to improve the texts by correcting any possible mistakes made by previous copyists; hence their copying involved a certain amount of interpreting.

Judges are also similar to modern literary scholars, both editors and critics: just as judges may depend on the decisions of previous judges that they interpret in order to decide certain cases, so scholars and critics depend on previous critics, whose editorial judgements and theoretical positions they interpret in order to form their own stance (cf. Dworkin 2011.130–31, 142). Admittedly, literary critics differ from judges in terms of what is expected of their interpretations: the pursuit of the best possible interpretation in literature allows room for more than one reading of a given text, while a judge must arrive at a single right interpretation of the law (cf. Gana 2003.320–21). But the efforts of both groups rely on precedent. The interdependence of interpretive approaches in ancient literature is indeed amply demonstrated in the history of literary theory (cf., e.g., the dependence of “new historicism” on Foucault or of Kristeva’s intertextuality theory on Bakhtin’s dialogism). Literary scholars interpret seriatim; the accumulation of interpretations arguably constrains the choices and freedom of subsequent critics.

The conception of interpretation as a chain enterprise allows us to contemplate the similarities between the role of the interpreter and the role of the author, between interpreting and creating: each contains a fair [End Page 104] amount of the other (Dworkin 1985.158).24 The following section about authorial intention will develop this idea. It will also offer proof that interpretations and interpretive theories are sensitive to normative disagreement about the value of the literature interpreted.


Dworkin invites us to review the role of authorial intention in literary interpretation. This is one of the greatest concerns of Dworkin’s interpretive philosophy and is addressed in all three of the essays discussed here. It is also part of his discussion of the “psychological state theory of interpretation,” which holds that “interpretive claims are made true . . . by facts about the mental states of one or more people” (Dworkin 2011.128ff.). Can facts about intentionality make an interpretive judgement true? Dworkin appears cautious: the aesthetic hypothesis forces us to reconsider the weight of the author’s intentions and realize that maybe what is valuable in literature cannot be limited to what an author intended to put there.25 Clearly, literary interpretation is a multi-layered process (Dworkin 1985.154–58 and 1986.51ff.). One needs to take into account the unsolved mysteries in the psychology of creation; Dworkin reminds us of the author’s cliché: characters seem to have minds of their own, and he rightly wonders whether this can be explained as a “subconscious intention” or “change of intention.” Also, a discussion of authorial intention cannot ignore this other level of intention: to create a work of art whose meaning is not fixed and hence is independent of its creator—obviously this type of intentionality is hard to pin down (Dworkin 1985.156–57). The theory that interpretive claims about literary texts might be made true by facts about the mental (even if not necessarily conscious) states of their creators has a further serious [End Page 105] limitation: it cannot be used in all interpretive genres—certainly not in historical interpretation, which cannot depend on the psychological states and thoughts of historical actors (Dworkin 2011.129).

As Dworkin notes (2011.150), the authorial intent theory was fashionable in the nineteenth century.26 The discovery of ancient authors’ intentions has indeed been a mainstay of classical scholarship for a long time. Today, however, authorial intention is mostly treated as unknowable, and not many critics would understand interpretation as the discovery of the author’s intention; the assumption is that interpretations are produced not by confronting the author’s self but the work he produced. The tendency to accord to the author no authority on the meaning of his work features in a number of theoretical approaches to literary interpretation: according to Paul Ricoeur, who developed a hermeneutic phenomenology that influenced narratology, the author is only the “first reader.” Another major contributor to narratology, Roland Barthes, spoke of “the death of the author”; the American current of reader-response theory known as New Criticism developed an interpretive approach that isolates the text from its surroundings and circumstances (close reading).27 The focus of the interpretive lens on readers is shared by such varied approaches as narratological, deconstructionist, and intertextual readings, and feminist and psychological analysis (Schmitz 2007.93).

The focus on readers is not universal: psychoanalytic approaches to the text claim to shed light on the “psychology of creation” (but this claim is far from uncontroversial). In 1967, Eric Hirsch argued that the author’s intentions should hold a central place in the interpretive process if the latter is to achieve determinacy and validity. There is also a new interest in authorial intention in literary theory (Burke 2010 and Jannidis 1999). Of course, it is hard to deny that aspects of the author’s self are reflected in his or her work. Heath (2002.59–97, esp. 83–94) sensibly defends a version of intentionality as “conditionally necessary if we wish to achieve certain [interpretive] purposes” (2002.60; he argues, for example, that the understanding of a classical text’s cultural and social context may involve [End Page 106] intentionalist questions). Knowable intentions include those that are relevant to the genesis of a literary work;28 for example, Pindar was commissioned to write songs with the intention of praising winners at athletic contests. But otherwise intentions are not fully determinate or stable,29 and in the end, authorial intention is mostly, and as far as possible, gleaned from the sense expressed in the texts (this is especially true for ancient literature, whose authors are far removed from us in time), hence there is a risk of being caught in a vicious interpretive circle.

Dworkin offers an alternate way of viewing the place of the author: he suggests that what is reflected of the author is his interpretive beliefs (1985.155–58, 1986.51ff.). The author inevitably interprets as he creates and may form new beliefs about the story and characters he has created by interpreting his own work. This idea might, for example, suggest a new way of viewing the old problem of the composition of Herodotus’s Histories (alluded to earlier in relation to the principle of integrity). This work starts with semi-mythical accounts of the history and of the various parts of the Persian empire (Books 1–5), before moving on to the central topic of the Persian wars (Books 6–9). Scholars have long debated whether Herodotus had this arrangement in mind from the beginning or conceived of its different parts at different times. Certainty is hard to achieve, but it is reasonable to assume that the author of such a long and versatile composition as the Histories interpreted at each stage of his composition, forming opinions and beliefs that led to new aesthetic and thematic preferences. In other words, the Histories must reflect the author’s interpretive beliefs, which are unlikely to have been fully formed from the start or to have stayed frozen from beginning to end. This is not necessarily an “analyst” view, but may suggest the limitations of the “unitarian” approach.30

The author’s interpretative beliefs are not to be separated from intention, which Dworkin (1986.58; cf. 2011.125) sees as an important concept for the interpretive process; it is this concept that provides the formal structure for interpretive claims. Dworkin stresses that interpretation is by its nature the report of a purpose (not cause); even when historical authorial [End Page 107] intention cannot be retrieved, the purposes are those of the interpreter. In fact, it is conceptions of the interpreter’s aims that help us better evaluate the conclusions of the interpretive process (a philosophy of interpretation affects such conceptions, and this makes such a philosophy particularly useful). Dworkin goes on to argue (1986.58) that the concept of authorial intent inevitably engages an interpreter’s artistic convictions.31 This is the essence of “constructive interpretation”: the act of imposing purpose on a work of art in order to make of it the best possible example of its genre (and not necessarily what the author would have wanted it to be). This use of artistic intention may disguise some of the interpreter’s own views, but the interpreter acknowledges that an author might reject his readings. Dworkin further argues that an insight or reading belongs to the author’s intention if it fits and illuminates his artistic purposes (and makes his work better) in a way he would endorse even when he has not explicitly done so—this brings in the interpreter’s sense of artistic value and holds even for authors long dead. As an example of this interpretive method, Dworkin presents (1986.56ff.) an imagined conversation between Stanley Cavell (1969.230ff.) and Federico Fellini on the question of whether a character in the latter’s film La Strada is an intentional allusion to the Philomela legend. The filmmaker admits ignorance of the legend but acknowledges its affinity with the film character, which compels him to “accept it as part of the film he made.”

Dworkin’s analysis appears useful for the reading of ancient texts that show an interest in aesthetics and the issue of authorial intention.32 These ancient texts are not exclusively works of ancient literary criticism. A well-known scene from Old Comedy can arguably be read as a comment on the conflated roles of author and interpreter, as well as a satire of the interpretive beliefs of both (author + interpreter). This is the dialogue (agon) between Euripides and Aeschylus in Aristophanes’ Frogs, essentially a debate about what constitutes good and bad poetry, where each of [End Page 108] the two tragedians “interprets” the poetry of the other. The debate takes place in Hades: both great masters are dead, and Dionysus travels to Hades in order to bring one of the two back to Athens. To establish which one is worthier, the god judges a poetic competition that displays their skills and weaknesses. Authorial intention looms large in the contest: issues of intentionality are forced upon the reader by the work itself as Aeschylus and Euripides explicitly define their intention as didactic (cf., e.g., Frogs 1008–09 and 1053–55). Intention suggests an interpretive mode with a strong evaluative element, and the suggested audience’s ability to function in that mode has been recognised in a recent article.33 However, it would be preferable to assume that it is authorial interpretive beliefs (as imagined by Aristophanes) that are showcased here, as it is the poets themselves who act as interpreters, while the place of Aristophanes and of the audience in the debate can only be guessed at.34

One other example of a text that can be more richly understood as an expression of the blurring of boundaries between author and interpreter is Lucian’s imagined conversation with Homer (an author long dead in Lucian’s time) in True History. This work, which recounts a fantastic journey undertaken by Lucian and a number of comrades, includes a stop at the Island of the Blessed where Homer now resides. Unlike Aristophanes, who does not appear as a character in his own comedy, Lucian fills the double role of narrator and character, and interacts with Homer, “the author,” directly (Lucian True History 2.20, trans. A. M. Harmon):

Hardly two or three days had passed before I went up to Homer the poet when we were both at leisure, and questioned him about everything . . . I went on to enquire whether the bracketed lines had been written by him, and he asserted that they were all his own: consequently I held the grammarians Zenodotus and Aristarchus guilty of pedantry in the highest degree . . . I next asked him [End Page 109] why he began with the wrath of Achilles; and he said that it just came into his head that way, without any study.

This may be read as a dialogue between “interpreter” (Lucian) and "author" (Homer), who in this case also acts as interpreter (of his own work); other “interpreters” are also mentioned (Zenodotus, Aristarchus). Lucian appears disillusioned with contemporary scholarship on the Homeric poems and therefore asks the poet himself to clarify his intentions. Homer’s answers imply that authorial interpretive beliefs are wrongly understood by such critics as Zenodotus, Aristarchus, and those who try to interpret the poet’s chosen way to begin the Iliad; this is a case of an author rejecting certain interpreters’ readings (though the question remains open as to whether it is the author’s or the interpreters’ readings that show the poems as the best literature they can be).

Aristophanes and Lucian have in common the use of satire; they both comment on the nature of interpretation but do it with caricature. A more serious note is struck by a memorable passage from Philostratus’s Apollonius of Tyana (4.16), where, quite strikingly, the voice of a poet, Homer again, is substituted for by one of his heroes, Achilles, who answers Apollonius’s interpretive questions and thus acts as “interpreter” of Homer’s poetry (the idea that a poetic character is able to speak in the name of the poet outside the poetic context where he belongs and express the poet’s purposes and beliefs takes us back to the issue of the nature of poetic creation as independent from the creator).

The reading of this passage, too, can profit from the concept of authorial interpretive beliefs. During his visit to the Troad, Apollonius spends a night on the burial mound of Achilles, who appears to him and agrees to answer five questions on the poets’ treatment of the Trojan war.35 Three of these are about Homer, and two are of a particular interest here: “I asked, ‘Did Helen come to Troy . . . or did Homer choose to invent all that?’” Achilles’ answer is that Homer’s poetry reflected the true beliefs of the Achaeans, who were under the impression that Helen was in Troy, while, in fact, she had been in Egypt the whole time. That Homer’s poetry reflected real events is an interpretive belief, which explains the absence of a reference to Helen’s “true” whereabouts. Apollonius also asks: “How is [End Page 110] it that Homer does not know about Palamedes, or if he does, excises him from his account of you all?” “If Palamedes did not come to Troy,” [Achilles] replied, “Troy did not exist either. But since the wisest and most warlike of heroes was killed by a ruse of Odysseus, Homer does not bring him into his poem to avoid celebrating Odysseus’ crimes.” The hero’s answer suggests that the poet’s interpretive beliefs (in this case, his interpretation of an element of poetic tradition: the role of Palamedes) affect the shaping of his poetic material.36

To summarise: Dworkin maintains that different styles of interpretation are, in essence, different answers to the questions as to “where the value of a work of art lies and how far it has been realised?” (1986.60). When we interpret literature, we best understand disagreement in view of the question: where does the value of literature lie? The academic argument about authorial intention should be seen as one of the arguments about where that value lies, and it can be used alongside or in juxtaposition to other arguments, e.g., the argument of New Criticism that a poem has its deep value as an expression of an idea that has to be contained within the poem itself, independent of any information we can bring to it (about the poet). In this sense, interpretive practice incorporates value from which interpretive responsibility flows. Whatever the interpreter’s understanding of his or her responsibility, it should include a sense of the aim to uncover a work’s excellence. This view of interpretation erodes the difference between two questions: what makes a particular piece of literature good? And what does this particular piece mean? Value is the main parameter in Dworkin’s notion of objectivity in interpretation, which is the topic of the following section.


Perhaps the most important and most difficult question in interpretation is whether there are true and false, most and least accurate, interpretations, or only different ones. I shall present here some of Dworkin’s thoughts, in the belief that they can help increase self-awareness in the interpretation of classical texts regarding the complex relationship between interpretation and truth. [End Page 111]

As already mentioned, different interpretations often set out different questions to answer and thus do not necessarily compete for accuracy and truth. Indeed, as Dworkin notes (2011.139), they can be independent of one another (when they simply do not interact) or complementary to one another (adding insight and not competing). We have seen that interpretation relies on genre and on the interpreter’s pre-existing assumptions. Nevertheless, these considerations, argues Dworkin (2011.124–27), should not dictate interpretive scepticism: we should not assume that there are no right and wrong readings, or only different ones that appeal to different people (this view in itself constitutes an interpretation—just as non-demonstrable as the truth of the readings it questions). In such cases, we may speak of “interpretive uncertainty,” but we need not assume “indeterminacy.”37 An interpreter is responsible for the use of criteria that help form a best interpretation or choose between divergent ones. A classicist who translates and/or interprets classical texts bears a similar responsibility (Dworkin 2011.148–49 on translation).

The pursuit of interpretative truth is encouraged by the phenomenology of interpretation (how it feels to interpreters), which includes a sense that interpretation aims at truth.38 Literary theorists (and that includes classicists) frequently claim truth and assume disagreement, not just difference; their interpretations are often competitive (they take one another to be in some way defective; Dworkin 2011.139–40). In some cases, (the rightness of) a particular interpretation seems ineffable (it just feels right), but ineffability does not guarantee truth. Dworkin (2011.130) argues that interpretation succeeds insofar as it achieves the truth about an object’s [End Page 112] meaning, and that a successful theory of interpretation must account for the possibility of truth, for the clash of opinion about where it lies, and for its intractability; it should also account for all genres of interpretation.39 The aesthetic hypothesis in literature and art functions as a weapon against relativism: interpretative claims may remain subjective in the sense that they cannot be proven true or false in a way that convinces everyone, but disagreement is made easier to tackle if its root causes (the interpreters’ differing conceptions of value) are identified.

Awareness of the above points and limitations might suffice for the open-minded classicist, while the philosopher’s persistence in the notion of absolute interpretive truth may be felt as alienating. Such persistence is, however, hard to avoid, since, as Dworkin notes (2011.131), as we interpret literature, we inevitably also interpret the practice of interpretation. A systematic interpretation of interpretation is entailed in Dworkin’s “Value Theory” (2011.130–34), which brings to the foreground a general notion of interpretation as a process that is oriented towards an ideal target: objectivity. The process comprises a set of distinct stages (2011.131). As we interpret,

  1. 1). we first need to specify the genre of an individual interpretive act (e.g., literary interpretation).

  2. 2). we then attribute “some package of purposes” to that genre of interpretation.

  3. 3). finally, “we try to identify the best realisation of that package of purposes on some particular occasion.”

This schema echoes the stages of “constructive” interpretation that Dworkin proposed in one of his earlier works (1986.65); these stages refer primarily to the interpretation of law as a social practice, but it is implied that they are transferable to literary interpretation.

  1. 1). the preinterpretive stage, at which “the tentative content of the [interpreted object] is identified.” Dworkin states that in literary interpretation, this is the stage when literary texts (novels, plays, and so forth) are identified in terms of type and content and distinguished from other texts. [End Page 113]

  2. 2). the interpretive stage, “at which the interpreter settles on some general justification for the main elements of the practice identified” at the previous stage. In literary interpretation, this is presumably the process of providing aesthetic arguments for the point or value of (the main elements of) a literary work—in other words, the process of building an argument for it being worth creating and reading.

  3. 3). the postinterpretive/reformative stage, at which the interpreter “adjusts the sense of what the practice [or in our case, the literary work] really requires so as to better serve the justification he accepted at the previous stage.”

The preinterpretive stage for literary interpretation is explained by Dworkin himself clearly enough. The next two stages, however, are not described in terms of the relationship between literary and legal interpretation, and their content is not immediately graspable.40 Nevertheless, it seems obvious that the interpretive stage may produce several divergent perspectives regarding the purposes of interpretation and the value of the literature interpreted; the postinterpretive stage is where these perspectives are considered “in terms of [their] applicative relevance to the world, society or the culture [they] address themselves to or imbibe from or both” (Gana 2003.333). It emerges once more that different interpretations may compete with one another for truth and validity, as one’s conception of “best realization” or “justification” (and defense) of a literary work rests on different assumptions about value. Objectivity is connected with value; as Dworkin puts it (2011.132), the level of convergence and divergence between assumptions about value decides whether interpretation flourishes or falls into “argumentative sand” (perhaps he meant quicksand).

Applications of literary theory in the reading of classical texts have brought forth self-conscious, articulate theories concerning the point of their interpretive genre that try to avoid ineffability, but they generally lack a common “self-conscious value strategy,” i.e., a common clear view of where value lies in the literature they interpret. Value judgements in literary interpretation rarely take the form of simple, all-inclusive, maxim-like statements; we may, for example, state that the value of ancient comedy lies in its comedic features, but this is, of course, less than half the story. Dworkin accepts (2011.144–47) that the pursuit of objective interpretation in [End Page 114] literature will not necessarily lead to definitive interpretive claims,41 but it can enable stronger, better-grounded claims than would otherwise emerge. To paraphrase Dworkin’s words (2011.134) so as to form an analysis focusing on the interpretation of Aristophanes’ comedies: “The kaleidoscope of contradictory interpretation[s] [of the plays, as bearers of a serious message vs. solely and purely comic] does not reflect revolutionary discoveries about [Aristophanes’] artistic intentions. Nor is it helpful simply to say that later critics saw in the [plays] what earlier ones had missed; on the contrary, the fact that different critics saw different things is part of what needs explaining. If we are to make sense of what seems undeniable—that each of the long succession of critics took himself to be right and others seriously wrong about ‘what [Aristophanes] was up to’—we must study not the critics’ research into the [author’s] thoughts and ambitions but their sense of where value lies in art and of their own role in creating that value.” Dworkin provides the interpreters of literature (including classical literature) the awareness of an important desideratum.

Nikoletta Kanavou
University of Heidelberg


Aly, Wolf. 1921/1969. Volksmärchen, Sage, und Novelle bei Herodot und seinen Zeitgenossen (2nd ed., with additions by L. Huber). Göttingen.
Burke, Seán. 2010. The Death and Return of the Author: Criticism and Subjectivity in Barthes, Foucault, and Derrida (3rd ed.). Edinburgh.
Cavell, Stanley. 1969. Must We Mean What We Say? New York.
Davidson, Donald. 2001. Subjective, Intersubjective, Objective. Oxford.
Dewald, Carolyn, and John Marincola. 2006. “Introduction,” in Dewald and Marincola, eds., The Cambridge Companion to Herodotus. Cambridge. 1–12.
Dolin, Kieran. 2007. A Critical Introduction to Law and Literature. Cambridge.
Dworkin, Ronald. 1985. A Matter of Principle. Cambridge, Mass.
———. 1986. Law’s Empire. Cambridge, Mass. [End Page 115]
———. 2011. Justice for Hedgehogs. Cambridge, Mass.
Eagleton, Terry. 1996. Literary Theory: An Introduction (2nd ed.). Oxford.
Fish, Stanley. 1989. Doing What Comes Naturally: Change, Rhetoric, and the Practice of Theory in Literary and Legal Studies. Durham, N.C.
Gadamer, Hans-Georg. 1975. Truth and Method (trans. G. Barden and J. Cumming). New York.
Gana, Nouri. 2003. “Beyond the Pale: Toward an Exemplary Relationship Between the Judge and the Literary Critic,” Law and Literature 15.313–43.
Grossardt, Peter. 2009. “How to Become a Poet? Homer and Apollonius Visit the Mound of Achilles,” in Kristoffel Demoen and Danny Praet, eds., Theios Sophistes: Essays on Flavius Philostratus’ Vita Apollonii. Leiden. 75–94.
Guest, Stephen. 2012. Ronald Dworkin (3rd ed.). Stanford.
Heath, Malcolm. 1989. Unity in Greek Poetics. Oxford.
———. 2002. Interpreting Classical Texts. London.
Hirsch, Jr., Eric D. 1967. Validity in Interpretation. New Haven.
Hitchcock, Louise A. 2008. Theory for Classics: A Student’s Guide. New York.
Jacoby, Felix. 1913. “Herodotos,” RE Suppl. II.205–520.
Jannidis, Fotis et al. (eds.) 1999. Rückkehr des Autors: Zur Erneuerung eines umstrittenen Begriffs. Tübingen.
Jauss, Hans Robert. 1982. Toward an Aesthetic of Reception (trans. T. Bahti). Minneapolis.
Kennedy, George A. (ed.) 1989. The Cambridge History of Literary Criticism. vol.1: Classical Criticism. Cambridge.
Konstan, David. 2014. Beauty: The Fortunes of an Ancient Greek Idea. Oxford.
Lardinois, André. 1996. “Who Sang Sappho’s Songs?,” in Ellen Greene, ed., Reading Sappho: Contemporary Approaches. Berkeley. 150–72.
Leavis, Frank R. 1986. Valuation in Criticism and Other Essays (ed. G. Singh). Cambridge.
Martindale, Charles. 1993. Redeeming the Text: Latin Poetry and the Hermeneutics of Reception. Cambridge.
———. 2005. Latin Poetry and the Judgement of Taste: An Essay in Aesthetics. Oxford.
Nünlist, René. 2009. The Ancient Critic at Work: Terms and Concepts of Literary Criticism in Greek Scholia. Cambridge. [End Page 116]
Palmer, Richard E. 1969. Hermeneutics: Interpretation Theory in Schleiermacher, Dilthey, Heidegger, and Gadamer. Evanston.
Patrick, Dale, and Allen Scult. 1990. Rhetoric and Biblical Interpretation. Sheffield.
Posner, Richard A. 1988. Law and Literature: A Misunderstood Relationship. Cambridge, Mass.
———. 1996. “Law and Literature: A Relation Reargued,” in Lenora Ledwon, ed., Law and Literature: Text and Theory. New York. 61–90 = Virginia Law Review 72 (1986) 1351–92.
Powell, Jim. 2007. The Poetry of Sappho. Oxford.
Ramberg, Bjørn, and Kristin Gjesdal. 2014. “Hermeneutics,” The Stanford Encyclopedia of Philosophy (Winter 2014 Edition), Edward N. Zalta, ed., URL = http://plato.stanford.edu/archives/win2014/entries/hermeneutics/
Rosen, Ralph M. 2008. “Badness and Intentionality in Aristophanes’ Frogs,” in Ineke Sluiter and Ralph M. Rosen, eds., Kakos: Badness and Anti-Value in Classical Antiquity. Leiden.
Schmitz, Thomas A. 2007. Modern Literary Theory and Ancient Texts: An Introduction. Oxford.
Sluiter, Ineke, and Ralph M. Rosen (eds.) 2012. Aesthetic Value in Classical Antiquity. Leiden.
Speaks, Jeff. 2014. “Theories of Meaning,” The Stanford Encyclopedia of Philosophy (Fall 2014 Edition), Edward N. Zalta, ed., URL = http://plato.stanford.edu/archives/sum2014/entries/meaning/
West, Martin L. 2003. Greek Epic Fragments. Cambridge, Mass.
Whitman, Jon (ed.) 2000. Interpretation and Allegory: Antiquity to the Modern Period. Leiden.
Wihl, Gary S. 2004. “Fish and Dworkin on the Work of Interpretation in a Democracy,” in Gary A. Olson and Lynn Worsham, eds., Postmodern Sophistry: Stanley Fish and the Critical Enterprise. New York. 75–96.
Williamson, Margaret. 1995. Sappho’s Immortal Daughters. Cambridge, Mass.
Wimsatt, William K. 1954. The Verbal Icon: Studies in the Meaning of Poetry. Lexington. [End Page 117]


1. Thanks are due to the Alexander von Humboldt Foundation and to the Foundation for Education and European Culture (IPEP) for supporting my research. I am grateful to Harry Papadopoulos, lawyer, for introducing me to the philosophy of Dworkin as a subject of study and for his insightful comments on the final version of this paper; also to Jonas Grethlein and to the anonymous reviewer of Arethusa for a number of useful suggestions on earlier drafts.

2. Note the basic distinction between scientific and artistic interpretation (where literary interpretation belongs); Dworkin 1986.53: “We judge success in works of art by standards different from those we use to judge explanations of physical phenomena.” For Dworkin (2011.134–39), both literary and legal interpretation belong to the “collaborative” genre, in which the interpreter is in partnership with the creator in the sense that (s)he assumes functions that are similar to those of the creator. Historical interpretation is “explanatory” (the interpreter does not collaborate with historical actors), and philosophy interprets concepts (“conceptual” interpretation).

3. Cf. Gana 2003. Note also Dworkin’s long debate with Stanley Fish; see, e.g., Fish 1989.87ff., Guest 2012.86, and Gana 2003.333 on how some of the interpretive concepts developed by Fish may have influenced Dworkin.

4. Among the extensive relevant bibliography, see, e.g., Dolin 2007. Note Gana’s introductory comments and his response to Posner’s arguments against assuming a relationship between law and literature (which are largely based on an unfair trivialisation of literature; Gana 2003.313–14, 321–23). See also Gadamer 1975.539 on the view that literary interpretation differs from legal hermeneutics in that the former does not employ normative arguments. But we shall see that Dworkin’s interpretive concepts have a normative function. For a defense of the law-literature relationship, see, further, Guest 2012.88–89.

5. Cf. Dworkin’s programmatic remarks (2011.127): “In literary interpretation . . . new tribes of critics emerge daily claiming an entirely different—and better—way to read Spencer or Kerouac: we are treated to psychodynamic readings, all-within-the-text readings, reader response readings, cultural myth readings, Marxist and feminist readings. Can we make any sense of the competition among these tribes?”

6. Concerns about quality were expressed in classical antiquity (some classical examples in this paper demonstrate this). The various factors that guided ancient aesthetic appreciation (which included perceptions of beauty, a pleasurable effect, but also social and practical usefulness) are discussed in the essays collected in Sluiter and Rosen 2012; see now also Konstan 2014.5–7, 171. On modern “hostility to talk about beauty and to aesthetic criticism,” and on responses to this hostility, see Martindale 2005 (the quotation is from the prologue) and Konstan 2014.189–91.

7. This approach to meaning is intuitive and commonsensical. Dworkin does not (and I shall not either) discuss philosophical theories of meaning (on which a good introduction is Speaks 2014). The relevant philosophical discussion has penetrated the work of literary theorists as, for example, in the distinction between “meaning” and “significance” (see Schmitz 2007.125).

8. Cf. Dworkin 2011.102: the interpretive process is “the process of seeking meaning” [in a text].

9. See, e.g., Palmer 1969.118–21; Ramberg and Gjesdal 2014 sketch the evolution of this notion in philosophical hermeneutics.

10. For example, “good” literature must be unconventional and defy the readers’ expectations (Jauss 1982.25); see Schmitz 2007.89 for a summary of natural objections to this criterion.

11. This view goes back to the tradition of hermeneutics, e.g., Gadamer 1975.336, 358—in particular the notion of the “fusion of horizons,” which describes the encounter between the cultural horizon of readers and the authority of the text; cf. Martindale 1993.5–7. Note, however, Hirsch’s criticism of this notion for its relativist nature (1967.245–54; see also below, n. 38, and Ramberg and Gjesdal 2014 for an overview of further discussions). Cf. Fish’s (1989.26 and passim) concept of “interpretive communities” (cultural contexts of interpretive practice).

12. See, e.g., Martindale 1993.6 for objections to Wolf’s historicist approach to Homer in his Prolegomena ad Homerum (1795); cf. Eagleton 1996.x (an unprejudiced approach to texts is a mere illusion).

13. Cf. the centrality of questions and dialogue in the hermeneutic circle; the interpreter enters into a dialogical relationship with the object of interpretation (Gadamer 1975.370ff.). According to Heath 2002.40–41, we might eliminate pluralism in interpretation by determining which question(s) we should be asking.

14. These are similar to Heath’s “preconceptions” or “background assumptions” or “prior preliminary judgements,” which do not constitute “immutable truths” (2002.34–36, 100–07). They may also echo the preoccupation of reader-response criticism with the readers’ pre-existing assumptions and expectations about a text (Jauss 1982.22–25).

15. The context of this poem’s citation by Dionysius of Halicarnassus (Comp. 23) offers a good example of the ancient interest in aesthetics and artistic quality: Dionysius describes the poem as “a polished and exuberant composition” that possesses “euphony and charm.”

16. See, e.g., Lardinois 1996 (esp. 151, 164 on frag. 1).

17. For an example of a systematic application of the aesthetic hypothesis to non-legal texts, see Patrick and Scult 1990.84–102, who apply it to the interpretation of the Book of Job and identify a set of judging criteria that emerge from Dworkin’s presuppositions. A similar venture into the field of classical texts lies outside the scope of the present paper but is envisaged as a separate study.

18. Wihl 2004.86. Dworkin himself recognized this banality (1985.152).

19. Cf. Heath 2002.68: “We do not in practice look for the best possible answer to a given question; we look for an answer that is better than the one we have already and good enough for the purposes we have in hand” (emphasis in original).

20. On Dworkin’s debt to Gadamer and Fisher here, see Gana 2003.317–18; Dworkin cites Gadamer (1986.62) and T. S. Eliot (2011.141). For criticism (and Dworkin’s answer to it), see Guest 2012.85–86. A similar idea in different terms is expressed by Martindale 1993.4–10, who argues for the role of the history of reception in the practice of interpretation: later appropriations of classical themes and modern scholarship come between readers and text.

21. Dworkin 1985.159 mentions the pornographic novel Naked Came the Stranger that was written in this way. A typical chain novel is the modern Greek Novel of the Four (1958). Not every collaborative novel is a chain novel: collaboration has to proceed in the form of a chain.

22. See Fish’s critique (1989.88ff.) and the defense of the chain by Guest (2012.87–88).

23. Possibly also by rewriting the law—a controversial idea.

24. This is the essence of the “collaborative” genre of interpretation, to which, according to Dworkin, literary interpretation belongs (see above, n. 2). Cf. the post-structuralist notion of the “critic as artist,” on which see Martindale 1993.35–39, who draws on the influential theories of Harold Bloom (the notion goes as far back as Oscar Wilde and his essay “The Critic as Artist” [1891], which attacks the distinction between fine art and criticism).

25. A relevant question (Dworkin 1985.154) is whether discussion of authorial intention should uncover literary value or meaning; we have seen that an interpretation based on the aesthetic hypothesis does not separate the two. Intentionality in literary interpretation is a more complex issue than in other forms of interpretation (e.g., conversational, scientific), because of the different nature of the standards used to measure value or success (see also Dworkin 2011.128–30).

26. Citing Samuel T. Coleridge’s celebrated view that to answer the question “what is poetry?” is in effect to answer the question “what is a poet?” (Biographia Literaria [1817] chap. XIV).

27. Schmitz 2007.91–92, 126. Dworkin 2011.130 refers to the term “intentional fallacy,” used by New Criticism for the mistake of identifying the meaning of a text with its author’s intentions (the classic text is Wimsatt 1954.3–18).

28. Dworkin 2011.140 sees these as “independent of any interpretive reading of the work.”

29. Cf. Heath 2002.79–83 (and 60–78 on further limitations to the concept of intentionality).

30. Cf. Felix Jacoby’s analyst theory of the Entwicklungsgeschichte of Herodotus (1913.216ff.): the thematic versatility of the Histories may reflect the evolution of the author’s interests, from ethnography and geography to purer historiography (cf. also Aly 1921/1969). More recent Herodotean scholarship tends to emphasize the unity of the Histories; see, e.g., Dewald and Marincola 2006.1–3.

31. The idea of the central role of the interpreter in the process of interpretation was developed by hermeneutics (see, e.g., Palmer 1969.52).

32. See, e.g., Nünlist 2009.166 on an “intentionalist” reading of a Homeric passage by an ancient critic (Porphyry). Note that both Aristotle (Poetics) and Horace (Ars Poetica) extended their interpretive theories beyond the criterion of the author’s intention. Concern with authorial intention in certain periods of literary practice reflects an intellectual culture that attaches value to the process of artistic creation (Kennedy 1989.xi). Cf. Dworkin 1986.60: romantic faith in a creative genius explains our preoccupation with intention, sincerity, originality.

33. “Aristophanes . . . does seem to be aware that assumptions about an author’s intentions (including factors extrinsic to the work itself) can have an important effect on how one ends up judging it” (Rosen 2008.146 n. 6).

34. Dionysus (Aristophanes?) differentiates between the perspective of the poets and that of the actual theater-goer (Frogs 905–1097). An audience’s interpretation may well miss the meaning intended by an author (Frogs 934 offers a comic example of this). Authorial intention may equally fail to be grasped; in fact, reception lies outside the author’s control (cf. Rosen 2008.149–53 and notes 11 and 18).

35. On this passage, see, further, Grossardt 2009. The translator is C. P. Jones; my emphasis. Note that Philostratus probably visited the tomb of Achilles in 214/15. His concern with Homer and the Trojan heroes is more fully manifested in the Heroicus.

36. The story of the conflict between Odysseus and Palamedes must, in fact, be later than Homer. The latter’s fate was mentioned in the Cypria (frag. 27 West).

37. This counters the scepticism expressed, e.g., by Heath, who questions the feasibility of a single, correct theory of interpretation (2002.40–41; cf. 46: “Pluralism is . . . ineliminable in principle as well as in practice”), as well as its desirability: interpretive disagreement protects us from being “immobilised in error.” But is “error” the optimal term? Diverging interpretive attempts might rather express “uncertainty,” cf. the transformations of Heath’s own reading of Aristophanes (2002.102–03).

38. Dworkin 2011.126, citing Leavis 1986.277: “A critical judgment . . . means to be more than personal.” Objectivity is a locus of disagreement in hermeneutics: in Hirsch’s concept of validity in interpretation, objective interpretation is the ultimate goal; Gadamer refrained from a notion of absolute objectivity, placing the emphasis rather on the investigation of the nature and the possibilities of interpretation (see, e.g., Gadamer 1975.xxii and Palmer 1969.46). The interest in the connection between understanding and truth is a point of convergence between Gadamer and Davidson; the latter is also cited by Dworkin 2011.148–49, who did not see in Davidson’s conception of interpretation as indeterminate (2001.214) a negation of the possibility of a “unique theory of truth.”

39. Note, in comparison, the notion of a universal hermeneutics (which goes back to Schleiermacher): “a hermeneutics that does not relate to one particular kind of textual material (such as the Bible or ancient texts), but to linguistic meaning in general” (Ramberg and Gjesdal 2014; “universal hermeneutics” evolved to encompass ontology: see, e.g., Gadamer 1975.490ff.).

40. See also Gana 2003.333 for an application of these stages to literary interpretation.

41. Scepticism can be fruitful, if it flows from the interpreter’s critical responsibility (Dworkin 2011.146).