Dimensions of Interpretation

In: Verstehen und Interpretieren
Open Access

I. Meaning, Understanding, Interpretation

My allotted topic is interpretation, but I want to begin by saying a few words about how I take the concept of interpretation to relate to the other key concept for this workshop, the concept of understanding. Both of these words are ambiguous in multiple ways, and the purpose of these remarks is mainly to indicate the way that I intend to use them in this talk – a rather standard way, I think, but not the only way.

Understanding, first, I take to be a psychological state. Understanding is always someone’s understanding: a paradigmatic example would be the state that you enter into when you read or hear a not too difficult utterance in a familiar language. In ordinary usage, understanding is most often an achievement word, in Ryle’s sense – if you understand something you are getting it right. But there is also a neutral way of using the word, where misunderstanding is a mode of understanding. Here, I will take understanding in the latter sense, where you can understand something, for example an utterance, in different ways, and still be qualified as understanding it.

Understanding has an object – the text, utterance, picture or whatever – that you understand. And it has a content – the way that you understand it. The content of an understanding is what I will call a meaning: to understand something is to experience it as having a certain meaning.

What about interpretation? In standard usage, the word “interpretation” suffers from a process/product ambiguity. Interpretation may be an activity, something that you do, or it may be the product of that activity – a finished interpretation, the result of your interpretive labours. This ambiguity often colours accounts of interpretation, sometimes making them ambiguous, in turn. For example, when Gadamer writes about the hermeneutic circle, he often seems to be talking of the process of interpretation as it develops in time, of a movement back and forth between the parts and the whole. But he also, and perhaps more importantly, thinks of the hermeneutic circle as an unavoidable feature of every interpretation in the product sense, of the achieved interpretation – the hermeneutic circle in this sense is a close cousin to the circularity of a circular argument or a circular definition.

Here, I will concentrate on interpretations in the product sense, and primarily on interpretations as the products of scholarly activities of interpretation, as when we talk about Brandom’s interpretation of the Phänomenologie des Geistes or Kripke’s interpretation of the private language argument. One of my main contentions, however, will be that interpretation is not a basic notion, but that interpretations should be factorised into two components which I will call “readings” and “claims”. Roughly, a reading is an assignment of a meaning to a text, while a claim, even more roughly, is a reason why it is relevant or interesting to read the text in this way. One type of claim on behalf of reading may, for example, be that this is the way its author intended it to be read but readings may draw their interest from many other sources as well.1

So what is the difference between an interpretation in the product sense, and an understanding? Would it not be better to think of interpretation as the process that, when successful, leads to understanding? This, I think, is Gadamer’s main way of thinking: interpretation is an activity which we engage in when understanding is lacking or problematic, and which, when successful, leads to a new or better understanding. There is nothing wrong with this way of speaking, but here I want to look at interpretations in a slightly more objective way. An interpretation (in the sense of a reading) does not need to be the content of someone’s act of understanding, but I think of it simply as a specification of a meaning for a text, in a very wide sense of both words. Most of the time, we can think of an explicit interpretation as a text expressing the meaning of another text, sometimes through direct paraphrasis, sometimes as a metatext describing the meaning of the object text. Interpretations, in this sense, are embodied in books and articles produced by scholars and critics on the occasion of other works, though they may also be thrown up and discarded as byproducts in the production of finished interpretations, or they may be presented just in order to be refuted.

II. To the Lighthouse

To elucidate the distinctions that I wish to make, I will use one concrete and elaborate example: Virginia Woolf’s To the Lighthouse. Nothing of what I have to say hinges on this being a work of literary fiction, and I will comment on other domains of interpretation as we go along, but it is convenient and illustrates many of the points I want to make. I will start out with a few actually proposed interpretations of the novel. I will not take up the time with a description of the plot, but just take it for granted that you know it well enough to go along – actually, I don’t think that it matters much if you don’t happen to know it at all. I will give you four different readings of the novel, starting with:

Philosophical reading 1: “Virginia Woolf’s To the Lighthouse is a study of personality, of the relationship between the sexes, of time, death, nature and art.”2

The basis for this interpretation is, of course, that the story is about some persons of different sex, that time passes, that some of the persons die, that the characters think and talk some about death, that there are artists in the story, e.g., a painter and a poet, that produce works of art, of which Lily Briscoe’s painting of Mrs Ramsay is the most important. I think we all recognise this sort of interpretation – it is very common for critics and literary scholars to point to general and universal themes of this nature, that the literary work is supposed to treat in the guise of a specific and particular story. As an interpretation it is, of course, very sketchy, but here is a little elaboration of it, relating it more closely to the actual narrative:

Part I of To the Lighthouse deals with ideas about personality, the relationship between the sexes, and modes of escape from time. Primarily, these ideas are expressed through the Ramsay’s, whom Mrs Woolf has endowed with qualities she believed typical of the basic masculine and feminine characters. Their traits are complementary: Mrs Ramsay’s creative, intuitive femininity balances her husband’s courageous, intellectual masculinity.3

I will leave it there, for the time being, and proceed to:

Philosophical reading 2: “To the Lighthouse is really the story of a contest between two kinds of truth – Mr Ramsay’s and Mrs Ramsay’s. For him, truth is factual truth; for her, truth is the movement toward truth: since truth is always being made, and never is made, the struggle for truth is the truth itself. The form of this novel at once expresses and verifies Mrs. Ramsay’s truth.”4

This is a generalising, thematic reading, of roughly the same kind as the first one, though it is staged on an even more abstract level. The interpreter goes on to elaborate it in greater detail with the help of Bergson – in fact, the novel is largely taken to be a fictional rendering of Bergson’s philosophy.

Though different, the first two readings are similar in their general thrust, and it is not hard to envisage them to be combined in a synthetic Philosophical reading 3. The next reading, however, places the ball in a different park:

Theological reading: “Although there are ten people in the Ramsay family, and they have six guests, at the dinner – as at the Last Supper – only thirteen people appear. Mrs Ramsay represents Christ serving his twelve disciples at the Last Supper. She, like Christ, is soon to die.”5

If the two philosophical readings deal in interpretation as generalisation or perhaps “essentialisation” – from this man and this woman to men and women in general, from this painting to the nature of art – the theological reading is more like an allegory in the classical sense: the overt story is taken as a cipher for another, hidden, story.

The fourth and last reading is, again, very different, but of a quite common type. It is not a published, scholarly, interpretation, but given in a letter to the author, by her sister Vanessa Bell:

Biographical reading: “It seems to me that in the first part of the book you have given a portrait of mother which is more like her to me than anything I could ever have conceived possible. It is almost painful to have her so raised from the dead. […] You have given father too I think as clearly, but perhaps, I may be wrong, that isn’t quite so difficult.”6

This is a reading of the novel which Virginia Woolf seems to have happily accepted, and it is not only current in the circle of her family and friends. In biographical accounts of people in and around the Bloomsbury group one refers, seemingly without hesitation, to Woolf’s portrait of her father, Leslie Stephen, in To the Lighthouse, and compares the picture she gives there with what she herself or others say about him in other contexts. Almost any author biography is replete with these sorts of readings.7

III. Readings and Claims

I hope we can all see that these are indeed four different ways to read the novel, four different readings as I will say. Is any of them correct or perhaps wrong? A puzzling question, because we have so far given it no sense – for the question of correctness to arise, we must consider the reading together with some specific claim made for it, for example the claim that this is how the author intended her work to be read. So, this is the first distinction I want to insist upon: an interpretation should, for analytical purposes, be factorised into, on the one hand, a reading and, on the other hand, a claim on behalf of the reading.

Interpreters usually concentrate on the reading and are less explicit about the claims they make for it – sometimes leaving us with genuine unclarity, sometimes relying on context and default. I will come back to the many different types of claim, but let me first illustrate with some of our examples. For most scholarly readings of literary works, the default claim, I think, is indeed intentional, at least in a broad sense. Looking at the first philosophical reading, this is made explicit already in the quote I gave: “These ideas are expressed through the Ramsay’s, whom Mrs Woolf has endowed with qualities she believed typical of the basic masculine and feminine characters.”8

The theological reading is a more interesting case – it looks more outrageous on the face of it, but in fact its author has made a strong intentional claim explicit in this case as well:

Since the Lighthouse is used as a symbol, it could of course imply many things; but it has, I feel, a single intended essential symbolical meaning. Using the method of Joyce in Ulysses, in To the Lighthouse Virginia Woolf was writing a novel with the overtones of an epic. The work is an allegory, with possible reference to various literature, but based principally on the bible.9

Whatever we think about the plausibility of this claim, we would surely like to see some evidence for it, apart from the reading itself. And this brings me to my other distinction, closely related to the first. Just as we should factorise an interpretation into a reading and a claim, we should distinguish, among what are generally taken as arguments for interpretations, between the elaboration of a reading and the justification of a claim. If asked to elaborate the theological reading, an appropriate response would be to give more and more detailed biblical parallels, but clearly, that would not in itself do much to justify the claim that this is the reading that the author intended. To justify that claim, we must appeal outside the text. Indeed, with regard to the justification of an intentional claim a letter or a public statement from the author may carry quite some weight – as it obviously does in the case of the Homeric parallels in Joyce’s Ulysses.

The distinction between elaboration and justification can also serve to dispel a worry which is frequently raised about intentionalism in interpretation: that it threatens what Gregory Currie has called “the centrality of the text”10. If meaning is intention and one is allowed to use any sort of evidence to reveal it, does that not reduce the text to just one clue among many to what it means? The solution to this conundrum is that a reading is always a reading of the text, in the sense that it connects meanings to features of the text. “External” evidence may bolster the claim that this is how the author intended the text to be read, but it does not replace the reading itself. – I will come back to possible claims for the autobiographical reading later on.

IV. Readings

I will use the rest of my talk to elaborate the conceptual apparatus that I have hinted at, saying more about readings, texts and claims, and the interaction between them in interpretation. Nominally, a reading is just an assignment of a meaning to a text, but that leads to the questions of what is a text, what is a meaning and what does it mean to assign the latter to the former?

Types of Meaning

One thing that should be clear from the examples is that the meaning sought through literary interpretation is seldom linguistic meaning or anything very similar to linguistic meaning. There are no linguistic or literary conventions for readers to rely upon in order to figure out that To the Lighthouse is about time, truth or art, and even given these topics, there are no rules to establish more precisely what it has to say about them. The relevant semiotic relation for the two “philosophical” readings seems to be what the rhetorical tradition calls exemplum (Aristotle’s paradigma): features of a specific example are appropriately generalised to cover a wider class of cases. The theological and biographical readings trade in different species of allegory: behind the overt story there is another one, accessible to those who have the key. And these meaning relations are by no means unique to literary interpretation – case law gives a particularly clear example of the relevance of the exemplum in juridical interpretation. Conversely, I do not mean to imply that literary interpretation is all about generalisation and allegory – for example, the symbolic meaning of the lighthouse in Woolf’s novel does not fall into either category.

For the moment, the main point of drawing attention to these forms of meaning is to break the spell of the linguistic paradigm, the idea that linguistic semantics or speech act theory can be expected to provide a blueprint for textual interpretation in general. One pertinent difference is that linguistic meaning, generally speaking, is arbitrary – there is no other way to find out what a word means than to consult the lexicon for the relevant language – while most other forms of meaning are motivated. To be sure, a hint about the relevant allegory or topic for generalisation may be handy, but the more of it you can figure out for yourself, the better reader you are.

The distinction between arbitrary and motivated signification also gives a clue to why the author’s intention rules supremely for some types of meaning, while it can be plausibly contested in other cases. We don’t quarrel with authors about the meaning of words. There is no point in supplying another arbitrary meaning for the one that the author has chosen – it is just too easy. But for motivated meanings there has to be an argument, and arguments can always be contested: “I understand that you intend Mr Ramsay to represent the masculine principle, but for me he is just a Victorian upperclass boor.”

The Relativity of the Text

A reading assigns a meaning to a text. But if there are many kinds of meaning, standing in different semiotic relations to their textual basis – what, then, is a text? In the case of literature, the question of what a text is seems deceptively simple – is it not there on the page? The literature on literary interpretation is full of would be platitudes like this:

Whatever else it is, a literary work is first of all a text, a piece of language. So what the interpreter reveals is the meaning of a text.11

[…] literature is, first, a linguistic phenomenon and, further, an artistic one – works of literature are linguistic works of art.12

I think that this is importantly wrong. I do not want to deny, of course, that language plays an important part in literature but linguistic interpretation is not a dominant feature of literary interpretation. The misidentification of the object of literary interpretation as always a verbal text, is one of the main sources of the mistaken assimilation of all interpretation to linguistic understanding, and of the tendency to look to one’s favourite philosophy of language for the clue to literary interpretation. But if it is not the verbal array that is the text, what is it then? Look again at our first example:

Part I of To the Lighthouse deals with ideas about personality, the relationship between the sexes, and modes of escape from time. Primarily, these ideas are expressed through the Ramsay’s, whom Mrs Woolf has endowed with qualities she believed typical of the basic masculine and feminine characters.

There is no mention here of words or sentences: the basis of the reading is the characters and the events of the novel, not the words used to describe them. This is not to deny, of course, that we get access to the events and characters through written descriptions. But even that is not necessary – on this level, by and large, the same readings are available for the movie version as for the written novel.

If we look closer, we will see that the object of a reading, what I will here continue to refer to as the “text” in a very generalised sense, is relative to the type of meaning ascribed. The text for a generalising or allegorical reading, like any of the examples that we considered for To the Lighthouse, is, so to speak, the top layer of a hierarchy of other readings, operating on other texts. A sketch of the relevant hierarchy, for these examples may look like this. First, we have the actual ink marks on the page, which must be read as letters of a certain alphabet – if you think that this is always trivial, consider the travails of deciphering hand written sources, or the efforts to decipher cuneiform script or linear B. Second, these letters must be read as constituting words in a particular language – again, the early controversies whether the language of Linear B was Greek or not provides a vivid example. Third, these words must be assigned linguistic meanings – a task that may involve difficult problems of disambiguation, and of identifying the semantically relevant historical state of a language, a sociolect or an idiolect. Fourth, comes the task of assigning real or fictional reference to referring expressions identified at the previous stages. Fifth, comes assigning illocutionary force. Sixth, a string of assertions or the like must be organised into a coherent whole, for example an argument or a narrative, by filling the gaps in the information explicitly supplied, in the narrative case invoking mechanisms along the lines of David Lewis’ Truth in Fiction.13 Seventh, we are finally at the stage where we can start to discuss generalising, symbolic or allegorical readings of the narrative, of the type that we started out with, taking, in this case, the complete narrative, with its characters, events and situations as the text.14

I mentioned the fact that the novel and the corresponding film may present the same narrative by means of different signifying media, and thus offer the same possibilities of higher level readings. This obliteration of lower levels is a general phenomenon. The printed text must be in some particular font but the same linguistic text may be printed very differently. The linguistic text must be in some particular language, but the same linguistic meaning may be given by a translation into another language, and so on. But, specifically in aesthetic contexts, obliteration of lower levels is often tempered by the phenomenon that Nelson Goodman calls “repleteness”15 – in a poem, for example, phonic or graphic features do not lose their relevance, once the meaning is established.

Interpretation and Description

The relativity of the text throws some light, I think, on the relation between interpretation and “description”. These are often distinguished in epistemological terms – the description accounts for “given” features of the object, features that are accessible “before” any interpretation, while interpretation is supposed to imply some degree of uncertainty or choice between different possibilities. But just like the distinction between text and meaning, the distinction between description and interpretation is relative. The very fact that the text is in most cases not a physical fact, but is itself the product of a “prior” interpretation, makes it necessary to present it through a description, which indicates the level at which the object is already read, and in particular which features of it are taken as relevant for the meaning on the next level.

Perhaps, I need to say something as well about the sense in which the text and the description are prior, in relation to the meaning. The priority is a logical one, with respect to the reading: the reading ascribes meaning to the text. But it is not necessarily an epistemological or a temporal priority – part of the reason for accepting a description of a text may well be that it sustains the meaning that we want to ascribe to it on higher levels.16

V. Types of Claims

Let that be enough about readings, for the time being, and let me turn to the different types of claims that can be and are made on behalf of readings. This is where large parts of the contemporary debate about interpretation is played out, for example in discussions of the respective merits of actual versus hypothetical intentionalism. As already hinted, I am an all out pluralist when it comes to interpretational claims, a pluralism that I think is borne out by actual scholarly practice. I will sort the types of claim in a few, not mutually exclusive groups, and then I will discuss some possible mixtures and interactions between claims. The main distinction is between historical claims and value claims. A historical claim concerns the empirical relation between a reading and a text, and is justified by empirical evidence that the relevant relation holds, while a value claim is supported by arguments that the reading in itself has some valued property: truth, interest, beauty, or whatever is relevant in each case. Of particular interest among the historical claims are intentional claims, which are themselves a subspecies of what I will call explanatory claims, but, as we will see there are also historical claims that are not explanatory. Among value claims, I will look primarily at those that are usually invoked in connection with the principle of charity, or Gadamer’s “Vorgriff der Vollkommenheit”17 – claims that readings are true, rational, coherent, consistent, and the like.

V.1 Historical Claims

A historical interpretation is a reading which is claimed to match some historically occurring understanding of the text. Foremost among the historical claims are the intended meaning claims – the claim on the part of a particular reading that it matches the author’s own understanding of the text.

Intended Readings

In the debate about the intentional fallacy there is a good deal of confusion about what the author’s intended meaning is supposed to be. For example, intentional interpretations are often taken to favour biographical readings – a confusion initiated by Beardsley and Wimsatt themselves – or to encourage the reader to delve into the private and perhaps abstruse associations of the author. But, in most cases, the main reason to reject such readings is precisely that they are not intended. The best way to define an intended meaning claim is to take a hint from Grice’s theory of communicative intentions18 :

Intended reading claim: The reading is claimed to match what the author intended the audience to understand.

Within that general formulation there is obviously room for a lot of specifications – room that scholars often exploit to make more nuanced intentional claims. What is, for example the intended audience? It is quite possible to intend the same text to be read in different ways by different audiences, sometimes differing in degree of literary sophistication, and sometimes in more radical ways, along the exoteric-esoteric dimension. Vivid examples of this occur in politically oppressive situations, where utterances may be designed to have both an innocent and a politically charged reading, intended for different audiences. Another example is irony, which usually has an appeal to multiple audiences built in: the humour of irony is the invitation for the insider to laugh at the expense of the outsider.19

It is noteworthy that intentional meaning claims, when understood in this way, are by themselves enough to block many concerns that anti-intentionalists regularly worry about. Look, for example, at the false opposition between intention and convention and the fear that intentional interpretation would open for a Humpty Dumpty semantics. If the claim that the author intended the audience to read in a certain way is to be taken seriously, the author must also have intended to provide the means that would make it possible for the audience to find that reading. Among the most important of these means, obviously, will be adhering to linguistic and literary conventions that she expects the audience to be familiar with. Conversely, the most important motivation for the interpreter to appeal to a certain background of convention and tradition is the fact that this is the background that the author expected the intended audience to use. What better reason could be given for interpreting Shakespeare according to the conventions of Elizabethan English than the fact that this is the language that he wrote? In a similar manner manner, the most straightforward argument against a biographical interpretation is usually that it was not intended by the author, and that in most cases the author cannot reasonably have expected the audience to have enough background knowledge about the author’s life to produce the relevant reading.

One more thing that is worth touching on in passing, is the possibility of unsaturated intentions. Again, Virginia Woolf supplies a good example, reflecting on the symbolic meaning of the lighthouse, in her eponymous novel:

I meant nothing by The Lighthouse. One has to have a central line down the middle of the book to hold the design together. I saw that all sorts of feelings would accrue to this, but I refused to think them out, and trusted that people would make it the deposit for their own emotions – which they have done, one thinking it means one thing another another. I can’t manage Symbolism except in this vague, generalised way. Whether it’s right or wrong I don’t know, but directly I’m told what a thing means, it becomes hateful to me.20

So, to give a symbolic meaning to the lighthouse, is to read according to the author’s intentions, and not to read it that way is to go amiss of what is intended – but there is no particular symbolic reading that can rightly be claimed to be the intended one. There is, so to speak, an intention with an empty slot in it to be filled by the reader: an unsaturated intention. The author relies on the readers capacity to travel from text to meaning in an independent way, driving the car of motivation herself.

Secondary Readings

Let that be enough, for the moment, about intended meaning claims, and let me mention one other type of historical claim that is common and important in scholarly reading.

Secondary reading claims: The reading is claimed to match some historically occurring reading of the text, which is not the author’s.

Many of the texts that scholars read have a long and rich Wirkungsgeschichte, supplying it with theological or philosophical content that in many cases goes way beyond anything that can reasonably be ascribed to the author. The founding texts of the great religions are obvious cases in point, but the dialogues of Plato or the works of Kant and Hegel provide more of the same. For the student of Chinese philosophy, establishing the author’s intended reading of the Dao De Jing is, of course, of great interest, but it is just as important to map the way the text was understood by Wang Bi, or at different stages of neo-konfucianism, or within Ch’an Buddhism, and so on. Each of these readings may be different, and some of the later ones would probably have been quite inaccessible to the author, but each of them may be historically correct relative to the different claims.

Explanatory Claims

The next group of possible claims, with regard to readings, is not disjunct from the first group, but it highlights other motives for interpretation and it opens for a wider class of readings.

Explanatory claims: The reading is claimed to explain some features of the text; or the action of uttering or issuing it; or its reception; or some other aspect of its history.

Intended meaning claims, of course, are generally explanatory with regard to the act of uttering a text – they provide a partial intentional explanation of some features of the utterance.21 But such explanation is, at best, only partial and other readings my claim to be explanatory as well – and some perfectly legitimate readings may not be explanatory at all, at least not with regard to the text or its utterance.

I will just give a few example of explanatory claims, that need not qualify as intended reading claims. My first example are biographical readings, like the one that Vanessa Bell sketches for To the Lighthouse. It is implausible that this is a reading that Virginia Woolf intended her audience to make – though there remains, of course, the possibility of a special esoteric reading intended for a select group of family and friends. But it is very plausible to think that she actually used her family as models for some of her characters, in roughly the same way that a painter could use his brother as a model for a picture of Christ. The reading of Mr and Mrs Ramsay as portraits of Leslie and Julia Stephen may explain some of the features that they are ascribed in the novel, just like the facial features of the painter’s brother would explain the looks of Christ in the hypothetical painting. Their role in the explanation of features of the text or utterance, is what makes such readings useful as sources of information about circumstances external to the text itself – which, of course, is why biographical readings are so cherished precisely by biographers.

Many modern strategies of reading are primarily interested in reading “behind the back” of the author, in order to uncover presuppositions and mechanism that the author may not be aware of. Much feminist and post-colonial reading would fall into this category: sexist and racist attitudes of historical authors may have been so ingrained in the historical context that they were all but invisible to the authors themselves. The text is taken as carrying information about the circumstances in which it was produced or circulated, in a way reminiscent of how a speaker’s accent may reveal her native language or geographical provenance.22

There are, of course, many unclear and borderline cases, where the finer nuances of the concept of intention come into play. As explanatory attitudes may often be taken as less than fully conscious, with regard to the various agents whose actions or reactions they are supposed to explain, we are approaching the subject of psychoanalytic readings, and other variations of “the hermeneutics of suspicion”23. Psychoanalytic readings, for modern texts, may of course be intended: the author has read Freud and inserted clues to the psychoanalytic reading of her characters. But in many cases – like for example Bruno Bettelheim’s readings of children’s stories –24 intended meaning claims would be rather implausible, and we would be better off looking for other suitable explanatory claims, relating, for example, to the propagation or reception of the story.

V.2 Value Claims

Readings may not only be claimed to be more or less adequate in relation to some historically occurring reading, or to play an explanatory role in relation to the features or the occurrence of a certain text. Readings may also be in themselves more or less interesting, true, elegant, make “good sense”, be funny or boring, and so forth. Claims that readings exhibit such values, to some degree, or that certain readings score better or worse than other readings in such evaluative dimensions – that they “make better sense” or are more true or interesting, as the case may be – I will call value claims.

With historical claims, the relevant reading is claimed to have some type of real presence in the world, the claim looks backwards, so to speak, to situations in which the text has originated or been used. Value claims look in the other direction: forwards to some insight or application to be derived from the text.

A vivid illustration of the difference between historical claims and value claims may be derived from the performing arts. A modern staging of a Shakespeare play often has little regard for historical accuracy, aiming instead for a reading that will be relevant for a contemporary audience. Conversely, in classical music there has been considerable interest in historically accurate interpretations, often played on “original instruments” – that such performances are at the same time appreciated by many as aesthetically superior is an added benefit that does not obfuscate the distinction.

One type of value claim that has often been appealed to in debates about interpretation is claims to the effect that readings are rational or true, with regard to their content. That readings should exhibit such properties is the basic requirement of Gadamer’s “Vorgriff der Vollkommenheit”, and of the “Principle of Charity” often appealed to in the analytical tradition, from Neil Wilson to Quine and Davidson.

Value claims can, of course, be conjoined with historical claims of various kinds, either as mere additions or as parts of arguments for historical claims: if I think highly of a certain author, the fact that a certain reading of her work makes better sense of it than another reading can make it more plausible to attribute it to her as the intended reading; or conversely, one may be reluctant to ascribe a very sophisticated reading of a passage to an author whom one takes to be incapable of comprehending or coming up with it.

While factual claims seem to be natural fits for most scholarly readings, often driven by a historical interest, and with emphasis on what is verifiable and amenable to empirical evidence, value claims seem more naturally at home in contexts where the emphasis is on communication and application. As Gadamer emphasises: the natural home of interpretation and understanding is not where we are trying to learn things about the text and what it means, but where we are trying to learn things from the text, about whatever “Sache” we take it to be about. The sort of expertise needed to find and defend valuable readings, in this sense, is not primarily textual or historical scholarship, but competence in whatever application one may be looking for. To find a philosophically valuable reading of Plato you need to be a philosopher, first, and a historian only secondarily; to find a reading of the law relevant for a particular case you need to be versed in the practice of law.

VI. In the Mixer

The distinction between readings and claims and the parallel separation between elaborations of readings and justifications of claims can elucidate many controversies in the theory and practice of interpretation, but not in the way, of course, that we can expect actual interpretations to propose exactly one type of claim for exactly one reading. I have already alluded to the fact that claims are often not explicit, but left to be inferred from context or default – sometimes leading to unclarity and ambiguity about what claims are actually made. But we can also expect real-life interpretations to present mixtures of different readings and corresponding claims.

One such mixture I have already alluded to, in the section about levels of reading and the relativity of the text. In order to have anything to interpret at the levels of reading that usually occupy critics and scholars, we need to take the reading of lower levels for granted. Generally speaking, the lower the level the more we will stick to (usually implicit) intentional claims – to construe the wording or linguistic meaning of a text with disregard for the author’s intention will almost always be a pointless exercise. The same thing goes, I believe, for narrative interpretation of the sort that seeks to establish what the story is rather than what it means.25

It may perhaps be objected that all of this makes the act of reading a much more complicated affair than it seems to be, particularly with regard to immersive readings of narratives. The phenomenology of reading is a separate story, but I will make two short remarks on it here. The first remark is that on a primary level the process of narrative reading is claimless: we just live the story as it presents itself to us. The second remark is that on a reflective level, readers are able to simultaneously entertain different readings and claims and play them out against each other, without much apparent effort. When thinking about Kurtz, in Conrad’s Heart of Darkness, we have no problem of forming our own view of what he is and what he stands for, while simultaneously reflecting upon what perhaps very different symbolic role Conrad intended him to play, or what his real-life models might have been.

There may also be a mixture of claims for one and the same reading. Look for example at the recommendations of “hypothetical intentionalism”, as formulated by Jerrold Levinson. The reading that the interpreter is aiming for should fulfil the dual requirement of being “the most explanatory plausible and, to a lesser extent, aesthetically charitable construction we can arrive at – regarding a works intended import”26.

In the present terminology this amounts to a combination of an intentional claim with a value claim: the interpreter should seek a reading that is as aesthetically valuable as possible, within the historical constraint of possibly being intended by the author.27 Similar mixtures of historical and vale claims are without a doubt very common. Indeed, most interpretations claiming to be true to the author’s intentions have an apologetic element where it is taken for granted that the proposed reading is not only historically adequate but also valuable in itself.

VII. Summary and Conclusion

I introduced the notions of reading and claim as two, usually not clearly distinguished, components in what is generally called an interpretation. Correlated to this I distinguished between two different activities that are usually lumped together as “arguing” for an interpretation: the elaboration of a reading and the justification of a claim.

In line with this, there are two groups of questions to be answered in connection with any proposed interpretation.

First: What reading is proposed? If it is incompletely specified: how can it be elaborated? What is taken as the text and what features of it are taken as significant, according to what modes of correlation? How does the reading of specific features and elements relate to the reading of the whole?

Second: What claims are made on behalf of that reading? How can they be justified? Are there several different claims involved that should be separately argued – for example historical claims that would demand empirical evidence and value claims that need to be defended on the level of the subject of the reading, “die Sache” in Gadamer’s sense?

1

Kripke famously abstains from claiming that his reading of the private language argument matches Wittgenstein’s intentions, motivating his reading instead through its philosophical interest, although without claiming that the argument as reconstructed is actually valid and sound.

2

Kaehle, Sharon/German, Howards, To the Lighthouse: Symbol and Vision, in: Bucknell Review, 10, May 1962, 328-346. Reprinted in: Beja, Morris (Ed.), To the Lighthouse. A Casebook, London 1970, 189-209, 189.

3

Ibid.

4

Hafley, James, The Creative Modulation of Perspective, in: Beja, To the Lighthouse, 133-148, 137f. The chapter is an extract from: The Glass Roof: Virginia Woolf as Novelist, Berkeley 1954.

5

Overcarsh, F. L., The Lighthouse, Face to Face, in: Accent 10 (1950), 2, 107-123. Reprinted in: McNees, Eleanor (Ed.), Virginia Woolf: Critical Assessments, Vol. III, Mountfield 1994, 493-509.

6

Nicholson, Nigel/Trautmann, Joanne (Ed.), The Letters of Virginia Woolf, Vol. III, London 1977, 572.

7

For plenty of examples involving Virginia Woolf herself, one can consult Hermione Lee’s excellent and exhaustive biography Virginia Woolf (London 1997), where we, among many other things of the same nature, learn that Orlando “is” Vita Sackville-West.

8

Emphasis added.

9

Emphasis added.

10

Currie, Gregory, Interpretation in Art, in: Levinson, Jerrold (Ed.), The Oxford Handbook of Aesthetics, London 2005, 291-306, 295.

11

Beardsley, Monroe C., The Authority of the text, in: Iseminger, Gary (Ed.), Intention and Interpretation, Philadelphia 1992, 24-40, 25 (extract from: The Possibility of Criticism, 1970).

12

Iseminger, Gary, Introduction, in: Iseminger, Intention and Interpretation.

13

David Lewis, Truth in fiction, in: American Philosophical Quarterly 15 (1978), 1, 37-46.

14

I do not intend this to be a complete enumeration of all possible text-meaning pairs, nor of the relevant dependence relations. The early stages of the enumeration are kin to Austin’s distinction between the phonetic, the phatic and the rhetoric speech act. In similar way, Alan H. Goldman distinguishes four hierarchically ordered textual levels, with narrative completion as the highest. Cf.: The Sun Also Rises: Incompatible Interpretations, in: Krausz, Michael (Ed.), Is There a Single Right Interpretation?, University Park 2002, 9-25.

15

Goodman, Nelson, Languages of Art: An Approach to a Theory of Symbols, London 1969, 229f.

16

This is made vivid in Gadamer’s favourite example of text emendation: in philology we sometimes argue for a particular specification of the wording of a manuscript by pointing out that it gives “better meaning” to the text.

17

Gadamer, Hans-Georg, Wahrheit und Methode. Grundzüge einer philosophischen Hermeneutik [1960], Tübingen 1965, 278.

18

Grice, Herbert Paul, Utterer’s Meaning and Intentions, in: Philosophical Review 78 (1969), 147-177.

19

A good example of writing for multiple audiences is described by Paisley Livingston in an extended analysis of Edgar Allan Poe’s “mesmeric” writings. See: Livingston, Paisley, Literature and rationality: ideas of agency in theory and fiction, Cambridge 2008, 70-78.

20

Nicholson, Nigel; Trautmann, Joanne (Ed.), The Letters of Virginia Woolf, Vol. III, London 1977, 385.

21

It is sometimes claimed that some form of inference to the best explanation could yield a general account of interpretation, but I think that idea fails on several counts. First, there are interpretations that are not plausibly construed as explanations of the text or the utterance at all – secondary reading claims belong here, as well as the value claims that I will come to in a little while. Second, there are so many types of explanations that may be invoked in this context, that the notion of “the” best explanation fails to do much useful work.

22

This is some of what Foucault means when he says that modern thought “transforms documents into monuments” (The Archeology of Knowledge, London 1989, 6).

23

See Ricœur, Paul, De l’interprétation: éssai sur Freud, Paris 1965.

24

See Bettelheim, Bruno, The Uses of Enchantment: The Meaning and Importance of Fairy Tales, London 1976.

25

This is not always true: there are cases where narrative interpretation becomes problematic on a literary level. The examples that spring to mind involve potentially unreliable narrators. Are the supernatural elements in Henry James’s The Turn of the Screw fictionally real or figments of the narrator’s imagination? Is the confession of the narrator in Agatha Christie’s The Murder of Roger Ackroyd genuine or intended to protect the real murderer, as Pierre Bayard argues in Qui a tué Roger Ackroyd (Paris 1998)? But even in these cases one generally takes it for granted that the author is the arbiter, even if only, like Henry James, by leaving his intention unsaturated. Even Bayard does not raise the possibility that Agatha Christie herself was mistaken about the identity of the killer, but only that she intended a more sophisticated solution than first meets the eye.

26

Levinson, Jerrold, Hypothetical Intentionalism: Statement, Objections, and Replies, in: Michael Krausz (Ed.), Is There A Single Right Interpretation?, University Park 2002, 309-318, 311.

27

That there will be a space for the aesthetic judgement to exploit is safeguarded by the strictures imposed by hypothetical intentionalism on “admissible” evidence.