Intertextualidade verbo-visual : como os textos multissemióticos dialogam ? / Verbal-Visual Intertextuality : How do Multisemiotic Texts Dialogue ?

The objective of this work is to understand how multisemiotic texts interact with each other to produce meanings, observing the complex intertextual relations among genres from various artistic and/or audiovisual fields. Therefore, I initially present a brief review of the literature on intertextuality, critically discussing how leading scholars address this issue. Then I argue that it is necessary to understand intertextuality in an integral and non-discretized way through a typological continuum of relationships between verbal-visual texts. Thus, I develop a model for understanding this phenomenon by means of a graph in which two continua intertwine: the representation of intertextuality through form (Implicitness/ Explicitness) and function (Approach/ Distance of the quoted voice) assumed in communicative situations. To test the model, four music video clips of American singer Madonna were selected so we can verify how music video texts rely on other texts to build their discourses and evoked identities.


Verbal-Visual Intertextuality: How do Multisemiotic Texts Dialogue? / Intertextualidade verbo-visual: como os textos multissemióticos dialogam?
Leonardo Mozdzenski * ABSTRACT The objective of this work is to understand how multisemiotic texts interact with each other to produce meanings, observing the complex intertextual relations among genres from various artistic and/or audiovisual fields.Therefore, I initially present a brief review of the literature on intertextuality, critically discussing how leading scholars address this issue.Then I argue that it is necessary to understand intertextuality in an integral and non-discretized way through a typological continuum of relationships between verbal-visual texts.Thus, I develop a model for understanding this phenomenon by means of a graph in which two continua intertwine: the representation of intertextuality through form (Implicitness/ Explicitness) and function (Approach/ Distance of the quoted voice) assumed in communicative situations.To test the model, four music video clips of American singer Madonna were selected so we can verify how music video texts rely on other texts to build their discourses and evoked identities.KEYWORDS: Intertextuality; Dialogism; Verbal-Visual Texts; Music Video RESUMO O objetivo deste trabalho é compreender como os textos multissemióticos dialogam entre si para produzir sentidos, observando-se as complexas relações intertextuais instauradas entre gêneros dos diversos campos artísticos e/ou audiovisuais.Para tanto, apresento inicialmente uma breve revisão bibliográfica sobre a intertextualidade, discutindo criticamente como os principais estudiosos abordam esse tema.Em seguida, defendo que é necessário entendermos a intertextualidade de forma integral e nãodiscretizada através de um continuum tipológico das relações entre textos verbovisuais.Assim, desenvolvo um modelo de compreensão desse fenômeno por meio de um gráfico em que dois contínuos se entrecruzam: a representação da intertextualidade por meio da forma (Implicitude/Explicitude) e da função (Aproximação/Distanciamento da voz citada) assumidas em situações comunicativas.Para testar o modelo, foram selecionados quatro videoclipes da cantora norte-americana Madonna, tendo por fim constatarmos de que modo os textos videoclípticos se apoiam em outros textos para construção de seus discursos e das identidades evocadas.

Introduction
Although nowadays it is commonplace to state that "contemporary culture is largely visual"1 (PELLEGRINI, 2003, p.15), incorporating image and other semiotic resources into linguistic research is still a taboo and has found much resistance in some more traditional approaches.As pointed out by Kress, Leite-Garcia and Van Leeuwen (2000), discourse analysis has historically focused on linguistic properties realized in texts, valuing verbal language in its oral and written modalities to the detriment of other semiotic modes.
Therefore, it is possible to establish that the elevation of the status of multisemiotic genres as an object of linguistic studies has just happened recently.
Although in other disciplines the examination of signs has been done since the Fifties and Sixties, it is just now that a more systematic and integrated research on multimodal texts can be observed in the field of Linguistics, embracing all semiotic resources that compose those texts and considering their inclusion in the so-called society of the image (KRESS; VAN LEEUWEN, 2001).
This work can thus be inserted in these new proposals of analysis by considering that the processing of textual information can only occur with the whole reading of the verbal text, the visual material (photographs, graphics, designs, colors, layout), the sound material (music, noises, intonation, rhythm), and so on.Otherwise, the defective reading can significantly affect the overall understanding of the unity of text.
In particular, I am interested in understanding how multisemiotic texts dialogue with each other to produce meaning.In other words, my purpose here is to observe the complex intertextual relationships between genres from various artistic and/or audiovisual fields.Thus I firstly present a brief review of the literature on intertextuality, critically discussing how leading theorists address this issue.Then, I argue that it is necessary to understand intertextuality in an unabridged and nondiscretized way through a typological continuum of relationships between verbal-visual texts.
To evaluate the operationalization of this analytical model that will be developed further, I have selected four music video clips of American singer Madonna in order to investigate how these multisemiotic texts rely on other texts to construct their discourses and the evoked identities.The choice for music video was made inasmuch as that it is a multifaceted audiovisual genre that reveals a current trend of integrating a large number of multisemiotic strategies to capture and hold the viewer"s attention.
Moreover, the researchers of this genre have already emphasized the importance of intertextuality in music videos: [I assume] the supposition that the meaning of video music clips lie in the analysis of the texts that preceded them, and that intertextuality, rather than indicating an "origin" of meaning implicates texts within a multiple variability of meaning based on both the context of the video music clip as text, and the context of the viewer (STOCKBRIDGE, 1987, p.158).

Reviewing the Notion of Intertextuality
Initially related to the study of literature, the concept of intertextuality was coined by Julia Kristeva (1980) by arguing that literary work redistributes earlier texts in a single text, so it is necessary to think of it as an "intertext".The author, however, from Bakhtin"s notion of dialogism, goes even further by considering that every text is an intertext in a succession of texts already written or yet to be written.
Thus, an efficient reading can never be done in isolation and that is why it is so important to understand how the origins, forms, themes, etc. of a text dialogue with several other texts (CHARAUDEAU; MAINGUENEAU, 2004, p.288).In this sense, Dominique Maingueneau (2005, p.21) sustains the primacy of interdiscourse over the discourse, arguing that "the most relevant unit of analysis is not the discourse itself, but a space for exchanges between several discourses conveniently chosen"2 .In fact, the idea that every utterance is constitutionally dialogical is already presented by Vološinov (1986), originally published in the 1920"s.In this perspective, the dialogic orientation consists of a characteristic feature of any speech, which is always traversed by the speech of others.
For Mikhail Bakhtin (1996, p.69), "any utterance is a link in a very complexly organized chain of other utterances".In other words, no utterance in a concrete discourse (i.e., speech) is told from a communicative "zero" or "vacuum".It is always in a constant dialogue with all that has been said about a specific topic, as well as everything that follows this "language presents the picture of a ceaseless flow of becoming" (VOLOŠINOV, 1986, p.66).As Cunha explained (2003, p.168), any "utterance is a response to an already-said, either in an immediate situation or in a broader context"3 .
Specifically resuming the notion of intertextuality, Bazerman (2004)  Piègay-Gros (1996) divides intertextual relationships into two types: relations of co-presence between two or more texts and relations of derivation of one or more text from a primary text.In the first group, the author lists: citation (the text is explicitly inserted in another); reference (similar to a quote, but with no literal transcription of the source text); allusion (the primary text is resumed in a subtle way by indications that the reader should realize); and plagiarism (the primary text is transcribed with no credits).
In the second group, there are parody (the structure and the theme of the text are reproduced in other situations with effects of carnivalization and playfulness); burlesque transvestism (rewriting a style in a "displaced form" from a book whose contents are preserved); and pastiche (imitation of a style using the same form of the text imitated).
For her part, Koch (2004) postulates the distinction between intertextuality and/or polyphony lato sensu and intertextuality and/or polyphony stricto sensu.While the former is constitutive of every discourse, the latter occurs when, within a text, another text (intertext) previously produced is inserted and forms a part of our social memory or of the interlocutors" discursive memory.The author also argues that intertextuality can be explicit or implicit.In this last case, the producers of the text do not mention the source of the intertext they introduced, hoping that their reader/listener recognizes its presence through the cognitive activation of the source text in their discursive memory; on the other hand, in the first case (explicit intertextuality), the source of the intertext is clearly mentioned.
In a later work, Koch, Bentes and Cavalcante (2007) resume these categories proposed by Koch (2004) Some criticisms can be made of these previous proposals that try to explain this phenomenon.Firstly, it is important to emphasize that the main problem of the classifications above lies in the fact that all tend to "discretize" intertextuality, grouping it into categories that seem to be constituted by separate, stagnant and well defined units.That is, according to these traditional classifications, intertextuality can apparently only be considered a "quote" or a "reference" or a "plagiarism" or a "reference", and so on.There appears to be no gradation or continuum between these categorical types.One can get the impression that the text is seen as a "semantic monoblock", which should be categorically enclosed in one of the possible discretized and non-integrated classes of intertextuality.
Secondly, a great part of these proposals of classification also uses, in theory, dichotomous categories to explain the phenomenon: intertextuality of similarities x of differences; implicit x explicit intertextuality; manifested x constitutive intertextuality; intertextual relations of capturing x subversion; shown x constitutive heterogeneity; marked x unmarked heterogeneity, and so on.It is clear that in our everyday discursive practices, we do not perceive texts as if they were intertextually divided and grouped into two -at first glance -antagonistic categories.On the contrary, we perceive as if they were on a continuum in which all these possibilities of occurrence of intertextuality can happen concomitantly.
Thirdly, as a final point, it is possible to observe the absence of more consistent and coherent criteria for grouping each type of intertextuality in the same category.In other words, phenomena like quote and paraphrase (mainly associated to the form of intertextuality) are treated as phenomena such as parody and pastiche (mostly related to the effects of the meanings produced by intertextuality).
It is from these reflections I present the following analytical model as a proposal to understand and examine intertextuality more systematically, seeking to advance the state of the art in discourse studies on this topic.
and Authier-Revuz, suggest that "it would be more appropriate to consider various degrees of explicitness" [originally in Portuguese: "seria mais adequado considerar variados graus de explicitude].It would be, in fact, a blend of what Bazerman (2004) calls "levels of intertextuality", "intertextual representation techniques" and "intertextual reach".

A New View of Intertextuality: Proposing an Analytical Model
In order to present a new analytical model of intertextual relationships, I initially seize upon the notion of explicitness, as elaborated by Marcuschi (2007, p.40): "to make explicit means to offer a discursive formulation that contains the conditions of appropriate or intended interpretability"6 .Accordingly, to make explicit implies to promote forms to make a text interpretable in contexts of use by creating some access conditions.
From an intertextual perspective, this means to state that the role of the speaker/writer is to gradually offer (or to refuse to offer) discursive-cognitive clues that make the text interpretation possible.Such clues are given according to the interlocutors" contexts, that is, their shared knowledge, their (inter)subjective interpretation of the communicative situation, their proposals, etc.7 .In synthesis, as to the form that intertextuality might assume in a text, I propose the typological continuum shown in Graphic 1.

Graphic 1. Typological Continuum of Intertextuality Regarding its Form of Occurrence
Therefore, as we can observe in Graphic 1, in terms of the source text"s explicitness, a text may ideally vary from plagiarism (a fraudulent presentation of other people"s work as if it were one"s own), where deliberately there are not explicit identification marks of the source text or its author, to authorized copies (a full reproduction, legally permitted, of an already existent text), as in the case of a compilation of scientific articles previously published in different academic journals.
It is crucial to emphasize that those traditional categories listed in this continuum (plagiarism, allusion, indirect mention, etc.) are merely illustrative and non-discretized.
In other words, a single text may simultaneously present one or more occurrences of any of these types of intertextuality, or any combination between these more or less stabilized categories and other "intermediate" classes.
The other criterion to observe the intertextual relationships regards the function performed by each intertextuality occurrence.More specifically, this criterion is linked to the positioning of the quoting author"s voice in relation to the quoted author"s voice in order to build his or her own discourse.
This idea of different voices inhabiting discourse is borrowed from the notion of polyphony in Bakhtin (1984), concerning the existence of several polemical voices in polyphonic dialogic genres, which are retaken, resignified, ratified, confronted and ironicized all the time.To use a metaphor by VOLOŠINOV (1986), discourse is conceived as a miniature arena, where these voices with several orientations (agreeing, contradictory, satirical, etc.) cross and fight each other.We might dispose, through the typological continuum of Graphic 2, the way how speakers operate with other people"s voices to produce certain meaning effects.where the defendant"s discourse is retaken by the prosecutor to defend the victim, or in journalistic articles, where the less privileged"s speech is made illegitimate through strategies of access -or the absence of access -to the discourse space (FALCONE, 2005).
On the other hand, the mention of authority happens, for instance, when a proverb is quoted to invoke "popular wisdom" as a rhetorical persuasion resource, or when the voices of the groups of power are introduced aiming to give credibility to the statement: "The government says that...", "According to specialists" opinions...", etc.
It is not worthless to say that, in all cases shown in Graphic 2, what is under analysis is the 'function' of intertextuality, that is, the meaning effects produced when the other"s voice is incorporated to the new utterance, and not necessarily the way this phenomena has occurred (object of Graphic 1)8.In addition, I ratify my understanding that the classic categories already exposed in Graphic 2 ("negative" quotation, "negative" paraphrase, parody, irony, pastiche, etc.) are listed only illustratively, and never taken as rigid, not interchangeable classes.
From these two continua (Graphics 1 and 2), I finally propose Graphic 3: Regarding the way this imagetic dialogism occurs, French scholar Jean-Jacques Courtine proposes the notion of intericonicity.According to Milanez (2004), Courtine starts from the principle that any image may be part of individual"s visual memoryand the relationships established between them are called "intericonicity".To Courtine (2006), more than a merely synchronic and technical conception of image analysis, the study of this phenomenon aims to show that, just as texts are textiles of intertextuality, images are permeated by this intericonicity, whose forms and displacements must be rebuilt from the investigation of their modes of production, circulation and reception in visual culture at a given historical moment.
The application of this term is naturally extended to any form of artistic manifestation.When talking about the literature-painting interface in Surrealism, for instance, Arbex (2000) calls our attention to the eminently intertextual nature of this movement"s artistic production, showing "works that are built as a mosaic of quotes".9 The author also makes use of this notion of intericonicity, defined in the same terms as the concept of intertextuality, that is, as a process of production of an image built as absorption or transformation of other images.Arbex (2000) sustains that, in the same way Bakhtin wanted for the literary text, intericonicity is related to the social context where it happens, being a mark not only of history and ideology, but also of aesthetics and visuality.
It is in this sense that, in an interview with Nilton Milanez, made in 2005 at the Université Sorbonne-Nouvelle (Paris 3), Courtine argues that intericonicity supposes the relationships of images external to individuals, as when an image might be inscribed into a series of images, a genealogy as an utterance in a network of formulation, according to Foucault.However, this also supposes taking into consideration all the individual"s image memory catalogues.Of all memories.Those may even be dreams, images that were seen and forgotten, images that reappeared, and also those imagined images we found in the individual (quoted by MILANEZ, 2006, p.168-169) 10 .
In this scenario, the importance of the role of memory for intericonic analysis is thus crucial: "Our memory is built, therefore, from the intertwining of places where we look for images that form the substance of our recollections" (MILANEZ, 2006, p.173).
How can this notion of intericonicity be applied to our analytical model, presented in Graphic 3, concerning the representation of intertextuality in terms of its form and function?To answer this question, I illustratively make use of four music videos starring American singer Madonna11 .
Firstly, from the comparison between the images shown in Figures 1 and 2 (below), it is clear that the music video Material Girl (Madonna, 1985 -Figure 1) visually dialogues very explicitly with Marilyn Monroe"s iconic performance in Gentlemen Prefer Blondes (Howard Hanks, 1953 -Figure 2).However, with a closer analysis of these works, it is possible to perceive a flagrant sliding of meaning between those two discourses, especially regarding the female identity built by Madonna in her video.
In the musical comedy directed by Howard Hawks (1953), dancer Lorelei Lee (Monroe) is a typical gold digger, interested mainly in her millionaire fiancé"s fortune.
On the other hand, in her music video Material Girl, Madonna "interprets" a rising star.
The video takes turns between two more evident fictional cores: as its main feature, it shows Madonna"s Marilyn-esque performance, almost reproducing the original choreography from the musical number Diamonds Are a Girl's Best Friend12 .In parallel, it shows the pop star in the backstage, refusing luxury presents and seeming dissatisfied with her wealthy suitors.If the music video -as Madonna intended -was a big irony, few people noticed that (TARABORRELLI, 2001).On one hand, in the lyrics of Material Girl (composed by Peter Brown and Robert Rans), Madonna"s image is very categorically built: "you know that we are living in a material world, and I am a material girl".On the other hand, in the video, Madonna refuses expensive presents and decides to choose her apparently poor suitor in the story, played by actor Keith Carradine.
In spite of the evident intericonic dialogue between the two productions, it is possible to notice that in the music video Material Girl there is a clear rupture from the original meaning of the film starred by Marilyn Monroe.If in the film the protagonist, adorable as she is, marries out of financial interest, in the video there is an evident parodic subversion of meaning, as Carradine"s character -who is actually a wealthy man -has to pretend being poor to win Madonna"s heart.
In other words, the music video ends up cognitively promoting the deconstruction of the original gold digger image -a foolish platinum blonde, dancer/singer at a nightclub (personified by Monroe"s Lorelei) -repositioning the character through the image of a powerful, independent woman (Madonna), whose happy ending does not depend on her suitor"s bank account.
That is, between these two multisemiotic texts there is a visible intertextual/intericonic explicitness regarding form (almost identical costumes and set decoration, a very similar dance routine, etc.), but we can notice a distance from the voice of the source text-film.The video is located therefore in the quadrant (4) of Graphic 4.

Graphic 4. Intertextuality between the film Gentlemen Prefer Blondes and the video Material Girl
With the live version of the song Vogue in MTV"s 1990 Music Video Awards, which happened in Los Angeles (USA) on 06/09/1990 (Figure 3), Madonna surprised her fans by retaking an important pictoric icon in History: the last French queen, Marie Antoinette (Figure 4)."Madonna [...] goes back to the etymology of the word "vogue", dressed up like Marie Antoinette, her dancers moving around her in the powdered faces, wigs, and heaving bosoms of the eighteenth-century French court", describes one of the singer"s biographers, Lucy O" Brien (2007, p.140).
According to historian Evelyne Lever (2000), Marie Antoinette Josèphe Jeanne de Habsbourg-Lorraine was a controversial and fascinating woman.Her enchantment with a luxurious way of life and total alienation from the people"s problems earned her the reputation of a futile, selfish and arrogant queen.However, in spite of being hated by the Versailles court, Marie Antoinette knew how to leave a personal mark, strongly influencing the court"s habits and fashion (dresses, jewellery, hairdos) and the architecture and decoration of French monuments.Madonna"s reading of Marie Antoinette"s image for her MTV"s 1990 Music Video Awards performance is eminently "pastiching".Here, there is not the irony of the music video Material Girl or the attempt to cognitively deconstruct the source image.
On the contrary: the exaggerated, farcical characterization only reinforces the collective imaginary of the French queen"s frivolity and excesses.The singer incorporates Marie Antoinette and her court"s look merely for the audience"s entertainment, without any intentions of political or artistic criticism or satire.It is, therefore, a clear pastiche.
According to Charaudeau and Maingueneau (2004, p.371), pastiche consists of an "imitation practice"13 , with a ludic objective, hence distinguishing from parody, which has an eminently subversive, contesting and oppositionist nature.The pastiche user normally leaves clear evidence of the purposes of their text, by express indication, the charicatural nature of the contents, or stylistic marks.
Thus, from the live music video Vogue, it is possible to establish two simultaneous intericonic relationships between the discourses and the images built for the characters.Firstly, there is explicitness regarding its form: an "authentical" scenic composition of the costumes (obviously respecting the adaptations needed for a musical performance), makeup, hair styling, set decoration, habits (e.g., one of the dancers does snuff), etc.Secondly, it is also possible to observe a certain degree of approach to the source text"s voice -or, at least, to the social imaginary existing about the text -since Madonna would be "paying a tribute", although a playful one, to the French queen.The video is, therefore, in the quadrant (2) of Graphic 5.

Graphic 5. Intertextuality between Marie Antoinette and Madonna's live version of the video Vogue
The music video clip Cherish (Madonna, 1989), in turn, was entirely filmed in Malibu, a beach in California (USA).Making use of a beautiful black and white film, director Herb Ritts -up to that moment one of Madonna"s favourite photographers, with no experience directing music videos -puts the singer in a paradise-like setting, surrounded by mermen.The song is, in fact, a simple happy pop love song, celebrating the passion and devotion of a woman for her lover, whose destiny is to be forever by her side.Madonna appears in a very comfortable, self-confident manner in the video, wearing a tight black dress and looking repeatedly at the camera, alone (smiling, dancing, showing off her muscles, almost always in contact with the ocean) or playing with a child mermaid, while mermen do acrobatics in the water.The tone is light, sweet and fun, and Madonna transmits the image of a radiant, playful woman, who according to the song"s lyrics is totally in love (Figure 5).

Some art and movie critics noticed a clear similarity between the music video
Cherish and the short feature film At Land (1944), by Ukrainian moviemaker, naturalized American, Maya Deren15 .Having her production concentrated between the 1940"s and 1950"s, Deren was also a poet, writer, photographer, dancer, choreographer and film critic, being considered a brilliant artist and strongly influencing contemporary directors of music videos and movies (TURIM, 2001).In her film At Land, Deren not only directs, but also plays the main character, a distressed, restless woman.The film starts with Deren bathing in the sea, and from that moment on we see a visually narrated journey (it is a silent movie), very surrealistic, in which she meets several people -many of them ignoring her presence -and different versions of herself.The tone of the movie is sombre and often disturbing.The image created for the woman is of a distressed, anxious person, always pursuing (and being pursued by) something that is not clear.
In spite of building very different female identities, the music video Cherish and the film At Land establish a visual dialogue, at least implicitly -since neither Madonna nor the director Herb Ritts have acknowledged any influence of Maya Deren"s work.In the famous blog "Madonna Scrapbook", Whacker (2010) noticed the similarity between the two productions (Figure 6).And in the contemporary art criticism website "Freak Show Business", Santos (2009) presented an interesting comparative photomontage, where the evident imagetic dialogue between the two works is also noticeable (Figure 7 attached).
We can come to the conclusion that there is an imagetic approach between the video Cherish (Madonna, 1989) and the short film At land (Deren, 1944), since both display similar forms: black and white photography, the protagonist"s identical costumes, similar body movements and settings (a beach, the ocean, waves), etc.These analogous forms, however, are implicit, since there was never any indication (either by clues in the music video or in interviews with the artists involved) that Maya Deren"s work served as visual influence for the composition of the music video.This is, in fact, a "discovery" -or, to be more precise, a supposition -of fans and critics.
Regarding the image built for the woman, we observe that there is an evident distance between those two female personas: whereas Madonna plays a happy woman in love, Deren embodies an anxious, tormented woman.The video is situated, therefore, in the quadrant (3) of Graphic 6. Music video Hollywood (Figure 8), besides its overwhelming plastic beauty, has another peculiarity in Madonna"s videography: it was formally accused of plagiarism by Samuel Bourdin, the owner of the copyrights of his father Guy Bourdin"s photographs.
It is not an exaggeration to say that Guy Bourdin (1929Bourdin ( -1991) ) was one of the most important, influent and innovative fashion and publicity photographers of all times.For the critic Alison Gingeras (2006), Bourdin was responsible for redefining concepts and establishing new aesthetic standards in the fashion and publicity worlds, mixing surrealism, glamour, irony, sensuality and a lot of controversy in his work.Madonna and director Jean-Baptiste Mondino retook a series of images originally produced by Guy Bourdin, presenting in the music video Hollywood an unaccredited "re-reading" that was dangerously close to the source text, as it can be seen in the photomontages in Figure 9, attached.
In an interview featured in the Brazilian TV show GNT Fashion (GNT Channel, 14/09/2009), TV presenter and fashion editor Lilian Pacce asked Samuel Bourdin to clarify the story of the legal process against Madonna.He said that in spite of the respect he holds for the singer and the music video director, the plagiarism is evident: "[...] what they did was a collage [of Guy Bourdin's works] with modifications, and there is no credit, no acknowledgement of the source.My father"s influence was not recognized"18 (BOURDIN, 2009).
In intericonic relationships established in artistic productions, the borderline separating "inspiration" from mere plagiarism is not always clear.Even though in the legal realm the definition of plagiarism seems to be in principle precise, when we take into account the complex intertextual webs built between verbal and non-verbal discourses -especially involving works of art -this evaluation is always a subject of controversy.To avoid unnecessary discussions on the theme, I assume here as a criterion of definition of what plagiarism is (or not) the legal or administrative decision establishing a text (verbal or non-verbal) as original or result of illegal appropriation.
As for the intertextual function of our model, it is clear that both the iconoclastic pictures of Guy Bourdin and the corrosive lyrics of the song Hollywood follow the same direction in terms of giving a critical look at the status quo.With his lenses, Bourdin managed to unite a highly sophisticated glamour and a ferocious tone about the fashion world, through images "full of sexuality and violence, escaping from the obvious in daily life scenes" (PACCE, 2009).In the lyrics of Hollywood (composed by Madonna and Mirwais Ahmadzaï), the singer is clearly ironic: "Everybody comes to Hollywood / They wanna make it in the neighborhood / They like the smell of it in Hollywood / How could it hurt you when it looks so good?".
On the other hand, regarding the form of intertextuality between Bourdin"s and the music video"s images, it is possible to see an obvious visual and "attitude" similarity between the female identities built in both cases.Madonna, however, by not acknowledging the authorship of the source text, had the intention of leaving this intericonicity only implicit.The video was conceived, therefore, as pertaining to the quadrant (1) of Graphic 7 -naturally, before the revelation of the plagiarism 19 .

Final Considerations
We have explained throughout this paper how the proposed analytical model for intertextuality operates.There are, however, some issues yet unsolved, raised here as a provocation for possible future debates.One of them is the following: if there is an equally evident formal visual similarity between Figures 1 and 2

(Madonna in Material
Girl and Marilyn Monroe in Gentlemen Prefer Blondes) and Figures 8 and 9 (Madonna in Hollywood and Guy Bourdin"s images), why is that the music video Hollywood is considered as plagiarism and Material Girl isn"t?
A possible answer to this question might be found in the notion of memory socially shared in a given culture.In this sense, hence, Marilyn"s exuberant image singing Diamonds Are a Girl's Best Friend is already a part of the Western culture"s social memory, thus constituting a founding discourse.That is, although it is possible to identify the source text in a precise way (and obviously its authorship), it is already part of our collective imaginary, regardless of being legally public domain (something that also happened in the case of the live music video Vogue in relation to Marie Antoinette"s image). 19Under an out-of-court settlement agreement, Madonna and others accused of plagiarism recognized the formal intericonicity between the works and the music video ended up, therefore, in quadrant 2, similar to Graphic 5.
A flagrant evidence of this hypothesis may be noticed in the several "rereadings" of Marilyn Monroe"s iconic performance in different media, such as printed and video advertisement, movies, etc.On the other hand, Guy Bourdin is still restricted to a small group of fashion and publicity professionals.Despite having his images printed in various international publications for over thirty years, the French photographer remains unknown to the larger audiences.
In other words, because they are not part of our collective memory, the female images created by Guy Bourdin -differently from Marilyn"s anthological performance -still possess a very strong authorial mark, which prevents them from being borrowed without the due authorization and acknowledgement.Besides, it is also necessary to consider the fact that the wide divulgation of a film, especially when it features a renowned star like Marilyn Monroe, enables a more solid and lasting power of cognitive assimilation and social crystallization, compared to advertisement campaigns in fashion magazines or video artworks produced by Bourdin.
Therefore, as we can observe from the discussion of the examples of music videos previously analysed, it is crucial that we take into consideration the role played by intertextuality in the meaning making process of any text.Evidently, not every music video will fit so precisely in one of the graphics above.It is possible to find music video texts dialoguing with other texts in a multifaceted, unclear way, which Fairclough (1995, p.15) calls "mixed intertextuality" where texts or discourse types are welded in a more complex, less easily separable way.
Hence we can see the importance of having selected, for this paper, examples from an artist with extensive videographic work.Based on the choice and discussion of Madonna"s music video clips, it was possible to make a "didactic" analysis of how intertextuality might be understood in terms of two continuums, regarding form and function (i.e., meaning effects built from the retaking of the source text"s author"s voice).Finally, it is relevant to emphasize the need to change the still prevailing dichotomized perspective that "texts are for reading" x "images are for seeing".With the presence of increasingly multisemiotic texts in contemporary culture, it is important to pay attention not only to their verbal elements, but also to all other components (images, sounds, etc.).This is the only way to realize an integral and appropriate interpretation of text and the different ways how it dialogues with other texts socially circulating.

Graphic 2 .
Typological continuum of intertextuality regarding its function of occurrenceTherefore, as it can be seen in Graphic 2, a quoting text may ideally vary from the situation when the other"s voice is disqualified until the moment it is used with the function of authority to guarantee the validity of the new utterance.The first case (disqualification) typically occurs in political and scientific debates, as well as in trials,

Graphic 3 .
Representation of intertextuality by its form and functionGraphic 3 synthesizes, within the discursive-sociocognitive approach adopted here, the representation of intertextuality through both form (Implicitness ↔ Explicitness) and function (Approach ↔ Distance of the quoted voice) assumed by this phenomenon in communicative situations.3Intertextuality in Music Video Clips: Testing the ModelOnce one understands the manner how intertextual relationships in socially circulating discourses take place, it is necessary to deepen the discussion about how images specifically dialogue with one another, before we apply the proposed model to the analysis of music video clips.