Art's historical nature

Abstract In this paper, I examine some of the arguments that have been put forward against Levinson’s historical account of art. These arguments focus upon the conceptual impossibility of defending a historical account of art when we take into consideration issues regarding the identification of art in other cultures or worlds. Levinson has tried to make his position clearer defending himself from these criticisms in a way that seems quite persuasive. However, as I argue at the end of this article, Levinson’s attempt cannot successfully grasp a meaning of art that could be identified universally. At the end, his account can only works if we restrict our application of «art» to works that already belong to our own history of art and that, correspondingly, satisfy one or more of the artistic regards that have been relevant in our culture. Key words: art’s definition, artistic regards, art’s nature, historical definition, other cultures, possible worlds, intension and extension of «art». Resumen. La naturaleza histórica del arte En este artículo, examino algunos de los argumentos ofrecidos en contra de la definición histórica del arte defendida por Levinson. Estos argumentos subrayan la imposibilidad de defender una teoría de este tipo una vez que tomamos en consideración cuestiones relacionadas con la identificación del arte en otras culturas o mundos posibles. Levinson ha tratado de aclarar su posición a la vez que ha proporcionado algunas respuestas a estas críticas que parecen convincentes. Sin embargo, como trato de argumentar al final de este artículo, Levinson no puede dar cuenta de un significado universal del concepto de «arte». Finalmente, su teoría sólo funciona correctamente cuando la aplicamos a obras que ya pertenecen a nuestra historia del arte y que, consecuentemente, satisfacen algunas de las consideraciones artísticas que históricamente han constituido el significado de «arte» en nuestra cultura. Palabras clave: definición del arte, consideraciones artísticas, naturaleza del arte, definición histórica, otras culturas, mundos posibles, intensión y extensión del término «arte». Enrahonar 38/39, 2007 159-169 Art’s historical nature1 María José Alcaraz León Universidad de Murcia (Spain) [email protected] 1. This paper has been possible thanks to the finantial support of the research projects La expresión de la subjetividad en las artes (HUM2005-02533) and La expresión de la subjetividad: teoría y práctica (03089/PHCS/05). Summary 1. Art’s historical nature: an introduction One of the ideas that, in some form or another, has entered into contemporary reflection about art nature is that art is essentially historical. This claim can be understood in many different ways ranging from the trivial sense in which we claim many other practices to have a history, to a stronger sense according to which what art is at a certain moment is historically determined. One of the authors2who has currently defended this claim is Jerrold Levinson who has attempted to provide a definition of art which makes intrinsic reference to art history. Actually, his definition takes art historicality as its core. The idea that art has a historical nature has a twofold root. It is partly due to certain claims held by art historians such as Wöllflin whose dictum «not everything can become an artwork at any time»3has become a well-known commonplace within art theory; the second one is derived from the philosophical response to a certain, important, amount of art production which defied common assumptions about art before the avant-garde movements. Among its merits, avant-garde art made possible that a plain object could be regarded as art thanks in part to the creation of Duchamp’s ready made; this fact strained common assumptions about art and showed certain theoretical reflection about the concept of art as necessary in order to bring these new works into art realm. Actually, both Arthur Danto and Levinson have recognized avant-garde artistic production as a prompter of the idea we are trying to pursue here4. 1. Art’s historical nature: an introduction 2. Jerrold Levinson’s intentional-historical definition of art 3. Gregory Currie’s criticism to art as an historical concept 4. Stephen Davies’ criticism against artworld relativity 5. Does Levinson’s historical definition provide a sufficient condition for art identification across different worlds or different traditions? 6. One more possible world example Bibliography 160 Enrahonar 38/39, 2007 María José Alcaraz León 2. Others definitional attempts which take into consideration historical concerns may be attributed to Arthur Danto (1964), Noël Carroll (1993), and James Carney (1994). 3. This claim, first formulated by Enrich Wölfflin, has been appropriated by many contemporary art theorists. 4. Danto has defended the idea that the real problem a philosophical characterization of art must solve could not be recognized as such before some works of art set the conditions for it to appear as such problem. Levinson, following Danto in this point, claims that «it is curious […] that the highly abstract, self-referential condition I propose as the minimal essence of art-making is one that could not have been evident to us before the deconstruction, if we may call it that, performed on the practice of art by the avant-gardists of the early twentieth century» («Extending Art Historically». In: LEVINSON, J. (1996). The Pleasures of Aesthetics. Ithaca: Cornell University Press, p. 150-171, p. 152). These two traditions converged at one point: as well as certain material conditions must be given for a certain object to be produced, certain theoretical atmosphere is required for something to be regarded as art. Certain objects already considered among the paradigms of art, like the ready-mades by Duchamp, could not have been regarded as art if they had been done, say, three centuries earlier. The kind of artistic atmosphere which allowed a ready made to be art had a totally different structure from the one that made possible Renaissance painting or, to press back into the past, the cave paintings of Lascaux (if these latter can be considered art at all). Quoting Levinson at this point: «what was understood by the term “art” in 1795 is not the same thing as is understood by the term today […]; what items or activities would have counted, the reasons why they would have so counted, and what would have been the paradigms with reference to which counting would have been assessed were dramatically different»5. Someone may argue at this point that most of our practices can parallel art in this respect; specially, science and philosophy have a history in which the only relevant aspects to take into account are not merely the different theories proposed within each domain, but also reflections about what counts as science and philosophy seem to be relevant to trace their respective histories. However, even if those considerations might have a key role in the task of reconstructing these practices’ history, it still seems true that we can define them independently of the historical considerations about their nature; they can be defined, for example, functionally6. However, for those who take the historicality of art seriously, no functional definition grasps everything that counts as art because, as its history has widely shown, the range of concerns that have been involved in art practice is so diverse that no functional theory is able to account for it. So, an historical definition not only has the virtue of letting these various concerns to enter into the meaning of art, but also leaves open the possibility for new concerns to arise as relevant for art production7. It must be noticed that the true historical account of art does not only rest upon this assumption, but it underlies the contingent character of the concerns that have historically provided the content of art. These make up a set that helps present art identification, but this set could have contained different concerns if art history had been different. Historical account of art must then take into account the enrichment of art meaning through its history and its contingent character. So, in principle, Levinson would defend the uniqueness of the concept of art in this sense. Most of our practices are historical in a weak sense, but only Art’s historical nature Enrahonar 38/39, 2007 161 5. LEVINSON, Jerrold (1996), «Extending Art Historically». In: The Pleasures of Aesthetics. Ithaca: Cornell University Press, p. 150-171, p. 150. 6. «Philosophy and science, however varied their manifestations, retain some intrinsic or functional differentia, which is not the case with art» (Levinson, ibid., p. 155). 7. In this sense, Levinson’s definition would be the best proof against the non-definition stance of most neo-wittgensteinians who, taken as a fact the wide diversity of artworks and artregards, claim that art cannot be defined without violating its essential disposition to include new art-regards. art is historical in the much stronger sense which implies that art’s meaning cannot be specified without making explicit reference to its history. And what the art theorist must do is to find the shape of our concept of art so that it fits this historical condition. I will first examine Levinson’s definition paying special attention to its historical aspect. Second, although he has proved an impetuous energy in providing answers to the many and diverse criticism against his view, I would like to focus upon Gregory Currie’s criticism formulated in his article «A note on Art and Historical Concepts» and Stephen Davies’ concerns against what he names the «artworld relativity» problem. I think both authors have provided some arguments against Levinson’s definition, and have supported the idea that a certain concept of art not dependent upon a certain history or tradition seems necessary to identify art across different traditions. Besides, these authors’ arguments can be taken as contributing to one of the positions of a wider debate concerning art’s nature. This debate, which is far from ended, is settled between those who think that art’s nature can be fixed through a priori reflection about art, and those who embrace the view that art’s nature is essentially historical, so that what is art at a moment t depends upon certain relationship between the object and the art history within which it is created. This second view makes art’s nature relative to what historically have been considered as art and brings art history within the very content of the concept of art. We do not need, for our present purposes, to settle the endpoint to this debate. I merely hope to show that there is a certain intuition about art identification which is kept in these criticisms against Levinson that should be preserved and that will render Levinson’s view insufficient. 2. Jerrold Levinson’s intentional-historical definition of art Jerrold Levinson has provided an intentional-historical art definition which makes both artists’ intentions and historical issues essential to art. I will pay attention to the historical feature of his definition, that is, to what he terms his «intrinsic historicist» claim about art. For Levinson, art is a special kind of concept whose content cannot be given independently from what has been historically taken as art. Thus, his definition has a structure which provides «the meaning of “art at t” […] in terms of the extension of “art prior to t”»8 that is, were the extension of art prior to tdifferent from the one it actually is, the meaning of art would be different. Levinson’s definition has a recursive structure which aims at providing the necessary and sufficient conditions for something to be an artwork at a certain time. His complete definition is the following: 162 Enrahonar 38/39, 2007 María José Alcaraz León 8. LEVINSON, J. (2004). «Defining art Historically». In: Peter LAMARQUE and Stein HAUGOM OLSEN (eds.). Aesthetics and the Philosophy of Art. The Analytic Tradition. An Anthology. Oxford: Blackwell Publishing, p. 35-46, p. 40. X is an artwork at t = df X is an object of which it is true at tthat some person or persons, having the appropriate proprietary right over X, nonpassingly intends (or intended) X for regard-as-a-work-of-art —i.e., regard in any way (or ways) in which objects in the extension of “artwork” prior to tare o were correctly (or standardly) regarded9. Thus, he introduces past art-relevant regards and past artworks into the conditions that take part of the intention of an artist to make an artwork; that is, he brings art-historical considerations into the conditions for something to become an artwork. Note, however, that Levinson’s definition is meant to avoid the charge of circularity for the meaning of «artwork» is not included in the right hand of the definition but merely its past extension to some point. Art is characterized in such a way that its content cannot be specified regardless its actual art history. As Levinson’s puts it: «that art is necessarily backward-looking (though in some cases not consciously so) is a fact that the definition of art must recognize»10. We can see that his definition provides a function which, once properly filled, will provide the meaning of art at t. It does not refer to any specific property of the object which aspires to arthood, neither it appeals to any function a work is supposed to satisfy. It aims, nevertheless, as providing an essential feature of art’s nature: the historical enrichment of its essence. As with other definitions, we expect from Levinson’s proposal that it provides necessary and sufficient conditions for a work to be (regarded as) art, and, in principle, we also expect it adjust to our intuitions about works that aspire to be included within art realm but that fail to do so, that is, we expect certain normative character from an art definition. Once we have seen how Levinson’s incorporates into his definition this backward-looking of art, we can introduce some criticisms against it. As I hope to show, these share a common worry concerning the lack of specificity of the notion of regard-as-a-work-of-art introduced by Levinson. Finally, I will offer one more reason against Levinson’s definition sufficiency for art identification. 3. Gregory Currie’s criticism to art as an historical concept Gregory Currie11 has argued that Levinson’s definition cannot properly characterize art as an historical concept in the way he intends because historical concepts must have a structure which, in fact, overrides the very aim of Levinson’s attempt. Following Currie’s argument, in order to make intelligible the idea that our concept of art is historically determined, we must admit that our concept would have been different had its history been different. If we are to make Art’s historical nature Enrahonar 38/39, 2007 163 9. LEVINSON, ibid., p. 40. 10. LEVINSON, ibid., p. 35. 11. CURRIE, G. (2000), «A Note on Art and Historical Concepts». British Journal of Aesthetics, vol. 40, nº 1, January, p. 186-190. sense of a different concept of art which, by definition, is not ours, we must have a way to identify it as an art concept all the same; this implies, in his view, we need to make reference to an overarching concept of art that binds our concept and other possible concepts of art together. Otherwise, how can we identify such concepts as concepts of art? Thus, he claims that for our concept of art being really historical it should have something like the following structure. (i) We need a concept of art that it is our actual concept. (ii) Where its being our actual concept depends on some contingent matter of history. (iii) We need a concept, or some concepts, that we might have had, had things been different with history of art. (iv) And we need some overarching concept that binds them altogether —that shows that they are all, indeed, concepts of art12. The necessity of (iv) is shown after applying to the concept of art the bidimensional structure for a semantic of possible worlds that, in Currie’s view, can help us elucidate the structure that historical concepts must have. He claims that given that a historical concept must have these elements, its structure is the same as natural concepts’ identification across possible worlds; moreover, this analogy, when properly developed, shows that in order to make sense of the identification of art across different worlds we need to make reference to an already possessed concept of art which serves to bind different artistic practices. This proves, in his view, that the very attempt at characterizing a concept as historic requires that we already have a certain characterization of that very concept. Thus, if Currie’s argument is right the very conditions that make intelligible the concept of art as an historical concept undermine its essential historicality, for it is part of the structure of such concepts that we already possess a concept of art which, somehow, relates the different art concepts identified across different worlds. Although Currie’s analogy between «art» and «water» —which is the term used in his analogy between natural kinds’ identification across possible worlds and historical concepts— may be questioned13, we can retain by the moment the idea that certain reference to an overarching concept of art must be necessary if we are to characterize different art-histories as art histories at all. 3.1. Levinson’s answer to Currie’s criticism Levinson’s response to Currie is twofold; first, he argues against Currie that he misidentifies his target. Our concept of art is that of a practice which, what164 Enrahonar 38/39, 2007 María José Alcaraz León 12. CURRIE, ibid., p. 187. 13. In his answer to Currie’s criticism, Levinson focus upon the illicitness of the analogy between «water» and «art» that Currie holds in order to obtain his conclusion. As Levinson shows the bi-dimensional semantics of possible worlds valid for natural kinds identification does not provide a good frame for analysing historical concepts such as «art». ever the particular regards that have become part of its content, «would have (not) been different if actual art history had been different»14, simply because, in Levinson’s account the concept of art is one whose structure would not be different in other possible worlds, what would be different is its content, that is, the specific set of art regards relevant in each case. We must distinguish, then, between the concept of art —which, according to Levinson would have the structure he offers in his definition— and the meaning of art at t, which is given once we specify the extension of «art prior to t». The former does not suffer any change across different worlds, while the second would be receptive to the different histories in which its content is specified. The second line of argument developed by Levinson may be regarded as a consequence of the first. Thus, when trying to make sense of the role —demanded by Currie— of an overarching concept of art for identifying other possible concepts of art, he claims that this concept of art cannot be other than ours, and the art regards according to which we will identify other regards as artistic cannot be but ours as well. So, where these answers leave us then? We have seen that, according to Levinson, the problem raised by Currie is not an actual problem for his definition, for he claims our concept of art would be the same in other possible worlds; that is, it will be a concept defining a practice whose objects will be identified recursively according to the extension of the concept so far. The concept has a structure such that its content expands historically, regardless the specific art-regards which the artist takes into account as part of his or her intention. What would vary, from one world to another, would be the specific art-regards considered as relevant in each case, since different histories would yield different set of art regards. «Art» for them would mean something different from what «art» means for us, although the concept of art for both of us operates in the same manner, that is, recursively: through the relationship that an object has with the art history within which it has been forged by an artist. But then, Levinson acknowledges that it is likely that our means to recognize the artistic regards relevant for other worlds cannot be but relating them to our own relevant artistic regards. These need not be understood as features of our concept of art reached through a priori reflection, but as a set of features which contingently have constituted the content or meaning of our concept of art. That our relevant art regards are fixed a posteriori does not deprive them from their role at all; that they are contingent (some others could have been taking part of the relevant set and some of the included could have been absent if art history had been different) makes them no less essential. Moreover, that we cannot recognize others’ relevant art regards without making reference to our own relevant art regards seems a fact that we can hardly avoid. How else could we? Our regards are contingent but normative within our artistic practice given its actual history. Art’s historical nature Enrahonar 38/39, 2007 165 14. LEVINSON, J. (2002), «The Irreducible Historicality of the Concept of Art». British Journal of Aesthetics, vol. 42, nº 4, October, p. 367-379, p. 376. I believe that part of the resistance showed against Levinson’s proposal has something like the following thought behind: we expect our definitions to have a certain normative character, that is, they have the power to include or exclude art. In Levinson’s the art regards that provide art’s meaning are contingent, because historical, and thus we get that art’s essence is contingently determined, instead of aprioristically. The reason in favour of an a-historical account of art is, I think, that we are able to identify art in traditions different from our own tradition. But, once we identify art meaning with art history we seem unable to tell what art can be for others, since art has been identified with art «for us». There remains the point that it is likely that we cannot proceed differently in this matter: given art history, this is the way it is; so what we have to do is to give up worries about others’ art. 4. Stephen Davies’ criticism against artworld relativity Although differently stated, Stephen Davies15 has formulated a similar worry concerning definitional attempts that bring historical constituents into the very conditions of art identification. For him, Levinson’s definition would fall under the «artworld relativity problem» which, briefly stated, consists in the following: once we admit that what art is must be specified in relation to an artworld (a concrete art history in Levinson’s account) and, also, that there can be temporal and geographically distinct artistic traditions, artworlds, or art histories, we lack a criterion to identify artworlds as such from other similar structures whose modus operandi can be parallel to the one specified in Levinson’s definition. Since the specific art-regards that enter into art practice can vary from one history to another, we cannot, on the barely basis of a relational property such as the one specified in the historical definition, discriminate between different histories as art histories rather than other kind of practice histories, which in a similar way extends its range of objects through relating new items to past items within that practice. Shortly, the problem focus upon the content of the art-regards and the necessity of specifying what are they if we are to be able to separate art histories from other practices’ histories. It is not difficult to see that the element that will help grouping different histories as art histories is something very similar, if not the same thing, that could help identifying art across possible worlds in Currie’s argument. What we need is something that provides a criterion for art identification independently of the recursive mode given in Levinson’s definition. And this something seems to point to the content of the art regards mentioned by Levinson but unspecified within the definition. 166 Enrahonar 38/39, 2007 María José Alcaraz León 15. DAVIES, S. (2001), «Definitions of Art». In: GAUT, Berys and LOPES, Dominic McIver (eds.). The Routledge Companion to Aesthetics. P. 169-179. 5. Does Levinson’s historical definition provide a sufficient condition for art identification across different worlds or different traditions? Currie and Davies’ criticisms seem to converge at one point. Given the historical recursive structure of the concept of art we will be unable to discern other traditions as artistic, where the meaning of art, given their different histories, is different from ours. And we will not be able to discriminate between art and other possible historical practices, since they would have same structural definition. It will not be sufficient to detect a certain recursive enrichment of their concept of «art», for we should first of all recognize that that is their concept of «art» at all. That is, recursive definitions make other art traditions opaque to us. Either they have our concept, or we cannot recognize a certain practice as art on the mere basis of their practice’s recursive self-enrichment. As I see it, both problems belong to the same family insofar as they demand certain specification of the regards according to which an object can be legitimately related to previous art practice. But then, if those regards are specified, it seems that Levinson’s definition is not as purely historical as it aims to be. For, once we start specifying some regards as artistic over other non-artistic ones, but equally present in art practice, we surreptitiously introduce the kind of properties relational definitions tried to take off from the definitional ground, such as providing an aesthetic experience, being expressive, having significant form, etc. Both criticisms enact the conflict between two different kinds of art definitions I pointed out at the beginning of this paper: one is deeply engaged with providing a concept of art which focuses upon what Levinson calls art’s regards, that is, the features that have been traditionally thought to be features of art, and another, more formal as it were, which provides the formal structure of our concept of art so that it can be filled in with what we have been historically calling art; but are we leaving outside art history all that needs to be kept distinguished from art? Certain artistic regards have become obsolete not merely due to our habit changing, but also because art has traditionally served to purposes which are out of its strict domain now. For example, some of the regards under which art have been done are religious representation or propaganda and, although some of the works we acknowledge as art now could not have been done without these aspects playing an important role in their production, I think we can intuitively reject these regards as artistic. But how can we? The mere fact that certain regards have been operative in the past and have had a role in constituting artworks is enough for them to be regarded as sufficient to constitute new artworks, regardless of the fact they are considered as regards of the relevant artistic type or not. Thus, given the structure of the historical definition, we cannot provide an answer to this question because, in spite of Levinson’s attempt at showing that we do not need any concept else apart from the one we already have, we need to refer, at least implicitly, to certain intuitions concerning art we have in order to trace a borderline separating art relevant regards from non relevant regards for art. Art’s historical nature Enrahonar 38/39, 2007 167