An Ontological Justification for Contextual Authenticity

In this paper I defend a contextualist interpretation of authenticity in musical performance: we judge a performance as authentic not in respect of a stable set of requirements but according to contextually determined factors. This solution is the natural outcome of an independently supported ontological account of musical works: Musical Stage Theory. The aim of the paper is to give new momentum to the debate concerning the notion of authenticity and to challenge a monistic interpretation of authenticity: there is not one authenticity but many .


Introduction
We all seem to know what we mean when we describe a musical performance as 'authentic'. Yet, when comparing our intuitions, we often find out that each one of us informs the concept of authenticity with a different meaning. While I may deem a performance authentic because it respects the score, you may disagree and claim that a performance is authentic when it expresses the composer's intentions. We could go on: a performance may be authentic because it represents the performer's idea of the piece, because it is played on period instruments, because it takes place in the original venue for which it was conceived, and so on.
Without doubt the notion of authenticity plays a relevant role in guiding performers in their musical choices and listeners in their evaluation of performances. However, informing the concept of authenticity with conflicting meanings risks rendering authenticity an empty word that loses its regulative value.
In the literature there has been an effort to find a single way to characterize authenticity in music and to determine its regulative role. 1 I suggest that we should abandon this quest and acknowledge that there is no single way to define authenticity. In proposing a contextual assessment of the criteria behind the judgement of a performance as authentic, my proposal follows the path opened by Peter Kivy. 2 Yet, I justify an otherwise arbitrary contextualist claim on the basis of a novel and independently supported ontological account of musical works: Musical Stage Theory (henceforth MST). Elsewhere, I have argued for the benefits of adopting MST as an account for the ontology of musical works. 3 Here I am interested in showing how it can help justifying the contextualist view of authenticity.
In Section 2, I outline briefly the historical context behind the adoption of the notion of authenticity. I then show how the abundance of criteria that may be recognized as determinative of the notion of authenticity results in a multiplication of the ways in which it is interpreted. In Section 3, I explain how, strictly speaking, the ontological framework of MST justifies the understanding of authenticity only as Personal-Authenticity-in other words, the expression of the performer's ideas, independent of historical factors. The contextual interpretation of authenticity is developed in Section 4 on the basis of the more encompassing notion of work-as-construct. The various kinds of authenticity are examined and examples from actual performance practices are provided. 4 The final aim of the paper is not only to give new momentum to the debate regarding authenticity but also to exploit a new view in musical ontology-MST-in order to justify a context-dependent interpretation of authenticity.

Which Authenticity?
The criteria that informs the audience's expectations towards an authentic performance began to exercise their strength when the concept of authenticity itself gained relevance in the late nineteenth century. The interest in a faithful recreation of Early Music, promoted by the Early Music Movement (henceforth EMM), has been developing up to the present day, together with a more specific interest in the performing practices needed in order to deliver an 'authentic' performance. 5 The ideal of authenticity promoted by the EMM can be summarized as follows: The key to defining authenticity is specifying the sort of thing to which a historically authentic performance is faithful. Three main proposals have been presented. According to the first proposal, an authentic performance is one faithful to the sound of performances at the time of composition. … Alternatively, an authentic performance of a work is one faithful to the intentions of the work's composer (or author). The third proposal suggests that the authentic performance of a work involves fidelity to a score and the performance practices employed at the time of the work's composition. 6 This definition highlights the nature of authenticity as a relational concept (in other words, a performance is authentic in respect to something, whether the score, the composer's intentions, or other parameters). 7 The conception of authenticity as a relation between the performance and a pre-determined set of rules spread among the audience as the EMM gained more and more supporters, thus becoming the standard way of interpreting authenticity. This ideal of authentic performance, though, makes it difficult to accept a more pluralistic understanding of authenticity and, in particular, an understanding of authenticity where the performer plays a relevant role. Under these conditions, in fact, the performer's role is that of following obsequiously what the editions of the score or the treatises prescribe. If an ideal of purity and authenticity should be met, then, according to the dictates of the EMM, this should happen at the cost of the performers' freedom.
Philosophical literature is divided over the question of which one of the criteria of authenticity just mentioned is more important than the others. 8 Even though this lack of consensus among theorists of authenticity does not question the usefulness of authenticity as a regulative concept, it may question the strength of the concept itself and our capacity to find one way to interpret it. Kivy notoriously solves the conflict precisely by refusing to stand for one single authenticity. Instead, he suggests four different ways to define the concept: 'Authenticity-as-Sound', 'Authenticity-as-Intention', 'Authenticity-as-Practice', and 'The Other Authenticity'. 9 For the sake of clarity, in the rest of the paper I refer to the latter kind of authenticity as Personal-Authenticity.
The first authenticity, Authenticity-as-Intention, refers to an understanding of authenticity as 'faithfulness to the composer's intentions'. Authenticity-as-Sound refers instead to a level of respect for the composer's instructions-these instructions usually are provided by a score. The third meaning of authenticity, Authenticity-as-Practice, refers to the respect not only for historical sounds but also for the actual practices which were put into place in order to realize them. Finally, Personal-Authenticity has to do with the individual interpretation that the performer can give of the work. 10 The role of performers thus seems to acquire a relevance that the other kinds of authenticity do not acknowledge and, in fact, seem to diminish. 11 I agree with Kivy's contextualist interpretation of authenticity. The conventions that contribute to structuring the concept of authenticity may be different and, as a result, the concept of authenticity itself may be interpreted in a different ways. The expectations concerning the authenticity of a performance may vary depending, for example, on the 7 This thought was also expressed by a participant in one of the interviews I conducted (see n.4 for details): 'Authenticity is a two-place predicate. It is authenticity to something and you can fill the blank in different ways.' style to which it belongs: we would not apply the same criteria for evaluating the authenticity of a Jazz solo and of an early Classical Sonata. Or they could vary depending on the context of the performance. We may judge a performance strictly because we read on the programme that it is supposed to be an authentic performance of Handel's Messiah.
Probably we would not be so strict if we were to judge a performance of the same piece played by a high school's orchestra and choir.
In the next sections, I strengthen Kivy's contextualist interpretation. Indeed, I justify the legitimacy of the existence of more than one authenticity by including the concept within the bigger picture of a novel account of the ontology of musical works: MST.

Authenticity and Ontology
In philosophical literature, authors have tried to make sense of the notion of authenticity by grounding it in ontological considerations. For example, Levinson's contextualist account of musical works as indicated sound structures which are partly constituted by their means of performance allows him to say that: Performances are partly authentic in virtue of being performed on the instruments for which they were intended (or envisaged), for a reason wholly other than, and distinct from, their thus procuring a sound that matches what an ideal contemporary performance would have delivered. 12 For Stephen Davies, instead, authenticity is faithfulness to the intentions expressed in a score by a composer. The variable thickness of a musical work, reflected by the level of detail in the score's instructions, determines the norms to follow for an authentic performance. 13 In what follows, I analyze the notion of authenticity through the ontological account put forward by MST. The result is unexpected: the strict interpretation of the nature of musical works as performances, independent of any historical considerations, cannot accommodate the evaluative notion of authenticity that we are used to employing in our language. Thus, authenticity gains independence from ontological considerations. 14 Although this does not mean that MST cannot accommodate an evaluative understanding of authenticity. The shift operated by MST from a strict reading of musical works as performances to a more encompassing notion of 'work' will, in fact, be enough to explain why we often use the concept of authenticity in our musical discourse and why we can make sense of it as a context-dependent notion. In the next section I provide a quick outline of the main arguments proposed by MST. This will be crucial for later discussing how we can justify a contextual interpretation of musical authenticity.

Musical Stage Theory: An Overview
MST argues that the work we refer to with the expression 'Mahler's Symphony No. 4 in G major' is a sound event, namely a performance (from now on referred to as workas-performance). 15 As a consequence, strictly speaking, every performance is a different work. Yet, MST grants that the act of grouping performances together according to a certain relationship also plays a role in our talk about music.
The presentation of MST and, consequently, the explanation of how it overcomes certain apparent difficulties that ensue from identifying works with performances, takes as its starting point an analogy with theories expressed by Theodore Sider about persistence in space-time. 16 According to MST, musical works are spatiotemporal stages, just as continuants are in Sider's view. Different arguments have been given to support MST: it suffices here to mention its straightforward consistency with everyday intuition that means we immediately grasp a musical work by listening to it. 17 Every work-as-performance is related to its counterparts through the Repeatability-Relation (henceforth R-Relation)-in other words, the sort of relation reflecting the set of intuitions that traditional theories (I am thinking of Platonism and Nominalism especially) explain in terms of exemplifiables and their instances. So, although the work (say, 'Mahler's Symphony No. 4 in G major') is a performance-stage, this entity is related to other performances-stages by means of the R-Relation, parallel to Sider's Identity-Relation.
The components of the R-Relation are (i) a Causality-Relation between works-asperformances and which connects works-as-performances to the relevant act of composition, (ii) an Intentionality-Relation which expresses the performers' intention to play exactly that performance, and (iii) a Similarity-Relation which calls for a sufficient degree of similarity between the works-as-performances. 18 As stated, for MST our terms for works primarily refer to stages-in other words, performances. Yet, MST needs to explain our uses of terms such as 'Mahler's Symphony No. 4 in G major' in ways that do not seem to refer to a single work-as-performance but 18 I discuss the contextual assessment of these requirements in Section 4. Downloaded from https://academic.oup.com/bjaesthetics/advance-article-abstract/doi/10.1093/aesthj/ayz020/5540151 by University of Nottingham user on 31 July 2019 to a more encompassing understanding of 'work'. In order to resolve this, with a strategy similar to Sider's move from stages to works, MST explains how our linguistic attitudes systematically shift between the reference to single sound events, works-as-performances, to a more general concept of work: the work-as-construct.
Leaving aside further independent details, what is important is that, not unlike Sider's approach to persons, the shift of reference between work-as-performance and work-asconstruct is importantly contextual. Deciding whether 'Mahler's Symphony No. 4 in G major' refers to a single stage, to a collection of stages, or to other information regarding the work, depends on the speaker's linguistic intentions within the context of the expression at stake. For example, if I say 'Mahler's Symphony No. 4 has been performed many times', I am referring to a collection of R-related stages. If, instead, I say 'I enjoyed Mahler's Symphony No. 4' I am referring to a performance-stage. 19

Personal Authenticity
For MST, a musical work is a performance. In this picture, the performance is not judged with respect to previous renditions of the same score, to the composer's instructions or intentions, to the performing traditions, or to the use of period instruments. The majority of views about authenticity mentioned in the previous sections, then, are excluded from the consideration of MST, strictly speaking. The performance stands by itself and we do not need to judge it as authentic in respect to sound, intentions, or practice, since in order to be deemed a work it does not need to be included in a web of historical or contextual relations. 20 There is, however, one kind of authenticity which may still apply within the strict layout of MST: Personal-Authenticity. Personal-Authenticity is the authenticity of the performance in respect of the individual interpretation given by the performer. This idea of authenticity, as I mentioned, is different from the traditional idea of authenticity as the respect for parameters such as compliance with the score or with the composer's intentions. It is, however, nearer the concept of authenticity which is adopted in our common discourse-namely authenticity as genuineness, as a transparent display of intentions and meaning. The emphasis that the interpretation of MST gives to authenticity as Personal-Authenticity shifts the focus of the criteria for determining authenticity from the original act of composition to the personal contribution made by the performer.
I will soon return to the questions of how MST can accommodate the notion of Personal-Authenticity and how it can still make sense of a contextualist interpretation of authenticity. But, first, for the discussion that will follow, it is important to note the differences between the notion of Personal-Authenticity and what Julian Dodd calls 'interpretive authenticity'. Dodd argues in favour of the relevance of interpretive authenticity, defined as 'a way in which e can be faithful to W in performance that consists in e's displaying understanding of W: that is, in e's interpreting W in a perceptive or insightful way'. The notion thus indicates the synthesis of the thought processes that underpin a performer's decision-making process-for example, knowledge of styles, knowledge of ways to realize notations, previous experiences, etc. Dodd contrasts this value of authenticity with score-compliance authenticity which, as the name suggests, presents authenticity as the accurate rendition of a score into sounds. Dodd claims that the insights provided by interpretive authenticity on works are to be preferred to a rigid respect for the instructions provided on the score. Deviations from the score are thus allowed if they make it possible for the performance to provide a better understanding of the work. 22 In this paper, Dodd only hints at the divergence between his interpretive authenticity and Kivy's Personal-Authenticity. In a recent talk he goes into more depth into this discrepancy. 23 Kivy describes Personal-Authenticity as being true to the performer's artistic values in the performance. 24 Dodd qualifies interpretive authenticity, and the understanding of the work that it provides, as a performance value. On the other hand, he does not consider Personal-Authenticity a performance value. He motivates his claim by showing how Kivy's argument is undermotivated: the individual style and originality transmitted to a performance by a specific interpretation can be achieved not only through Personal-Authenticity, as Kivy claims, but also through what Dodd has presented as interpretive authenticity. Moreover, blindly following her own personal style in performances of different repertoires may lead a performer to provide uninteresting and self-centred performances. Thus, Dodd's final claim is that seeking Personal-Authenticity, understood merely as following one's own style, cannot be deemed a performance value. Interpretive authenticity, instead, with its richer connotation, may trump other authenticities and take the lead for delivering a more profound understanding of the work. 25 As I noted at the start of the section, the kind of authenticity that MST, in its strictest ontological reading, can accommodate is Kivy's Personal-Authenticity-a kind of authenticity, that is, devoid of any relation to pre-determined instructions, intentions, or practices, but which reflects only the performer's personality. As Dodd rightly points out, we might struggle to accept this interpretation of authenticity as a performance value, though. In our musical discourse, in fact, we make use of the attribute 'authentic' to show how a performance relates to something else-in other words, a score, a set of instructions, a performing practice, and so on. The notion of authenticity that we use is a normative concept and, if we interpret it just as Personal-Authenticity, potentially any performance might be authentic. 26 26 Personal-Authenticity defined as the respect for the performer's artistic personality can still be a performance value in the case of free improvisation. When improvising freely (even if following some rules) the performer is indeed expressing her own personality without the need to respect any scores, performing practices, or composer's intentions.
Downloaded from https://academic.oup.com/bjaesthetics/advance-article-abstract/doi/10.1093/aesthj/ayz020/5540151 by University of Nottingham user on 31 July 2019 Nevertheless, the fact that the ontology of MST, at first sight, cannot account for a normative notion of authenticity is not a major drawback. First, as stated above, this helps to free the evaluative notion of authenticity from ontological considerations. Second, the fact that we cannot ground authenticity on ontology can also explain why we struggle to find one kind of authenticity that applies to all performances. There is in fact no independent ground that might justify such a choice. Lastly, MST can still explain the normative role played by authenticity by resorting to the shift from the work-as-performance to the work-as-construct. The more generous understanding of musical works allowed by the work-as-construct can in fact explain why there are many authenticities, and Dodd's interpretive authenticity is among them, all equally persuasive.

Authenticity and Context
I opened this paper arguing that MST could justify a contextual assessment of authenticity. If this is not possible by considering the notion of work-as-performance, it is, however, possible through the shift to the work-as-construct. While at the level of workas-performance, as mentioned, the only valid authenticity is Personal-Authenticity, at the level of the work-as-construct, other values of authenticity may be applied. Depending on the context or the audience's or performers' interest, in fact, authenticity as respect for the score, for the performing tradition, for the composer's intentions, or for a more individual understanding of the work-all information which add up to the work-asconstruct-can be possible.
The contextual assessment of the notion of authenticity can be justified by MST through the R-Relation. In outlining the main points of MST, I referred to the R-Relation being a horizontal relation between performances which, thanks to its constitutive features, explains the rules that we need to respect when comparing performances and when referring to a performance as a performance of a specific composition. In this sense, the R-Relation plays a regulative role and, I suggest, this role is comparable to that played by different notions of authenticity.
To demonstrate the similarity between the R-Relation and authenticity, it is sufficient to compare the features of the three requirements posited by the R-Relation with the kinds of authenticity proposed by Kivy and Dodd. The (i) Causality-Relation has the same normative features as Authenticity-as-Sound, Authenticity-as-Intention, and Authenticity-as-Practice; the (ii) Intentionality-Relation has demands equivalent to interpretive authenticity; and the (iii) Similarity-Relation and Authenticity-as-Sound both require a similarity between the sonic profiles of the performances. 27 The R-Relation between performances may intelligibly depend upon contextual factors, such as the interests of the speakers or the musical background that she takes to be appropriate. The Causality-Relation may then be deemed more relevant than the others in a performance of Early Music. The Similarity-Relation may instead outplay the others if we are looking for discrepancies between performances of the same composition: for example if we want to compare Barenboim's and Pollini's renditions of Chopin's Nocturne n. 6 in G minor. It is this contextuality, intrinsic to the R-relation, that plays a relevant role for the contextual assessment of authenticity.
The choice of one kind of authenticity over the others is dictated by social norms, in the same way as the relevance assigned to one component of the R-Relation over the others is contingent on conventions. 28 Social norms play a fundamental role in shaping our perception of musical works and their authenticity. 29 The social norms associated with the EMM, together with the nationalistic ideologies of the nineteenth and twentieth centuries, led to an emergence of an ideal of authenticity as the respect for the score and for the composer's intentions. The social norms related to Rap or Folk music, instead, privilege the emergence of the performer's individual personality and, as a consequence, they put forward an ideal of authenticity as expression of the self and of the performer's original contribution. The relevance of social norms for the adoption of one kind of authenticity over the others may thus explain how authenticity is perceived differently in different cultures.
In the following sections, I show how performers and conductors may value one constituent of the R-relation over the others. This, in turn, leads them to privilege one kind of authenticity. I organize the discussion according to the three requirements of the R-Relation. I provide further reasons for claiming that none of these requirements can be taken as absolute and more relevant than the others. I then question the desirability of interpreting authenticity within such tight boundaries. Each authenticity is equally relevant, and its relevance depends on the context within which it is evaluated.

The Causality-Relation
The (i) Causality-Relation draws a link among performances and between performances and the act of composition. So, according to this aspect of the R-Relation, a work-asconstruct is to be understood in terms of certain performance's appropriate relation to the original composition. This ontological relation is reflected by at least three senses of authenticity: Authenticity-as-Sound, Authenticity-as-Intention, and Authenticity-as-Practice. A link with the act of composition asks for the respect for the composer's instructions and intentions.
The first two senses focus directly on the relationship with the score indicated by the composer and her intentions in doing so. composer's ideas about how it should be performed, in fact, it is possible for a performance to be authentic in these two respects. I discuss Authenticity-as-Sound further in Section 4.3. Authenticity-as-Intention, for its part, defines authenticity as respect for the original idea of the composer about the composition. 30 Compliance with this may certainly add greater detail to the performance and, undoubtedly, it would be desirable to have the possibility of hearing, for example, the Goldberg Variations (1741) as Bach wanted them to be played. Authenticity-as-Intention is privileged, for example, by the harpsichordist Ralph Kirkpatrick who claims that his aim is to perform 'harpsichord and clavichord music in a manner as close as possible to what could be ascertained of the intentions of the composers'. 31 However, it seems that this sense of authenticity is hardly in the position of doing the job by itself: the composer's intentions may be impossible to discern, and, even when they are, they may not engender the most rewarding interpretations. 32 An example is the episode between Claude Debussy and the pianist George Copeland: Debussy asked Copeland why he played the opening of Reflets dans l'eau [(1905)] the way he did. Copeland's response was that old performer's standby, calculated to make any musicologist see red: 'Because I feel it that way'. To which Debussy replied that as for himself he felt it differently, but that Copeland must go on playing as he, Copeland, felt it. 33 The third kind of authenticity connected with the causal dimension of the R-Relation is Authenticity-as-Practice. It can involve elements which extend from the use of period instruments, through the choice of venue and the phrasing and expressiveness employed, to the style of the period in which the piece was originally performed. 34 All the mainstream performers I interviewed agreed on the fact that they bring a different approach to the score depending the historical period and musical style in which the composer wrote the score. Some examples: I must say that, almost automatically after years of study, I approach the piece in a different way on the basis of the period of the author. I would never play Mozart like Beethoven or Brahms like Stravinsky. 35  Outside the Western Classical Music tradition, the respect for the history of performing practice of the piece is a criterion adopted by Blues: 'the authenticity of a blues performance turns on the degree of mastery of the idiom … . Evidence of authenticity can be sought "in and around the performance" for the performer's recognition and acknowledgement of indebtedness to sources of inspiration and technique'. 37 Achieving a way of performing which is near to the one practised by the composer's contemporaries may be desirable. However, it has limitations as a sole criterion for authenticity. For starters, the reconstruction of period instruments is difficult to achieve given the incomplete instructions that we may have and the difficulty of accessing the materials which were once used. Even supposing that we do possess instruments identical to the original ones, the treatises and performance manuals cannot give us complete information about how to perform. 38 This is partly due to the aforementioned presence of improvisatory practices in performances, which cannot be fully reported and transmitted, to changes in temperament and pitch, and to differences in listening conditions and expectations. 39 Bach's St Matthew Passion (1727) cannot feasibly sound as grand to contemporary audiences as it must have sounded to Bach's. The change in expectations, already indicated as one of the reasons behind the multiplication of authenticities, also affects the way in which the audience perceives a performance, thus making the possibility of achieving authenticity harder.

The Intentionality-Relation
The (ii) Intentionality-Relation gives relevance to the performer's intentions to play that specific performance as an essential condition for justifying certain groupings of performances under the same 'work-as-construct'. This relation can be equated to the demands of interpretive authenticity as described by Dodd: interpretive authenticity entails a deviation from the other understandings of authenticity as respect for the score or for the composer's instructions, in favour of an interpretation which is reflecting the personal choices of the performer.
We are arguably more comfortable in applying this kind of authenticity to genres other than Classical music. For example, when talking about the search for the validation of the self through Folk music, not only by musicians, but by the entire community, Emily Green says: 'The quest for authenticity in music, then, becomes a quest for truth about oneself for which there is no objective answer.' 40 The audience plays a fundamental role in this process. Listeners value the personal contribution of the performer equally and sometimes more than the melodic or lyrical profile of songs. 41 As an additional factor, in Pop more than in Classical music the performers are deeply influenced in their performance by the kind of audience that faces them. The different setting and degree of involvement of the audience are part of the equation as well. While at Classical concerts the audience sits in silence (even if this was not what audiences centuries ago did during concerts), during Pop, Rock, or (some) Jazz concerts the audience participates, expressing appreciation or disappointment for the performance.
In the case of Rap music, the performer's public persona, and the expression of her Self through music, inexorably affect the audience's perception. 42 On the other hand, the feedback from the audience inevitably affects the performance and is, therefore, an essential part of the personal expression of the performer and of the authenticity of the latter.
Privileging interpretive authenticity over other kinds of authenticity is also not uncommon in musicians of the Western Classical tradition. This seems the conception invoked in the following example. Reviewing Casal's Brandenburg recordings, Lionel Salter affirms: His conception of Bach is frankly one which does not find much favour today: turning his back on scholarship and totally unconcerned with all problems of style and textural interpretation he not only unabashedly uses modern instruments [...] but pursues an aim of playing Bach as expressively 'as if it were Chopin'. 43 A symptomatic declaration of this conception of authenticity among musicians can also be found in the Early Music scholar Nikolaus Harnoncour. He is against the idea of historical performances in which the subjectivity of the performer is obscured in favour of an alleged authenticity: 'The outcome is found in those familiar musical performances which are often historically impeccable, but which lack all vitality. Clearly, an interpretation that was historically uninformed but musically alive would be preferable.' 44 What emerges from the thoughts expressed by these performers and scholars-and also by many others that I do not have the space to discuss here-is that, in line with Dodd's view on interpretive authenticity, the performer's interpretation of a piece may have more weight than other, more widely acknowledged, criteria for the evaluation of the authenticity of a performance.

The Similarity-Relation
The last relation, (iii) similarity, asks for the R-related performances to be similar enough in their sonic profile. The same result is achieved by performers who want to strive for Authenticity-as-Sound. The first step for two performances 'of' Mahler's Symphony No. 4 in G major, in fact, is for the performers to follow the instructions indicated on the score. The dichotomy between authenticity as complete respect for the score and authenticity as partially informed by the performer's contribution was to accompany the EMM throughout its history. Some conductors, such as Stokowski and Furtwängler, rebutted the claim that authenticity meant unconditional respect for the score. They dismissed the most extreme claims of historically informed performing practice. Others, like Toscanini and Leonhardt, did however prioritize respect for the instructions provided on the score and tried to achieve that as an aim. 45 The slavish respect for the score may however lead to preposterous results. Peter Hill reports that, in a master-class he attended 'a respected singer had no hesitation in asserting that because Schubert dropped the staccato signs in the third bar of a repeating accompaniment figure, the piano part should thereafter be legato'. 46 The discussion about the extent to which the respect for the score should be pursued is especially interesting when applied to scores of contemporary music, which range from the scarcity of instructions provided by chance music, to the extreme precision and level of detail of Ferneyhough's scores. Composers' attitudes differ in regard to the effort performers should put into following the instructions they provide on the score: some, such as David Hockney, give free hand to performers, others, like Luigi Dallapiccola, supervize performers until the day of the concert. 47 Once again, the choice of respect for the score as a privileged criterion of authenticity may depend on social norms, personal preferences, or other contextual factors. Whether adopting it or not, however, may be controversial in itself.
Score-compliance is the interpretation of authenticity traditionally adopted in Western Classical music. 48 Scores are underdetermined, and they were even more so three or four centuries ago when improvisatory practices made an essential contribution to the final shape of the piece. Interpreting the faithfulness to the work as faithfulness to the notation is thus inaccurate. 49 To this must be added the historical transformation of the aspect of scores. Elements such as tempi and phrasing were not indicated on scores before the eighteenth century. 50 Nevertheless, performers must have been compelled to make performative choices regardless of the lack of markings in the score. I conclude this section by pointing out a potential hurdle. The similarity criterion of the R-relation is essentially vague. The vagueness in determining the degree of similarity between two performances is partly due to the underdetermination of the score, from which they both derive. Different performers, in fact, could give equally correct renditions of a score despite changing some non-essential details of it, such as a change in the tempi, nuances, timbre, and so on. The similarity between two renditions of the same score admits of degrees and can be assessed contextually: it may oscillate between totally acoustically indistinguishable renditions and total freedom of both parts in interpreting the instructions provided. Unless one adopts an absolutist view in this respect, it is difficult to place the boundaries around what can be deemed acceptable and what cannot. 51 The contextual assessment of the Similarity-Relation is however inherently different from the contextual assessment of the R-Relation as a whole and, consequently, of the notion of authenticity. 52 While the context-dependency of the notion of similarity derives from the vagueness of the concept itself, the contextual assessment of the R-Relation is a fundamental element of the ontology of MST which establishes its alleged superiority in respect to rival theories. The type-token account, for example, is metaphysically rigid and the relationship between works and performances is fixed once for all. There can be a certain leeway also in the type-token framework, but it ultimately leads to the dichotomy between correct and incorrect performances. The contextualism advocated by MST, instead, extends beyond the evaluation of performances as correct and incorrect, thus granting a greater freedom to performers than the one they could ever get from typetoken theorists. 53 It is this kind of contextuality, part and parcel of the ontology of MST, that justifies a contextual assessment of the notion of authenticity.

Conclusion
The normative aspects which we recognize as relevant to the notion of authenticity and which help us to orient our judgement of performances we listen to are therefore the aspects also shared by the R-Relation through its three components. MST is thus able to account for the regulative aspect of authenticity and for our everyday linguistic exchanges.
In addition, MST is able to provide a re-interpretation of authenticity which offers benefits to the overall explanation of the concept. The three main benefits brought by this reconsideration are: (i) to give new momentum to the discussion about authenticity, (ii) to justify the existence of multiple authenticities on the basis of the contextual assessment of the R-Relation, and (iii) to grant autonomy to the notion of authenticity from ontological considerations.
The most obvious benefit brought about by the reconsideration of authenticity through MST is that of advancing the debate over the topic of authenticity. The analysis of authenticity by MST, revisionary as it may seem, challenges the traditional role that authenticity has played, calling into question its application in musical discourse. The notion of authenticity, through the many discussions and arguments surrounding it, is now encrusted with meanings that are not necessarily the ones that reflect its actual role in musical practice. The conventions, and the subsequent audience expectations engendered by the notion within the EMM, fixed the meaning of authenticity as that of respect for the composer's intentions and instructions. But, as the review of Kivy's proposal has demonstrated, authenticity as it is actually interpreted is much more than this. The interpretation of the multifaceted nature of the notion of authenticity through the contextual assessment of the R-Relation given by MST has the benefit of justifying the conclusion reached by Kivy. 51 As Goodman does in Languages of Art.
52 I thank an anonymous referee for pointing out this potential worry.
53 For a more detailed discussion on this, see Moruzzi, 'Every Performance is a Stage', 348.
Downloaded from https://academic.oup.com/bjaesthetics/advance-article-abstract/doi/10.1093/aesthj/ayz020/5540151 by University of Nottingham user on 31 July 2019 This leads directly to the second, and I believe most relevant, benefit of this analysis of authenticity: the legitimization of multiple authenticities through the contextualism of the R-Relation. As our linguistic and evaluative attitudes show, there is no single value of authenticity that should take a lead over the others. If we expect to listen to an Early Music performance, we may choose to value authenticity as respect for the instructions on the score. If, instead, I deem that Bach would have preferred his works to be played on a more versatile instrument, then I would rather listen to the Goldberg Variations played on the piano. Similarly, if I am listening to the n th rendition of Beethoven's 'Hammerklavier' Sonata, then I may wish to listen to a performance where Personal-Authenticity comes to the fore. If, instead, I am listening to it for the first time, then I may prefer a plain, more straightforward performance. The adoption of a multiplicity of ways to judge a performance as authentic is a feature of our musical discourse which finds an explanation in the influence that contextual factors have for understanding the nature of musical works. Just as we may relate performances together contextually, according to their common history of composition, intentionality of performers, or similarity of sonic profile, we may judge the authenticity of performances according to contextually determined factors.
The last, arguable, benefit of the revisionary interpretation of authenticity is the autonomy that the notion of authenticity gains from any ontological considerations. MST can still pursue the aim of justifying a contextualist interpretation of authenticity within its framework, but it does so by resorting to the more inclusive notion of work-as-construct and not through its straightforward understanding of work as work-as-performance. Thus, the evaluative nature of the notion of authenticity is preserved by explaining it through the consideration of the historical, personal, and contextual factors which surround the concept of musical work. In addition, this result brings with it the benefit of explaining the difficulty of applying one single, overarching concept of authenticity: there is no independent ontological basis that would allow us to do so.
A great part of the debate over authenticity started because of a lack of consensus over which aspects of authenticity should be deemed more relevant and worthy than others. Should respect for the composer's intentions be more important than respect for the performing tradition? Or should Personal-Authenticity be more relevant, since it acknowledges the role of the performer? These and similar questions, which constitute an impasse for finding a definite explanation of what authenticity is, lose their sting within MST. Each of these authenticities, in fact, bears the same relevance for the latter. The application and prevalence of one or the other is merely dictated by the context in which the judgement over authenticity is carried out. MST thus accommodates the practicality of authenticity in regard to the work-as-construct and shows how trying to reach a consensus over one single authenticity is hopeless: there is not one authenticity but many. 54 Caterina Moruzzi University of Nottingham caterina.moruzzi1@nottingham.ac.uk 54 I thank two anonymous referees and the editors of the journal for their insightful comments on the first version of this paper. This work was supported by the Arts and Humanities Research Council grant 1504272.