Glossar-English:Syntactically Incorrect Images

Sub-Item to: Image Syntax

Original: Syntaktisch unkorrekte Bilder

---

Can there be syntactic errors in images? The concept of syntactic correctness is initially familiar in its linguistic application: Example 1: ‘colorless green ideas sleep furiously’; Example 2: ‘sleep ideas green furiously color­less’. The difference between a syntactically well-formed sentence on the one hand (see example 1), whether it has meaning and practical use or not, and a syntactically incorrect sequence of words on the other hand (see example 2) is evident to anyone with a basic knowledge of the English language ([Chomsky 1957a]Chomsky, Noam (1957): Syntactic Structures. Mouton, Den Haag. : p. 15). Although the historical example only refers to the English language, it is true that language in general – in the sense of ‘la langue’ – only classifies a small part of the multitude of possible sequences of syntactic elements as well-formed sign carriers and, thus, separated from syntactically incorrect combinations. According to Chomsky, being able to distinguish syntactic well-formedness should be based on language competence itself. Is it possible to argue successfully for an analog distinction in the case of images? First of all, it should be noted that this question is ultimately less about images (in the actual, i.e., full sense of the word) than about potential image carriers: The term ‘image’ is at best imprecise (and at worst simply wrong) if pragmatic and semantic aspects are to be explicitly disregarded. At first glance, there does not appear to be any comparable classification, i.e., one that only refers to the structure of the (potential) image carrier (cf. [Langer 1984a]Langer, Susanne (1984): Philosophie auf neuem Wege. Das Symbol im Denken, im Ritus und in der Kunst. Fischer, Frankfurt/M. ). A spatial rearrangement of syntactic (or morphological) elements alone, based on the language example, is obviously not enough: The result is still an acceptable image carrier – which, by the way, anyone can easily convince themselves of using a computer-aided image processing system. It is valid for specific subclasses of images, i.e., certain pictorial sign systems, which comprise only a small part of all possible image carriers, that the associated sign carriers can be divided into well-formed and non-well-​formed ones according to language-​like syntactic rules – the system of technical drawings may serve as an example. However, it is not easy to see how this reference to a rule could be transferred to images in general and thus be justified in image competence itself ([Plümacher 1999a]Plümacher, Martina (1999): Wohlgeformtheitsbedingungen für Bilder?. In: Sachs-Hombach, K. & Rehkämper, K. (Hg.), Bildgrammatik. Interdisziplinäre Forschungen zur Syntax bildlicher Darstellungsformen. Scriptum, Magdeburg. ). The presumed lack of a real criterion of syntactic well-formedness is closely related to the often-claimed lack of a pictorial alphabet (⊳ pictorial grammar). While it is always true for linguistic sign systems that a genuine subset of the theoretically possible phonemes or graphemes are used as building blocks of complex signs, it seems that no corresponding limited graphic alphabet can be given that pictures, in general, are based on ([Sachs-Hombach 1999a]Sachs-Hombach, Klaus (1999): Gibt es ein Bildalphabet?. In: Sachs-Hombach, K. & Rehkämper, K. (Hg.), Bildgrammatik. Interdisziplinäre Forschungen zur Syntax bildlicher Darstellungsformen. Scriptum, Magdeburg. ). Since the lack of an alphabet also makes the combinatorial rules of correct composition obsolete, but the syntactic well-formedness of language is based on the respective combination rules for complex characters from the basic elements of the alphabet under consideration, it follows that no corresponding criterion can be formed for images. For a more precise treatment of whether there can be syntactically incorrect images, the distinction between the geometric basic structure and the visual marker dimensions has proven helpful. Real syntactic errors: Disturbances in the basic geometric structure In fact, all changes, both in the quantity and selection of the existing geometric elements and their arrangement, as well as in the assignment of visually distinguishable markings of the geometric parts, always only lead to different image carriers, which can only be said of that they are better or worse to use with regard to certain pictorial-pragmatic functions, but not that they are syntactically incorrect, i.e., they are not actual images simply because of certain peculiarities of the material. As in the concept of syntactic density, in images, in principle, any number of small changes in the sign carrier lead to other images: It follows that every sequence of such minimal changes also leads to an image again. Consequently, there can be no changes in an image surface produced by moving or exchanging areas that do not lead, in turn, to a syntactically acceptable image surface. Figure 1: A syntactically incorrect image? (A.E. Arkhipov: «Peasant Girl» (1920s), with a tear in the picture carrier) However, there are entirely everyday actions on a picture carrier that directly affect its function as a picture carrier from the ground up, namely the perforation or partial cutting or tearing of the paper, the canvas (among others), and similar damage. A painting (e.g., by Rembrandt), a sketch (e.g., by Picasso), the paper extract of a thermal imaging camera, as well as a family snapshot, each with a cut or hole, does not simply become a different image, but the same image with damage (⊳ Identity criteria for image carriers and identity of pictorial signs). What happens here, seen syntactically and morphologically? Suppose an image carrier is injured in the manner mentioned. In that case, the topology of the two-dimensional geometric basic structure is partially severely disturbed: Instead of the homogeneous (i.e., evenly distributed) and isotropic (i.e., equally valid in all directions of the image surface) neighborhood relationships between the spatial locations, gaps now appear between originally neighboring locations that do not belong to the image carrier, so that these spatial locations are now simply missing actual neighboring pixels. The entire geometric organization of the image carrier is thus disrupted because all higher-order pixemes (⊳ image morphology) based on these neighborhoods are affected. Therefore, such ruptures particularly affect the basic geometric structure, which has now become incomplete and inconsistent. The assumption that trimmings at the edge are considered qualitatively less serious than cuts within the picture area, since this changes semantic and pragmatic aspects but does not disturb the connection between the locations within the picture area, speaks in favor of considering the absence of disturbances in the basic geometric structure as the basis for the syntactic well-formedness of pictures. The complete separation of a part of the picture carrier ultimately leads to two other (cropped) pictures, whereby the type of separation is rather secondary: a torn edge also forms an acceptable frame. In contrast, in partial ruptures, the two now separated sides remain part of the same picture surface, whose disturbed basic geometric structure impairs the perception of the image. Incidentally, a disruption of the basic geometric structure must be distinguished from transparent areas of the image, such as those that occur in stained glass or picture windows. Such transparent areas correspond to quite regular pixemes with a unique, albeit less frequently used marking compared to all images. In fact, syntactic special cases such as reflection and transparency in images have an implicitly deictic quality: Aspects of the environment of the image carrier play a role here. When looking at the picture (carrier) from just one perspective, they are therefore not necessarily noticeable – it could be a question of a corresponding regular coloring of the spot. Only a movement of the viewer and the resulting shift in perspective makes it clear that these places are variable and depend on the environment. This connection often makes citing syntactically incorrect images problematic, namely if an image of it is presented instead of the destroyed image carrier. As demonstrated in Figure 1, the mapping of a syntactically incorrect image is itself (usually) a syntactically well-formed image: The cut is represented by a regular pixem with normal color markings, and the topology of the basic structure is not affected. Pragmato-syntactic errors: Disturbances in marker assignment While damage to the spatial cohesion of the basic geometric structure leads to real syntactical errors, which can be distinguished solely based on syntactical properties, problems in the dimensions of the marker values ​​can only trigger syntax errors that can be determined as such merely by pragmatic factors. The splintering off of pigments, which makes the ground visible, can only be identified as an error in the assignment of the marker values ​​– and, thus, as non-pixemes – in relation to the supposed normal state of the picture under consideration. It could just as well be intentional, i.e., a pixeme intended in this form (cf., e.g., sgraffito techniques). That is also reflected in the fact that the pictorial citing of such a disturbed image is significantly less problematic: The color and texture values ​​that the quote shows at the damaged point are, in fact, in the damaged original at that point. Reflexive use of syntactically incorrect images As an indirect pictorial quotation, Figure 1 represents a typical case of reflective use of an image: The presentation of the image is not intended to refer to a farm girl but as an example of some aspects of the use of the image itself. Just as the example of syntactically incorrect word order is quoted in a book on language syntax, the example of a geometrically disturbed image carrier is shown as a negative example of syntactically well-formed image carriers. Figure 2: Lu­cio Fon­ta­na: «Con­cet­to Spa­zia­le» (1965) Otherwise, syntactically incorrect images used directly in a reflexive manner also appear in an artistic context: L. Fontana (1899–​1968), in particular, plays with the deviation from syntactically well-formed image carriers. His sectional images are reflectively used image carriers, with the help of which attention can be drawn to materiality as a primary condition of image use through a negative exemplification (cf. [Whitfield 1999a]Whitfield, Sarah (1999): Lucio Fontana. Hayward Gallery Publ., London. , [Hynes 2005a]Hynes, Maria (2005): Rethinking Reductionism. Culture Machine, 7. Webadresse: www.culturemachine.net/index.php/cm/article/viewArticle/33/41 Notiz: online journal and [Sachs-Hombach 2003a]Sachs-Hombach, Klaus (2003): Das Bild als kommu­nika­tives Medium. Ele­mente einer allge­meinen Bild­wissen­schaft. Halem, Köln. : p. 212ff). The requirement of an undisturbed two-dimensional spatiality as the basic geometric structure of the physical image carrier for the regular use of the image catches the eye precisely because of its violation. Accordingly, “pictures” of the kind shown in Figure 2 would be pictures – if one does not want to understand them as borderline cases to sculpture in any case – only in the same improper sense as the sequence of letters in example 2 above occurs in a merely improper sense as a sentence in a linguistics textbook. On the other hand, if one were to understand them as quite regular images, one would ignore the point that also constitutes the essential difference between the image ‘Figure 2’ itself and the work by Fontana reproduced there. Syntactic (in)correctness with special image classes Diagrams, engineering drawings, and other groups of structural images are usually associated with certain explicit conventions of representation. Some of these rules can be checked at the syntactical level so that a form of syntactical incorrectness can be defined for these groups of images as a result of violations of the respective representation convention. The same applies to other conventions for delimiting a subset of images or an image medium, as these conventions can be determined using syntactic features. That may concern particular aspects of the style so that with regard to a particular art-historical epoch, for example, a picture can be judged as syntactically well-formed or incorrect. The causes are usually syntactically discrete subsystems embedded in the syntactically dense dimensions of pictorial signs. That also applies to the simpler versions of digital images with their local (possibly also temporal) and color space rasterizations. Relative to the restricted image manifolds encompassed by such a notational formalism, images with pixemes beyond the grid must be considered syntactically incorrect. In practice, this case is of little relevance for computational visualistics since usually, the most similar syntactically correct image is simply substituted instead of such an image that is syntactically incorrect with regard to the digitization convention, which is often sufficient. However, various technical problems can also lead in computational visualistics to the appearance of syntactically incorrect images in the narrower sense. Of course, this is not about digital images of syntactically incorrect images (such as Fig. 1), which are usually syntactically correct themselves. For example, disturbances in the marker values ​​can be caused by errors in the projection process: “Blind” areas of the screen mean that the color display no longer works appropriately at certain locations. Even more severe are technical problems when transferring or saving that lead to errors in the notation: The syntactic description of the image carrier can become incomplete or inconsistent. In the borderline case, parts of the image surface can still be projected, but, as in the case of the torn canvas, there are surfaces within the frame that indeed do not belong to the image carrier at all – in other words, a massive disruption of the basic geometric structure of that image. See also: Artistic image and everyday image Attention Color as an image-syntactic category Diagramm Digital image Exemplification Framing, frame Identity criteria of image carriers Identity of pictorial signs Image grammar Image in reflective use Image media Image morphology Image quotation Material Materiality Notation Original Reflection and transparency Sign, sign carrier, sign system Space and geometry Sculpture Structural image Syntactic density

Notes Literature [Chomsky 1957a]Chomsky, Noam (1957): Syntactic Structures. Mouton, Den Haag.[Hynes 2005a]Hynes, Maria (2005): Rethinking Reductionism. Culture Machine, 7.[Langer 1984a]Langer, Susanne (1984): Philosophie auf neuem Wege. Das Symbol im Denken, im Ritus und in der Kunst. Fischer, Frankfurt/M.[Plümacher 1999a]Plümacher, Martina (1999): Wohlgeformtheitsbedingungen für Bilder?. In: Sachs-Hombach, K. & Rehkämper, K. (Hg.), Bildgrammatik. Interdisziplinäre Forschungen zur Syntax bildlicher Darstellungsformen. Scriptum, Magdeburg.[Ros 1979a]Ros, Arno (1979): Objektkonstitution und elementare Sprachhandlungsbegriffe. Hain, Königstein/Ts.[Sachs-Hombach 1999a]Sachs-Hombach, Klaus (1999): Gibt es ein Bildalphabet?. In: Sachs-Hombach, K. & Rehkämper, K. (Hg.), Bildgrammatik. Interdisziplinäre Forschungen zur Syntax bildlicher Darstellungsformen. Scriptum, Magdeburg.[Sachs-Hombach 2003a]Sachs-Hombach, Klaus (2003): Das Bild als kommu­nika­tives Medium. Ele­mente einer allge­meinen Bild­wissen­schaft. Halem, Köln.[Sacks 1996a]Sacks, Oliver (1996): Migräne. Rowohlt , Hamburg.[Whitfield 1999a]Whitfield, Sarah (1999): Lucio Fontana. Hayward Gallery Publ., London. Version 1: 2013 Responsible: Schirra, Jörg R.J. Editing: Liebsch, Dimitri Contributions by: Joerg R.J. Schirra und Marcel Lemmes — (Note) Quote as: Schirra, Jörg R.J. (2013). Syntactically Incorrect Images. (Ausg. 1). In: Schirra, J.R.J.; Halawa, M. & Liebsch, D. (Hg.), Glossar der Bildphilosophie. (2012-2026). [bildglossar.org/index.php?title=Glossar-English:Syntactically_Incorrect_Images].