Theories of Authorship and Intention in the Twentieth Century. An Overview Dario Compagno University of Siena (dario_compagno@neomedia.it) Abstract This article discusses some of the most important theories about authorship and the author’s intentions developed during the last century. It argues that initially Husserl, Croce and the New Criticism firmly divided private intentions on the one side and verbal meanings (constituting an ideal subject) on the other. Then, it introduces Derrida and Barthes who suggested a radical change in perspective by confuting the existence of an ideal conscious subject, of ideal meanings and of private intentions. Subsequently, Booth and Foucault looked for a surrogate of the author and found it in a discursive instance showing the reader a path to the author’s intentions. Lastly, Anscombe and Eco formulated a new concept of public and open intention completely redefining the whole issue. This article, in conclusion, suggests that, in spite of all statements about the ‘death of the author’, it is precisely thanks to the twentieth-century debate that the author was born. Keywords: Authorship, Intention, Interpretation, Theory, Twentieth Century. And regardless of the extent to which interpreters may decide not to be bound by what authors wanted to say about their artistic intention, or about their true or supposed aim (as many interpreters today claim to do), what they cannot do is get rid of the idea of intention, because without presupposing that idea, the text in question would not even exist; that is, it would not be a literary work. Carla Benedetti, L’ombra lunga dell’autore, 1999 I can know what someone else is thinking, not what I am thinking. Ludwig Wittgenstein, Philosophical Investigations, 1953 1. Preliminaries In the twentieth century the concept of author has undergone a radical redefinition, playing a pivotal role in the studies of language, writing and meaning. The increasing autonomy and importance of human sciences needed to better understand how people and words are related, and how the former can Journal of Early Modern Studies, vol. 1, n. 1 (2012), pp. 37-53 http://www.fupress.com/bsfm-jems ISSN 2279-7149 (online) 2012 Firenze University Press 38 dario compagno control what they wish to say or write. Today some philosophers and literary critics simply dismiss the author and his or her capacity to manage meaning for a conscious end; they tend to pay attention to words alone, almost as if writing were an unintentional or unconscious activity, like dreaming. And it seems that the author’s intentions – that is, what s/he really wanted to say with a text – has been plainly excluded by many philosophical and analytical disciplines (from phenomenology to hermeneutics, semiotics and literary theory). This also seems to be true for those disciplines not dealing with written texts (like art and cinema studies), but that entirely exclude all considerations about authorial intentions from their objects of study. Throughout the century, a number of complex theories on subjectivity and its linguistic and semiotic expression have been elaborated. These have passed through three main phases, each one strongly critical about the preceding one. The specificity of each phase is often ignored, and very different positions like those of Husserl, Derrida and Foucault are believed to exclude the subject in the same way. In this article I wish to highlight and discuss the drastic differences characterizing some of the most important twentieth-century theories about subjectivity and meaning. We will see that, far from having disappeared, the author is today what has to be understood if we want to interpret texts.1 2. Meanings are Not Intentions: the Intentional Fallacy and the Transcendental Subject In the first stage, literary theory on the one hand, and phenomenology on the other, set up a sharp divide between author and text – that is, between intentions and meaning. According to this perspective, it would be totally useless for readers to connect texts with their producers’ will, because will is never really expressed, or it is unknowable, or the meaning and value of writing is unrelated to it. This first phase of research, which elaborated this first absolute denial of the author is exemplified by the notion of ‘intentional fallacy’ of William Wimsatt and Monroe Beardsley (1954), major figures in American New Criticism. For the two critics, intentions do not live along with texts, and texts alone constitute the proper object of literary studies. They famously claim that ‘the design or intention of the author is neither available nor desirable as a standard for judging the success of a work of literary art’ (3; my emphasis). Thus, they make two assertions at the same time: one about the possibil- ity of knowing the author’s intentions and the other about the importance (or, rather, unimportance) of intentions for defining literary value. Theirs is the best-known formulation of contemporary anti-intentionalist criticism, focused on text analysis, and marginalizing biographical evidence, historical understanding and contextual analysis. For New Criticism, works of art are different from ordinary written messages. The intention behind them is fully 39theories of authorship expressed, ‘pure’ in a certain sense: ‘Poetry succeeds because all or most of what is said or implied is relevant; what is irrelevant has been excluded ... In this respect poetry differs from practical messages, which are successful if and only if we correctly infer the intention’ (4). The author’s intention is reduced to the text, and the text only is a valuable object for the critic, also because all other intentions would be practically unreachable: One must ask how a critic expects to get an answer to the question about intention. How is he to find out what the poet tried to do? If the poet succeeded in doing it, then the poem itself shows what he was trying to do. And if the poet did not succeed, then the poem is not adequate evidence, and the critic must go outside the poem for evidence of an intention that did not become effective in the poem. (4) The only way to know more about intentions is through biographical evidence, but it would be a mistake for literary scholars to find literary meaning in biographical episodes and historical contexts (5). In fact, the result of writ- ing can be much richer than what its producer had in mind while writing it. The value of literature lays in the formal configurations words produce by themselves, independently from subjective plans and associations in the writer. Literary value lays in language itself, and it can be reactivated at any moment by accurate formal analyses. Before Wimsatt and Beardsley, Benedetto Croce had argued in favour of exactly the same idea: the mundane intentions (intenzioni mondane) of poets are different from aesthetic intuition (intuizione estetica; 1941, 306-307).2 When reading a poem we need to focus only on poetry, on the core meaning that is worth appreciation, while avoiding what was contingently introduced by the poet (that is, what was due to the poet’s practical intentions and not to the pure aesthetic idea): ‘the intentions and ends of poets necessarily re- main extraneous to poetry; it does not matter what the poet means, wants or believes s/he is doing, but only what s/he actually does, however unaware or in contrast with the professed end’ (306; my translation). We can find in Logische Untersuchungen by Edmund Husserl (1900) the theoretical grounds for this divide between texts and intentions. Hus- serl formulated an essential distinction: the experiences of those who speak and write are not the same thing as the meaning of the words used (1970, 187-189). Words have a meaning-intention (Bedeutungsintention) that is autonomous from the speaker’s intention (Absicht or Intention). Intentions are part of the speaker’s experiences: they are the reasons why one speaks or writes, in view of reaching a practical end in a given situation. Intentions, for the philosopher, are private mental states, related to the contingent situation of utterance and they cannot be communicated to others. On the other hand, the meanings (Bedeutungen) of words and sentences are ideal: that is, autonomous from the private ends and feelings of those who use them, 40 dario compagno and from the actual circumstances of use. A word like ‘house’ has an ideal meaning, independent of all personal associations that the speaker or writer can feel, and from any reference to concrete houses (that may, or may not, be in front of the person pronouncing the word). This ideal meaning gets communicated to others: whenever I hear or read the word ‘house’, I grasp its ideal meaning in the same way as the speaker or writer does (we share the same Bedeutungsintention; 194). Later on, in his Ideen (1913), Husserl finds in this ideal dimension of meaning the most conscious and clear experience we have. We are perfectly aware of what we want to say in a given moment, and words themselves per- mit us to think clearly about objects and states of affairs: to see something as a ‘house’, means that it corresponds to the typical ideal house, and we recognize the occurrence thanks to our previous knowledge of the type. Our being awake and conscious of living a certain present, of thinking certain thoughts, is possible only thanks to the ideal meanings of words coming to our mind. Ideal meanings therefore constitute a conscious subject that has a full control of what s/he is and wants to say. The transcendental subject is an ideal speaker, with a perfectly clear will to speak; it has no private experiences and intentions, but thinks only pure concepts (Bedeutungsintentionen): ‘I, the “transcendental ego”, am who “precedes” anything worldly: as the Ego, that is to say, in whose life of consciousness the world, as an intentional unity, is constituted to begin with. Therefore I, the constituting Ego, am not identical with the Ego who is already worldly, not identical with myself as a psycho- physical reality’ (1969, 238). Therefore Husserl ‘builds up’ the ideal speaker, perfectly logical and totally expressed in the words s/he uses. In this way, his philosophical analyses of language aim to grasp all that is available and all that is important: this is the same assertion defended by New Criticism for the analysis of literature.3 To use the words of Paul Ricoeur, ‘reduction is the philosophical act that permits the birth of a being for meaning’ (1974, 246). New Criticism focuses on literature (literary language), while Husserl writes about the subject and the possibility to communicate (language); two very different dimensions, that will continuously be kept together during the whole century. In short, for this first phase of research on subjectivity intentions are private and contingent: they are important for everyday life, because they are linked to feelings and to the contexts of utterance, but also impossible to communicate to others. Language allows a private mental state (in the utterer or writer) to set up a link with another private mental state (in the listener or reader) because meanings are independent from private intentions: meanings live in sentences and texts, and resist time. Only these meanings are the real ‘will to say’ of who speaks and writes; they are what a person is fully aware of communicating, and they are the only thing valuable for literary criticism and for the analysis of thought. Private intentions, on the other hand, play a residual and marginal role for theory. The subject of writing is therefore 41theories of authorship an idealized instance, with an abstract will (Bedeutungsintention) capable of thinking all and only the ideal meanings. 3. There are No Ideal Meanings: the Death of the Author and Deconstruction Let us focus on a radical change, reversing what we have seen until now. Pro- tagonists of this second phase of theorization are Roland Barthes and Jacques Derrida. Around 1968, French critics receive the innovations proposed by New Criticism, but swiftly overturn their sense (while on the surface they are just radicalizing them). The ‘text’, for the Americans something that can undergo an objective analysis, becomes with Barthes the place for anarchical readings, rooted in linguistics, psychoanalysis and anthropology.4 Barthes’s reading practice has nothing in common with formal objective analyses. Ap- parently, Barthes (1967) restates that the author’s intention should be kept away from the critic; actually, he begins a sort of analysis of the subject in his or her writings. We should point out that Barthes’s proper interest is literature: for Bar- thes contemporary works have to be read as if they did not have an author.5 The author’s intention is just an undue limit to the reader’s freedom: ‘To give a text an Author is to impose a limit on that text, to furnish it with a final signified, to close the writing’ (1977, 147). There is no reason to confine readings within what we suppose the author wanted to say. There is no need for a ‘master of meaning’, as Barthes writes, because literary value lays in its potential to stimulate creative thoughts in the reader: ‘the birth of the reader must be at the cost of the death of the Author’ (147). If we want to really understand Barthes’s perspective, we need to use extreme care: Barthes’s author is very different from the author depicted by New Criticism. Then, the author’s intentions were the idiosyncratic variation against which one should look for an ideal and objective meaning; now, the author is the ideal and objective meaning, dull limit that has no value for criticism. So the word ‘author’ has completely changed sense, and Barthes’s meanings are at odds with Husserl’s. The latter looked for a perfectly object- ive, durable meaning, removing all subjective and contingent intentions, while Barthes looks for an ever-changing meaning, removing all well defined, organizing intentions. The ideal stability Husserl and American criticism looked for is now considered fake and undesirable. What matters in critical readings are those details observed after passing over simple sense and moral, showing hidden aspects of the psyche or society. It is not the pure will of the conscious sub- ject that is sought, but rather the traces of its unconscious. Unconscious that manifests itself in writing and society, because it is language that structures the unconscious. The ideal conscious subject constituted by Husserl and the objective meanings he should be fully aware of are ‘dead’. 42 dario compagno To say with Barthes that ‘the author is dead’ means exactly the opposite of raising a sharp distinction between meanings and intentions. The ‘intentional fallacy’ and phenomenology meant to divide objective meaning and private intentions: take the first and drop the rest. But if there are no objective meanings on the one side, we cannot even think of private intentions on the other. What was hidden in the individual – and above all to the individual – is now visible in texts. The result of writing is as messy as real thought. The ‘death of the Author’ is the end of an ideal consciousness that says only what it wants to say, and also all objective analyses of texts do ‘die’, because texts are not constituted by ideal meanings. How should we read, then? The critic follows traces and hints, reaching meanings and thoughts that could have been in the author. The most important thing is that the critic should never say s/he has found the only correct meaning (what Husserl and New Criticism actually aimed to do with their ‘objective’, formal analyses). Interpretation is mere guessing. Derrida (1967) has a very similar insight and uses it within philosophical discourse to overturn phenomenology, attacking the essential distinction between ideal meanings and private intentions. For Derrida there is no ideal dimension of language – for example a house autonomous from the actual thoughts of the person saying the word ‘house’, and from the real houses in which we live. There are no ideal meanings: signs tend to stable meanings (the word ‘house’ tends towards a shared idea), but signs are made of empirical uses, with all their imperfections and variations due to the contingent situation of use. No matter how we try and purify linguistic use through repetition, we will never reach a totally ideal and detached dimension of meaning. Bernet writes: ‘The entanglement (Verflechtung) of the expressive function and the indicative function [of ideal meaning and intentions] of the same sign is for Husserl only an accidental contamination … Derrida on the contrary holds this entanglement of the expressive and indicative function of signs to be essential’ (1990, 253). There are no ideal and objective meanings, and there is no ideal conscious- ness able to grasp them. The conscious subject is something very different from what Husserl thought. Signs build experience, and it is only thanks to their repetition in actual uses that we can think; therefore we humans live in balance between the ideal dimension of a pure consciousness (with an extremely clear insight on its ‘will to say’) and the real chaotic variation of unconscious drives, of the raw ‘will’ about which we do not know much. Husserl ‘stepped out’ of reality, setting an abstract standpoint from which he was able to partition ideal meanings (good) and lived intentions (bad). For Derrida we cannot make this ‘step out’, assuming the existence of ideal meanings from the start. Whenever we say or write the word ‘house’ we let our listener or reader access a net of links (deferments), connecting in a heterogeneous way memories, references and past uses. We cannot isolate an ideal dimension of meaning that excludes all contingent references, leaving only pure thought. Therefore, for Derrida there are also no private intentions, Husserl’s lived residuals of ideal meanings. 43theories of authorship If we cannot tell what an ideal ‘house’ is, we cannot even tell what private, idiosyncratic associations the speaker is living close to it: how can we tell what intentions are in the speaker but not in the ideal meaning of the word? How can we tell what shape the background has, without a standing figure? Barthes and Derrida undermine phenomenology and New Criticism. Those ideal meanings that had to exist in order for an objective analysis of consciousness and texts to be possible, have been ruled out. And consequently there are no residual lived intentions ‘in the author’ that had to exist and be subtracted to obtain ideal meanings. The will to say escapes in an indefinite deferment. And so the practice of reading has changed accordingly: decon- struction is the never-ending quest for details giving access to new traits of the writer’s unconscious. Passing from the first to the second phase of research we have passed from the primacy of consciousness to that of the unconscious. Formal analyses neglected all that was beyond the ‘text’ (maybe a revealing name or date, significant for the author’s life); now Deconstruction neglects any stable core of meaning, what the author most clearly wanted to say, and puts on the same level of relevance what the author did want to write and what the reader adds by him- or herself: ‘no longer reduced to a “single message”, the text is opened to an unlimited variety of interpretations’ (Burke 1998, 43). 4. The Need for a Surrogate: the Implied Author and the Author-Function It would be a mistake to think that twentieth-century research on subjectivity and writing ended with the ‘death of the Author’. Barthes and Derrida just opened the possibility of studying the author (Burke 1998; Benedetti 1999; Irwin 2002; Bennett 2005). We can study the author because intentions are not inaccessible experiences, but are actually expressed in language use. We must study the author because any analysis trying to do without it runs the risk of looking for an ideal meaning in vain. Husserl’s argument was that, if private intentions are unattainable, then we should look for an ideal author; Derrida grounded the inverse argument: if ideal meanings are unattainable, then we should look for an empirical, fallible author, with a will to say s/he does not fully control. As Sean Burke writes, the ideal, hypertrophic author has finally left room for a human author, with all its limits: ... what is put to question is the absolutely determinative hegemony of intention over the communicative act. Intention is to be recognized, and respected, but on the condition that we accept that its structures will not be fully and ideally homogeneous with what is said or written, that is not always and everywhere completely adequate to the communicative act. … Intention is within signification, and as a powerful and necessary agency, but it does not command this space in the manner of an organizing telos, or transcendental subjectivity. (1998, 140) 44 dario compagno The matter seems then to be how the author’s intention emerges in texts. I will now rapidly highlight the way in which some scholars have searched for a new concept of author. In the first place we will look at Wayne Booth and Michel Foucault, who tried to define an Ersatz for the concept of author: a textual or discursive instance showing to the reader a path to the author’s intention. Then I will look at two more radical thinkers, Elisabeth Anscombe and Umberto Eco, who completely redefined the very idea of intention. Foucault (1969) makes a function of discourse out of the author. He recognizes the need of having a person next to the text. As a matter of fact, whenever we read something, we need to construct ‘a certain unity of writing’ (1984, 111) behind it. But the subject does not have an origin before language: consciousness has its historical conditions, and the author is the result of cultural production and interpretation. Text and people are built together in interpreta- tion, step after step: ‘The author’s name serves to characterize a certain mode of being of discourse … that must be received in a certain mode’ (107).6 For Foucault, it is important to stress that the author cannot be found simply within the text, but on the contrary any text needs to be linked to its particular cultural context, to certain definite historical practices of meaning production, only within which does it have sense. A word like ‘house’ has a meaning only with respect to certain social regularities of use. The so-called author-function is then the interface between a text and the system of other relevant texts in which it is produced. It would be useless to read a text detach- ing it from the culture in which it was produced: it would only apparently show its ‘real meaning’, while its words would actually not be worth much. A text has value as long as it is understood as an element of large cultural regularities (these regularities are what really interests Foucault). This is also why the single author should not be over-emphasized; texts ‘have to be writ- ten’ and their conditions of appearance are easier to understand by referring to larger social trends, than to the biography of a single man: ‘In short, it is a matter of depriving the subject … of its role as originator, and of analysing the subject as a complex and variable function of discourse’ (118). Foucault stresses the role of the interpreter in constructing the author and the role of cultural regularities in shaping the subject itself, making of it just the superficial appearance of deeper social and historical processes of change. The author function ‘does not develop spontaneously as the attribution of a discourse to an individual. It is, rather, the result of a complex operation which constructs a certain rational being that we call “author” ’ (111). The main question is then: ‘How, under what conditions, and in what forms can something like a subject appear in the order of discourse?’ (118). The author is again on the agenda, just a couple of years after Barthes’s dismissal. What does it matter who is speaking? ‘Foucault’s answer is that, in fact, it matters very much who speaks, or who we think is speaking’ (Bennett 2005, 19). 45theories of authorship Foucault explains in this way the constraints on the author’s will, but he does not focus on the means by which a person manages to take partial control over what s/he says or writes. It almost seems that unintentional social structures produce texts by themselves, without the real need of a conscious human being. But texts are intentional products: we cannot but read them as being so. Structure alone does not speak: without an effective author we could not give any meaning to texts. The author-function is the sum of all constraints to writing; for this reason it is only the ‘negative half ’ of the human author. Wayne Booth (1961) focuses instead on the ‘positive half ’ of the author, elaborating on his masters’ ‘intentional fallacy’. Texts are intentional products, the results of series of choices: ‘all art presupposes the artist’s choice’ (53). Therefore the author expresses a writing strategy whenever s/he produces a text, and readers need to reconstruct this strategy from the words and sen- tences used (otherwise they could even read pages in random order). If we read the word ‘house’, this is evidence that its author chose to use this and not another word, and more so ‘the novelist who chooses to tell this story cannot at the same time tell that story’ (78). While reading, we find a sense in what is written by attributing to words and sentences the nature of items that have been chosen among alternatives. According to the choices we find in the process of reading, we gradually attribute an intention to the author. In this perspective, Booth defines the implied author as ‘the sum of his own choices’ (74)7, whose effect is the text we are reading. The author we need for understanding a text is just this series of choices; but a choice cannot simply be found within the text. To buy an old Fiat Cinquecento today, or in the Sixties, are very different choices: today it is the decision of a connoisseur or collector, while in the sixties it would just have been an ordinary selection among other cars sold. The object Cinquecento remains the same with time, but its meaning (the meaning of choosing and buying it) changes depending on how we think of the strategy behind it, its reasons and ends. Words can be reiterated an indefinite number of times, but it is their place in a textual strategy that is most important. In Booth’s perspective, meaning is neither the ideal dimension described by Husserl (detached from any communicative and contingent aim), nor an unmanageable expression of the unconscious (language is not ‘speaking us’). To speak and to write are conscious activities realized for some conscious end, however bound to cultural constraints. Most importantly, the intentions of an author can be seen in his or her works. Sheldon Sacks writes that ‘in making each choice, his ethical commitments, intuitive or conscious, not merely were but had to be revealed’ (1964, 254). Thanks to Booth and Foucault, the author acquires a new silhouette. They begin to look for a new concept of author, an author on the border between consciousness and the unconscious. This is the beginning of the third phase in the research on subjectivity and language: now the aim is a synthesis of 46 dario compagno the pure phenomenological consciousness and of the structural unconscious. But why do Foucault and Booth make use of the neologisms author-function and implied author, instead of referring simply to the author of a text? The two thinkers still appear reluctant to recognize for the subject in itself a fully linguistic and semiotic nature.8 Foucault and Booth implicitly rely on another figure, hidden inside the individual: a ‘real author’ that analyses cannot reach. Expressions of subjectivity in language only have a derivative existence, while there is an ‘original’ subject that cannot be found in texts. They still imagine a ‘real author’, an unexpressed intention beyond texts. For this reason they look for surrogates: however far we go, we will never reach the person behind the work. The image of the author emerging from a text is necessarily different from what the author really is. The grounding idea of phenomenology sur- reptitiously resists: there is a private, immediate intuition that can be reached only ‘from the inside’, while it remains inaccessible to others. Words cannot express this intimate conception of oneself, this private intention (the ‘real thought’) preceding writing. This degree zero of subjectivity has escaped from all the attacks of theory, and becomes visible in sentences like ‘this is not what I really meant’, or ‘her/his real intentions are lost’. 5. The Public Nature of Intention Charles Sanders Peirce had written, about a century before (1868), that in man everything is sign. But if this is true, how then could intentions not be signs? The way in which we think and want depends on the public signs al- lowing us to do so. The last step in twentieth-century theories of subjectivity recognizes the semiotic nature of intention, the fact that even what is most intimate in man is made of signs, and therefore lives a public life. There is no hidden ‘real author’, no inaccessible subjectivity hidden from language and from others. All of the man is in the work. Jacques Bouveresse calls the extremely diffused and persistent disposition ‘to grant an exorbitant explicative power to some events that we say are inte- rior, hidden, private, etc.’ (1987, 694; my translation) the myth of interiority, a disposition that seems to have survived all of the attacks of the twentieth century. For Bouveresse, it is Ludwig Wittgenstein (1953, 1958) who first overcame the ‘myth of interiority’ and made the subject’s intention completely public and sharable. Wittgenstein writes that: ‘Intention is neither an emo- tion, a mood, nor yet a sensation or image. It is not a state of consciousness’ (1967, § 45). Intentions are formed within a human community that uses language and that has certain habits; we cannot want and act, except within ‘enormously complicated tacit conventions’ (Anscombe 1957, § 43). We are what we are because we are put into a certain form of life by language. And above all: ‘Exactly like Peirce, Wittgenstein denies that the immediacy and the private nature of somebody’s conscious contents can justify by themselves 47theories of authorship the knowledge s/he has of these conscious contents … and a fortiori that this immediacy and this private nature must be considered as a criterion of all authentic knowledge’ (Bouveresse 1987, 78, my translation). Wittgenstein, Bouveresse argues, is, for this reason, the ‘anti-Husserl’, subverting the priority of private intuition on public expression (22). And it is Elisabeth Anscombe (1957), Wittgenstein’s student, who grounds an explicit theory of action onto what is publicly done and interpreted. To Anscombe, if we want to understand the sense of an action – or of the product of an action, as a text – we need to ask why (for what ends) the agent acted in that certain way. Action is characterized by means-ends reasoning: we give meaning to actions because we are able to identify a red thread, a chain of consequential ends, towards which action is directed. And we do identify these ends, at every interpretation, without ever going ‘inside the mind’ of anyone. Anscombe puts intentions back at the core of action descriptions, giving to intentions an exclusively public nature. Intentions are not things, res, hidden in the individual and impossible to fully communicate with words or public acts. Intentions only exist for someone, in public acts, from a point of view and under a description: ‘to call an action intentional is to say it is intentional under some description we give or could give of it’ (§ 19). This is completely the opposite of what common sense and phenomenology say about intentions: private, immediately clear intuitions we have before acting, that necessarily remain in the mind and do not get communicated to others. It would be misleading to think that there is an ultimate ground of our behaviour that cannot be explained to others. If there were one, how could we explain it to ourselves too? Everything that a person can reach from within, can also be reached from the outside, because what puts it into a conceivable form is language.9 To understand ourselves and others, we need to find an end towards which one acts – end that lives in the acts we perform and not in the brain. The linguistic and semiotic nature of intentions makes them exist only ‘at a certain distance’, for people able to use the specific grammar of means and ends: The only description that I clearly know of what I am doing may be of something that is at a distance from me. It is not the case that I clearly know the movements I make, and the intention is just a result which I calculate and hope will follow on these movements. (§ 30) Oedipus kills Laius, and does not know that Laius is his father. Can we really accuse him of having intentionally killed his father? The agent’s point of view is essential for attributing meaning to his or her actions and to the results of those actions: why did Oedipus kill Laius? The same applies to writing: to understand a text we need to give an intentional role to words, characters, events, reconstructing the writing act. The ‘myth of interiority’, the belief in 48 dario compagno unknowable intentions, today should be refuted also in reading practices. We need to ask why the author produced this text and not another, understanding the reasons and constraints behind this choice. The teleological question why is the core of a new way of conceiving the author. The author is a series of answers given to the question why, asked about a text from many possible perspectives. Without this question a text would not be such – it would not be an intentional product (as Compagnon 1998 and Benedetti 1999 write). 6. The Intentional and the Unintentional Theoretical simplifications in the nineteenth century attributed to the author either a precise conscious plan (well defined ‘in the head’ up to the smallest detail) or a messy net of idiosyncratic associations (over which s/he has no control). What emerged in the end is the need for an author, for a real person who could be held responsible for the sense of writing, and could therefore confute inappropriate interpretations. The author shows up in interpretation well before biographies and in genetic comparisons: it is a matter of seeing the result of human action in literary and artistic works. As we have seen, the first phase of approaches to the author implicitly postulated a hidden, interior intention that could not be reached by analyses. The ‘real intentions’ of writing were confined to an unreachable subjectivity, and the public aspects of writing were consequently belittled, given that they could not point to the hidden ‘real intention’ in the author. But, actually, in interpretation anything is better than nothing, and some indications about the author’s intention have been recognized to be available and needed. If this is true, then it is better to build up an author as best as we can. A pragmatic approach to the author recognizes the fallibility of interpretation, but values all clues that can help to understand intentions in the act of writing itself, and so to see alternatives and choices in the words and sentences actually used. The twentieth-century hunting season for the author showed above all that a minimal meaning attributed to a minimal author is very hard to find and define. On the other hand, the ends of speakers and writers do emerge from the historical and communicative context: they are not unavoidably lost after the author’s physical death, because they never lived inside the author in the first place. Authors live in their works. There was in Dante no hidden res, there was no private ‘real intention’; if there were one, then we would be justified to stop looking for it (given its disappearance), and only concentrate on language. But intention is public, and lives in the acts performed: it does not disappear, or all meaning would also disappear at the same time.10 There- fore, given (1) the need for an author, (2) the availability of intentions, (3) the difficulty in building a minimal author and meaning, then (4) we need to understand the author as best as we can, by all means necessary. Otherwise we would risk making the author say whatever we preferred. As a matter of fact, 49theories of authorship whenever we build up a formal or minimal author, we are just projecting our intention into the text.11 Referring to the author, on the other hand, permits us to evaluate the correctness of our interpretation. Luigi Pareyson blamed Croce because for the latter the productive act had no value per se, and was only a means to express the universal idea behind the work. Pareyson had a very different theory about artistic creation: it is human action that gives artistic works their value, and ‘it is necessary to consider art more as a “doing” than as an “expressing” or “contemplating” ’ (1974, 7; my translation). The artistic intention emerges in interpretation, and the interpreter needs to reactivate the author’s intention in order to understand the work. Intentions are formed along the creative process and persist in the work’s form. The matter is then how to understand the intentions of an author that make a work of art out of wood, or a piece of paper and some ink. Umberto Eco elaborates on Pareyson’s theory. Eco looks for an inten- tion also in those open works whose meaning apparently does not depend on their author’s project, either because random processes of creation played a major role in their production, or because the author left in them an ample space for interpretative freedom. Eco’s theory is far from Barthes’s and Der- rida’s ideas. For Eco, we need to look for and find an intention in order to recognize a work of art as such. Art, in other words, is always intentional, although some relevant parts of its expression and/or meaning can be left open to the interpreter’s freedom. The author of an open work ‘wanted to provoke a feeling of suspension, of indeterminateness’ (1989, 116). Not all works of art are open works; but there is also an author’s intention in open works: the intention to make an open work, in a certain way and for a certain end. Any work, ‘even before becoming a field of actualizable choices, is already a field of actualized choices’, and ‘the original gesture, fixed by and in the sign, is in itself a direction that will eventually lead us to the discovery of the author’s intention’ (101-103). To interpret is then a matter of cooperating with the author’s intention, as it is put into form in the work (Eco 1979). For Barthes and Derrida the author’s intention was a dull limit that the critic had to avoid, while Eco looks for an intention that is public (intentio operis, as in Eco 1990), and not ‘in the head’ of the author, defined up to the smallest detail before even taking up the pen (intentio auctoris).12 This public intention leaves some freedom to the interpreter, an open field of possibilities, but not complete freedom. It may happen that a work does really escape from the author’s control and begins to produce meaning by itself, but in that case ‘what remains then is no longer a field of possibilities but rather the indistinct, the primary, the indeterminate at its wildest – at once everything and nothing’ (1989, 93). Eco’s intention is perfectly compatible and complementary with Anscombe’s. Intentions are semiotic realities, existing only within our public interpretation of actions and texts. Without intentions we could not give form to acts and texts (that are both the result of intentional activities), and intentions 50 dario compagno are neither hidden res in the brain nor private experiences accessible only to oneself. Moreover, intentions can be (and for the later Eco always are, at least in part) open; whenever we act, our end is never defined up to the smallest detail, ‘because the intention itself was open, aiming at a plural communication’ (102; my emphasis). Openness is a property of language and of thought, depriving the will-to-say of the conscious perfection wanted by Husserl, but granting a partially-conscious directive power to it. We are not fully conscious of what we mean, whenever we speak or write, but we are conscious of at least some of the sense effects we want to obtain by speaking and writing. The consequences of writing are only partially foreseen, and writers try to understand at their best the ways in which interpreters will read their works. This indeterminacy does not make our intentions less legitimate: that of a perfectly defined, ‘crystalline closed’ intended meaning in the mind is a wrong idea. What we think is analogous to what we write, with all the uncertainties and blank spaces (intentionally or not) left to the reader. Anscombe and Eco show that intentions are public and open, against the two phenomenological ‘myths’ of a hidden interiority and of a perfect definition. Barthes and Derrida could not recognize that something in the text is intentional and something else is not, otherwise they would have had to admit that the (however polysemic) author’s intention remains in the text and in its interpretation. Anscombe and Eco actively look for clues to better interpretations and, when doing so, find an intention at work. Thus, the later developments of twentieth-century reflection on subjectivity and language completely overturned the original statements about an ‘intentional fallacy’. Interpretation is a matter of recognizing what is intentional: where the work begins and ends, and what is planned within it. There is also the need to recognize what is unintentional, above and beyond the author’s intention – otherwise the author would become a god-like entity with a perfect control over language and with an unrealistically clear idea of the ends and consequences of his or her actions. In practice, while interpreting a text, we need to distribute intended and non-intended messages: the way in which we realize this distribution determines the resulting meaning. For example, it is very different to understand a certain sentence as ironic or not; it depends on how we give shape to the author’s intention, on the basis of all the data we manage to find and to our reasoning. Attributing or not-attributing an intentional ironic message to a sentence depends on the hypothesis we make about the author (while the sentence itself remains the same in both readings). For example, to talk about our friend’s mismatched socks can be done intentionally or not: it is all important to understand what is the case, in order to interpret the actual sense and use of what we say. And this applies also in case of intentional ambiguity. This is the exact opposite of asserting that the author’s intention is ‘neither available nor desirable’, as New Criticism suggested. Intention is always at stake and in discussion. The author cannot be exhausted, fully understood, because interpretation (from a pragmatic standpoint) is a nearly never-ending activity. 51theories of authorship But intention remains the only limit to interpretation and the only possibility to negotiate our proper reading with others on a public basis. Intention is actually the opposite of what Husserl and the New Critics thought: intention is an entirely public benchmark, to be defined according to textual and contextual evidence against which diverse interpretations can be compared. Intention is public and visible, and not something hidden within the author. To be an author (and also to simply communicate) is a matter of publicly expressing something, and not at all keeping it private for oneself. There is nothing unreachable in communication, and especially not the author’s intention. I hope I have given an adequate idea of the richness and complexity of the debate on authorship and interpretation as it developed throughout the whole of the twentieth century. The subject, initially marginal, took a long time to become visible and describable. And every new idea of the author has been accompanied by a new idea of what a text is, and of how it should be read: realization of ideal meanings under the control of a transcendental sub- ject; agglomerate of citations under the drives of the unconscious; intentional product of an author with a partially conscious strategy in mind. Only this last idea needs an effective, ‘useful’ concept of author: a person writing for an end, even if s/he may not be able to foresee all the consequences of his or her choices. This makes a human out of the author. In this way the word ‘author’ can be pronounced again without prefixes or restrictive attributions, without seeing in it the residual par excellence, that about which it is not possible to talk. There is no hidden author. A scapegoat for an entire century, the author is actually its dearest son, a linking of consciousness and the unconscious. It was only following the twentieth-century debate that the author was born. 1 This article presents some of the results of my PhD thesis, discussed at the University of Siena in May, 2010. 2 Wimsatt and Beardsley (1954, 6) completely misunderstood Croce’s theory and in particu- lar ignored his ‘golden rule’ of dividing mundane intentions from the pure aesthetic intuition (see also Benedetti 1999, 41). Croce did not attribute the original formulation of the ‘golden rule’ to himself, but to the Italian critic Francesco De Sanctis and to Pëtr Kropotkin. New Criticism gave resonance to an approach that was well-known and recognized in nineteenth-century Europe. 3 Eric Hirsch proposes a form of intentionalism explicitly rooted in Husserl's phenomenol- ogy. Hirsch is only interested in meaning-intentions (Bedeutungsintentionen), and not in lived intentions (Absichten). From our perspective, Hirsch’s position adds nothing to the ‘intentional fallacy’. Hirsch’s author is an ideal author (literary transposition of the transcendental subject), emptied of lived intentions, and fully aware of objective meanings independent from historical change: ‘an unlimited number of intentional acts can intend the same verbal meaning’ (1967, 38). 4 Important sources for the ‘death of the Author’ proposed by Barthes, and for French post-structuralism in general, are Lacan (1966) and Benveniste (1966). 5 Understanding the real extension of Barthes’s ‘death of the author’ would require much more space than is allowed here. Let us simply say that it is not fully clear whether Barthes (1967) refers to all literary works, to all contemporary literary works, or to some contemporary literary works. 52 dario compagno 6 Foucault’s move drastically changes the perspective in which we observe texts and authors. It is a matter of viewing authors as functions of texts, by which we ‘build’ and understand what we read. On the contrary, texts have been usually thought as ‘signs’ of the author’s genius and creativity. As H.W. Gabler makes very clear in his contribution to the present volume, this revolution in perspective represents in a much better way how scholars actually deal with texts and promotes further effects in contemporary critical and interpretative practices. The concept of text itself emerges as a literate construction involving material documents on the one hand, and hypotheses about the author’s work and will on the other. 7 Booth’s author is implied and not implicit. Implied by the very existence of the text, as my coffee on the table implies that someone made it. Explicit and implicit messages are both to be referred to the author. 8 Actually, Foucault’s position can be interpreted as saying that the ordinary concept of author has to be substituted by the author-function, meaning that there is no hidden ‘real author’ but only an effective semiotic figure built by discourses. This interpretation would imply that Foucault also held that the author function manifests all of the author’s intention and ‘interiority’, as Elisabeth Anscombe more explicitly stated (see section 5). 9 Derrida (1967) complained that Husserl did not take the distinction between linguistic mean- ing and intuitive knowledge far enough. Language was important for Husserl (it shaped thought), but not important enough, and private intuition kept the ruling role. Wittgenstein and Anscombe, on the other hand, attributed overriding importance to language, and left no knowing capacity for private intuition. To know what I think, I need to listen to what I say: this idea is common to Der- rida and Wittgenstein. There are many other analogies between the two thinkers (see Staten 1985). 10 Whenever we become unable to read an ancient language, we lose at the same time the texts’ meaning and their authors’ intentions. 11 Eco (1990) calls this intentio lectoris. For example, if my wife tells me that she’s going to buy some cigarettes, and I believe this to be a subliminal message and that she will never return, very often I am just reading in the words something that is not there. A reference to the situation of utterance, and to our life in general, would be the only way to guide interpreta- tion (in fact, every sentence is potentially ambiguous, but not all are intentionally ambiguous; see Eco’s openness). 12 Compagnon (1998) believes that Eco’s intentio operis is just another way of talking about texts (against the complete freedom accorded to the reader by deconstruction) without including their authors. But Eco’s pragmatic theory of interpretation could not do without the author’s intentions, strategies and choices. The intentio operis is Eco’s proper way to define a purely semiotic intention, existing only in operari and not in mentis. Works Cited Anscombe Elisabeth (1957), Intention, Oxford, Blackwell. Barthes Roland (1967), La mort de l’auteur, Paris, Éditions du Seuil. Engl. trans. ‘The Death of the Author’, in S. Heath, ed. (1977), Image-Music-Text, London, Mill and Wang, 142-148. Benedetti Carla (1999), L’ombra lunga dell’autore, Milano, Feltrinelli. Bennett Andrew (2005), The Author, London and New York, Routledge. Benveniste Émile (1966), Problèmes de linguistique générale, Paris, Gallimard. Bernet Rudolf (1990), ‘Derrida et la voix de son maître’, Revue philosophique de la France et de l’étranger CLXXX, 2, 147-168. Engl. trans., ‘Derrida and His Master’s Voice’, in Z. Direk and L. Lawlor, eds (2002), Jacques Derrida. Critical Assessments of Leading Philosophers, vol. I, London and New York, Routledge, 250-266. Booth Wayne (1961), The Rhetoric of Fiction, Chicago, The University of Chicago Press. 53theories of authorship Bouveresse Jacques (1987), Le mythe de l’intériorité, Paris, Minuit. Burke Sean (1998 [1992]), The Death and the Return of the Author, Edinburgh, Edinburgh University Press. Compagno Dario (2010), Attorno alla ‘morte dell’autore’. Indagine sulla soggettività nelle teorie semiotiche del Novecento, unpublished PhD dissertation, University of Siena. Compagnon Antoine (1998), Le Démon de la théorie, Paris, Seuil. Engl. trans. (2004), Literature, Theory, and Common Sense, Princeton, Princeton University Press. Croce Benedetto (1941 [1936]), La poesia, Bari, Laterza. Derrida Jacques (1967), La voix et le phénomène, Paris, Presses Universitaires de France. Eco Umberto (1962), Opera aperta, Milano, Bompiani. Engl. trans. (1989), The Open Work, Cambridge, MA, Harvard University Press. Eco Umberto (1979), Lector in Fabula, Milano, Bompiani. Engl. trans. (1979), The Role of the Reader, Explorations in the Semiotics of Texts, Bloomington, Indiana University Press. Eco Umberto (1990), I limiti dell’interpretazione, Milano, Bompiani. Engl. trans. (1991), The Limits of Interpretation, Bloomington, Indiana University Press. Foucault Michel (1969), ‘Qu’est-ce qu’un auteur?’, Bulletin de la Société française de philosophie, 63, 3, 73-104. Engl. trans., ‘What is an Author?’, in P. Rabinow, ed. (1984), The Foucault Reader, New York, Pantheon, 101-120. Gabler H.W. (2012), ‘Beyond Author-Centricity in Scholarly Editing’, in this volume. Hirsch Eric (1967), Validity in Interpretation, New Haven, Yale University Press. Husserl Edmund (1900), Logische Untersuchungen, vol. I, Leipzig, Veit. Engl. trans (1970), Logical Investigations, vol. I, London and New York, Routledge. Husserl Edmund (1913), Ideen zu einer reiner Phänomenologie und phänomenologische Philosophie, Den Haag, Nijhoof. Engl. trans. (1982), Ideas Pertaining to a Pure Phenomenology and to a Phenomenological Philosophy, The Hague, Nijhoff. Husserl Edmund (1929), Formale und transzendentale Logik, Den Haag, Nijhoof. Engl. trans. (1969), Formal and Transcendental Logic, The Hague, Nijhoff. Irwin William, ed. (2002), The Death and Resurrection of the Author?, Westport, Greenwood. Lacan Jacques (1966), ‘Subversion du sujet et dialectique du désir dans l’inconscient freudien’, in id., Écrits, Paris, Seuil, 793-827. Pareyson Luigi (1974 [1954]), Estetica, Firenze, Sansoni. Peirce C.S. (1868), ‘Questions Concerning Certain Faculties Claimed for Men’, Journal of Speculative Philosophy 2, 103-114. Ricoeur Paul (1969), Le conflit des interprétations, Paris, Seuil. Engl. trans. (1974), The Conflict of Interpretations: Essays in Hermeneutics, Evanston, Northwestern University Press. Sacks Sheldon (1964), Fiction and the Shape of Belief, Chicago, The University of Chicago Press. Staten Henry (1985), Wittgenstein and Derrida, Oxford, Blackwell. Wimsatt W.K. and Monroe Beardsley (1954), ‘The Intentional Fallacy’, in The Verbal Icon, Lexington, Kentucky University Press, 3-18. Wittgenstein Ludwig (1953), Philosophische Untersuchungen, Oxford, Blackwell. Wittgenstein Ludwig (1958), The Blue and the Brown Book, Oxford, Blackwell. Wittgenstein Ludwig (1967), Zettel, Oxford, Blackwell.