On digital photo-index C.S.퍼스와 기호학

 * This paper was published in the book titled, Peirce in His Own Words: Semiotics, Communications and Cognition, edited by Torkild Tellefsen et al, Mouton de Gruyter, Germany, 2014

(…) as because it is in dynamical (including spatial) connection both with the individual object, on the one hand, and with the senses or memory of the person for whom it serves as a sign, on the other hand.(CP 2.305; c. 1901)


   Can digital photographs be regarded as indices like analogue ones, despite the fact thatthe technology of photography has changed drastically since its inception? In the search for the identity of photography, many contemporary photo-theorists have tended to regard it as a kind of index following the footsteps of Charles S. Peirce, and every so often, photographs have been treated as traces of the real in their writings.[1] But the validity of ‘photo-index theory’ has been questioned since the widespread of digital cameras in the 1990s.

Since Peircean index is usually defined as a sign which has a direct physical (i.e., causal)connection with its object, a number of photo-theorists have cast doubt on the notion that a digital photo can be an index on the grounds that no physical causality is found between a digital photo and its object.[2] It is not altogether clear, however, that this kind of direct physical relation is what Peirce had in mind when he said that an index is “a real thingor fact which is a sign of its object by virtue of being connected with it as a matter of fact and by also forcibly intruding upon the mind, quite regardless of its being interpreted as a sign” (CP 4.447). My doubt was raised when I saw Brian Walski’s documentary picture taken at the Iraq War in 2003. [See Figure1 and 2] Controversy surrounding this composite picture, a combination of two different source photos, eventually led Walski to resign from his job at The Los Angeles Times.

[Figure 1. the composite photo by Brian Walski appeared in The Los Angeles Times (2003)]

 [Figure 2. the source files]

All digital photos can be regarded as Peircean indices when Peirce's actual notion of index is closely examined, although some digital composite photos must be treated as sub-indices orhypo-seme depending on their types. Before addressing these different typesof digital photos and indices, I will first provide a contextual basis for abroad overview on what an indexical sign is.

Signs are classified in relation to their objects, giving rise to Peirce’s well-known trichotomy of signs: icon, symbol, and index. An icon is a sign whose appearance is similar to its object, since an icon and its object should share the same qualities. According to Peirce a symbol is identified with a sign “which refers to the Object that it denotes by virtue of a law, usually an association of generalideas.” (CP 2.249) On the other hand, an index is a signwhich is physically connected with, or, which “pointsto,” its object. Unlike the icon and symbol,which are considered general, the index must be individual; otherwise, it cannotphysically connect to its object.

Thedistinguishing attributes of the index, asdescribed above, were explored by Thomas Goudge in the 1960s. Among his several points aboutthe index, perhaps the most important is thatof its being the “identifying sign, because “identification is accomplished only by means of an index.”(52-53)Althoughevery sign represents its object, it cannot provide the whole truth about theobject since “a sign is something which stands to somebody for something insome respect.”(CP 2.228) According to Peirce, every sign not only hasits own interpretant, but also is itself an interpretant of the preceding sign. In other words, no sign canrepresent its object without mediation. An exception,however, could be made when it comes to the interpretation of the index sinceit is “in dynamical (including spatial) connection both with the individualobject, on the one hand, and with the senses or memoryof the person for whom it serves as a sign, on the other hand..” (CP 2.305) For instance, the Pole Star which is in spatial connection with the direction of north, isalso an index since it allows us to identify the northsky when we find it.

Goudge sorted out six distinctive features of indicesin the same paper and provided an initial guideline for differentiating indicesfrom icons and symbols. Among the six characteristics of indices, the first andsecond ones illustrate his most important points: “(1) an index has a direct physical [i.e., causal] connection with itsobject, […] (2) an index exerts a compulsive influence on its interpreter,forcing him to attend to the indicated object.”(53) With Goudge’scharacterization of the indices, however, one must exclude many examples that Peircehimself provided from the group of genuine indices because of their lack ofphysically direct connection with the objects.[3]Hence, Goudge concludes that Peirce could not establish a comprehensive andcoherent theory of indices. Pole Star, in accordance with his argument, is aninstance that betrays Peirce’s failure in providing a consistent theory ofindices, since it merely has spatial connection with north sky but lacks anycausal relation with it. Yet, Peirce wouldn’t have accepted Goudge’s definitionof genuine index, when he mentioned that “If the Secondness is an existentialrelation, the Index is genuine. If the Secondness is a reference, the Index isdegenerate.” (CP 2.283)

More recently, Albert Atkinhas identified five characteristics that all indices should have. (163-164) Unlike Goudge, he doesn’t argue thatPeirce regarded direct physical connection between an index and its object asan integral part of index theory. The first and foremost featureof the indices that Atkin distinguishes is called the‘significatory feature’: “Indices use some physicalcontiguity with their object to direct attention to that object.” This feature has two components:physical contiguity and attention directing. To grasp the significance of the feature with two components, you may recall Peirce’s triadic definition ofsigns, for instance, that which appeared in CP 2.228: “a sign is somethingwhich stands to somebody for something in some respect.” In his definition, Peirce distinguishes two major relationships – ‘sign-objectand sign-interpretant’ – and notes that every kind of sign,including icon and index, should be considered in the context of theserelationships, no matter how weak they may be. Based on the fact that Peirce has never restricted indexical relations - ifgenuine or degenerate - to the causal ones, Atkin calls the first component “physical contiguity,” notphysical causality.

Because an index has a direct connection with its object, the role of the interpreting subject is minimized. Minimal as the rolemay be,it also needs to be interpreted as a sign. Interpretation of an index is largely determined by its object because “the characteristic function of the index” is“forcing the attention upon its object.”(CP 2.357) Therefore, weinterpret an index in terms of its dynamical relationship to the object. Index’s directing attention to itsobjectis about producing interpretant of the sign. When Peirce said that “Index, which is a Sign whose significance of its Object isdue to its having a genuine Relation to that Object, irrespective of the Interpretant”(CP 2.92; my emphasis), his point was not that the index cannot have interpretants, but rather thatan index can function as a sign withoutproducing any actual interpretant.[4]

The five features of the indices that Atkin identifiesare good references to discern the identity of index. Besidesthe aforementioned significatory feature, the other four features of the indices are independence, singularity,indicatory, and phenomenological ones. (163-166) The independence feature,derived from Peirce’s statement that “an index […] is a real thing or factwhich is a sign of its object […] quite regardless of its being interpreted asa sign”(CP 4.447), refers to the relation between an index and its object whichis independent of its interpretation, because the index and its object are real.[5] Regardingthe third feature, singularity, Peirce elaborates on the index as a sign “whichlike a pronoun demonstrative or relative, forces the attention to the particular object intended withoutdescribing it.”(CP 1.369; my emphasis) It is due to the singularity feature thatan index and its object can make an inseparable pair since it involves theexistence of its object. (Goudge, 53-54) Fourthly,the indicatory feature is derived from Peirce’s statement that “indices assertnothing”(CP 2.291), but Instead they just point it out. The final feature thatAtkin identifies as indexical is concerned with the category of secondness: therelation between an index and its object shows the brute existence of aphenomenon. (CP 2.283) Indices, unlike icons or symbols, don’t have to resembleor share law-like relation with their objects. However, it must bemade clear that the possibility of classifying the signs don’t deny the fact thatall three signs are in continuum. The general such asan icon and a symbol cannot designate the object without indices that embodythe quality of an icon and exemplify the law of a symbol.

Following the above explanation, a certain phenomenoncan be classified as an elementary index if it satisfiesall of the features Atkin described. Conversely, others that have only some ofthe features can be treated as sub-indices such as proper names, personaldemonstrative, or relative pronouns.[6] Letme get back to the question I raised in the beginning: are digital photosindices? In order to address this question, it isnecessary to distinguish the different types of digitalphotos.

There are three different kinds of digital photos: originalpicture files and two other types of composite pictures. The first one refersto pictures that are not altered or fabricated. If we don’t take physicallycausal relation between an index and its object, an original file produced by adigital camera may be treated as an index according to Peirce, who said that “aphotograph, for example, not only excites an image, has an appearance, but, owing to its optical connexion with theobject, is evidence that that appearance corresponds to a reality.” (CP4.447; my emphasis) Following this, it is acceptable toconsider original files made by digital devices as indices. Then, what about digital composite photos? Some composite photos are sub-indices that have fictional references, therebylacking independence andphenomenological features, as Peirce described a fiction as opposed to anexternal reality. (CP 5.405) A composite photo of Hamlet, afictional character, is an example of this category whichdepends on the interpretational convention.

Not all composite photos,however, are to be regarded as sub-indices. To demonstrate this point, let us reconsiderBrian Walski’s documentary photo which was derived from two originals. Despite the alteration, it could be said that the final imageindicates the real stateof affairs at the Iraq War in 2003. Our common-sense allows us to consider the event that Walski has captured as a real, individual state of affairswhich occurred atthe time. One might say two different moments have been composited in one picture,but, following Peirce, it could be also said that “single units, singlecollections of units, or single continua” (CP 2.306) are particulars. Therefore, his digital photo can be classified as an elementaryindex just like the Pole Star and any other analogue photos. Peirce’s pragmatic realism provides a theoretical ground on which we can render Walski’s picture as an index of a certain real thing. In hisscholastic version of realism the universals are real, while the individualsare instances of reality. The state of affairs that Walski has captured can bedescribed by this statement: “An American soldier was deterring one of therefugees holding his child in his arm from standing up.” From Peirce’sviewpoint, Walski’s composite photo is an index that exemplifies the conceptreferred to by the statement, a symbol.

Manyidentify ‘indexicality’ as causalrelationship between a sign and its object. In particular, the defendersof photo-index theoryconsider photos as the evidencesof the past existence of people or facts. While their interpretation is nottotally misguided, it fails to present the whole picture regarding Peircean indices. To see the whole picture more clearly, we needto remind ourselves that when Peirce was talking about the three kinds of signs - that is, icon, index and symbol - he was mainly concerned with linguistic or symbolic signs. These were of specificinterest to him because of the function they served in his logical scheme ofsemiotics. As the particulars, the indicesindicate the general embodying the Firstness (or quality) and exemplifying the Thirdness (or concept). With this factin mind, I have explained how one can regarddigital photos as indices. In most cases, they are elementaryindices that have all the features of the indices, unless they refer to thefictional.



Atkin,Albert (2005), Peirce on the Index and Indexical Reference, Transactions of the Charles S. Peirce Society, Vol.41

Goudge, Thomas A. (1965), Peirce's Index, Transactions of the Charles S. Peirce Society, Vol.1

Kang, Mi-Jung, 2009, Habitand Meaning: A Study on C.S. Peirce’s Theory of Interpretant, Semiotic Inquiry, Vol.25

Krauss, Rosalind (1977), Note on the Index 1: SeventiesArt in America, October, Vol.3

Yoon, Joon-Sung (2004), Post-photography:Reconsideration of Photography as Contemporary Art Medium, AURA, No.11

Articles inbooks

Lee, Young-Joon (1998), Epistemological Change Contemplated on the Threshold of 21stCentury, in Photography in the DigitalEra, ed. by Joo-Suk Park, Seoul: Noonbit.

Lefebvre, Martin (2007), The Art of Pointing: On Peirce, Indexicality, andPhotographic Images, ed. James Elkins, Photographic Theory (The Art SeminarII), New York: Routlege

Lister, Martin (1995), Introductory Essay, ed. Martin Lister, The Photographic Image in Digital Culture,Routledge; translated in Korean by Seon-Ah Woo, Seoul: Vision and Language

Sebeok, Thomas A. (1995),Indexicality, Peirce and contemporary Thought, ed.Kenneth Laine Ketner, Fordham University Press

Tomas, David (1996), From the Photograph toPostphotographic Practice: Toward a Postoptical ecology of the Eye, ed. Timothy Druckrey, Electronic Culture, APERTURE foundation


Dubois, Philippe (1983), L’acte photographique, SEJER, Nathan; translated to Korean byKyung-Ryul Lee, Sajin Masil Pubilshing Co. Ltd

Mitchell, W. J. (1992), The Reconfigured Eye: Visual Truth in the Post-Photographic Era,The MIT Press.

Peirce, C.S. (1931-35), Collectedpapers of Charles Sanders Peirce, Vols. 1-6, ed. Charles Hartshorne andPaul Weiss, Harvard University Press

Peirce, C.S. (1958), Collectedpapers of Charles Sanders Peirce, Vols. 7-8, ed. Arthur Burks, HarvardUniversity Press

Peirce,C.S.(1998), The Essential Peirce: Selected PhilosophicalWritings, Vol. II, ed. Peirce Edition Project, Indiana University Press

Peirce, C.S. (1977), Semiotic and Significs: The Correspondence between Charles S. Peirceand Victoria Lady Welby, Indiana University Press

Short,T. L. (2007), Peirce'sTheory of Signs, Cambridge University Press

Internet ressources

Göran Sonesson (1989),Semiotics of Photography – On tracing the index http://www9.georgetown.edu/faculty/irvinem/theory/Sonesson-Semiotics_of_Photography.pdf


[1] Among those are Philip Dubois, RosalindKrauss, Denis Roche, Pascal Bonitzer, Henri Van Lier, Jean-Marie Schaeffer, etc. (Refer to Krauss, Dubois, and Lefebvre)

[2] Scholars who pursue this line of thoughtinclude W. J. Mitchell, MartinLister, David Tomas, Gören Sonesson, Joo-Whan Kim, Young-Joon Lee, Joon-Seong Yoon and many others.

[3] Goudge interpreted Peirce’s distinction between ‘genuine’and ‘degenerate’ indices as depending whether they have causal relations withthe objects. (55-56) His reference is Peirce’s distinction between ‘designation’and ‘reagent’ that appears in CP 8.368n. Yet, I cannot find any clue thatreagents or genuine indices are in the causal relation to their objects.

[4]An index could be a sign without any actual interpretant. Around 1905 Peirce discriminated three kinds of the interpretants such asimmediate, dynamical (i.e., actual), and final one. (SS. 111) According to him every sign has its immediateinterpretant as the potentiality for future interpretation. (refer to Short and Kang) So, we can say that everyindex has an immediate interpretant, if not actual one.

[5]A fuller account of Peircean concept ofreality would be beyond the scope of this paper. However, a few remarks seem to be needed here. For Peirce thereal is independent of interpretation, while it is conceived of as the objectof truth that should be found by the infinite community of interpreters in thelong run. This is the most controversial claim in Peirce’s pragmaticsemiotics because it doesn’t seem clear whether reality is independent of themind that interprets it or not. He convincingly solved this problem by dividingtwo kinds of objects: immediate and dynamicalobject. The one is “the Object as the Sign itself represents it, and whoseBeing is thus dependent upon the Representation of it in the Sign” and theother is “the Reality which by some means contrives to determine the Sign toits Representation.” (CP 4.536) For Peirce, the reality or dynamical object canonly be known by the inquirers’ community which would continue to pursue theinquiry, not by an individual inquirer. Consequently, it could be said thatreality is independent of the individual mind, but not of the mind in general.

[6]Although Atkin namedthis kind of indices ‘index simpliciter’, but I prefer the term, ‘elementaryindex’ as used by Sonesson. Atkin distinguishes 3 kinds of indices such asindex simpliciter, sub-index, precept, whereas Sonesson classifies indices aselementary and secondary ones. (See Atkin, 170 and Sonesson, 63) According toSonesson, secondary indices are the signs, “where the indexical relationshipsholds between objects which in themselves are signs [e.g., symbols] alreadyconstituted in other way.” I’ll consider his ‘secondary index’ as analternative term for ‘sub-index’ or ‘hypo-seme’ though Peirce didn’t used themvery often. (See CP 2.330, EP 2.274)


댓글 입력 영역