Semiotics is a system of signs and meanings; the study of signs. I don't mean like traffic lights (although those count too!), but all the ways in which people communicate with each other, and the ways in which people "read" the world. Semiotics as a modern discipline (talk about signs and meanings go way back in every culture) was formulated independently by a Swiss linguist, Ferdinand de Saussure, and an American philosopher, Charles Sanders Peirce. All the talk about signifiers and signifieds comes from Saussure's work, which discusses language as being composed of a system of arbitrary differences (ie, there is no "natural" relation between a dog, the three letters together in a unit, and the sound [dawg]).

Peirce's system was a little more complex (he has more than 100 individual sign categories: he was kind of obsessed, throughout his career, with categories and classification), but basically, he defined a sign as something which creates in the mind of another a new sign (an interpretant), while referring to an object, the whole process of which is organized against a system of meaning called the ground. Here's how he puts it:

"A sign[, or representamen,] is something which stands to somebody for something in some respect or capacity. It addresses somebody, that is, creates in the mind of that person an equivalent sign, or perhaps a more developed sign. That sign which it creates I call the interpretant of the first sign. The sign stands for something, its object. It stands for that object, not in all respects, but in reference to a sort of idea, which I have sometimes called the ground...." (CW II.135; ctd SS 14 brackets elided)

Peirce also wanted to be able to distinguish a sign by its various guises; he talks about the same sign in different ways, one of which is when it refers to its direct object. Because his classifying mechanism depends on a conceptual distinction between what he called "firstness," "secondness," and "thirdness," (also called "presence," "difference," and "representation") we have three types of signs as they refer to their direct objects: icons, indices and symbols.

An icon resembles its object in 1) sharing certain properties of the object (images, eg) and/or 2) duplicating (analogically) the organising principles of the object (diagrams, eg) Ex: photos, painting, sculpture, cinema, also graphs and algebraic equations. Icons are "the only means of directly communicating an idea." (CW II.158)

A symbol is an arbitrary, conventional and generally unmotivated relation of sign and object. Ex: natural languages and notational systems. Most representamens in practice are a combination of the three types; Peirce considers those signs the richest which combine all three modes.

So, in Janet's dream of the glass key, we can discuss the function of the key as a sign in all three modes: the dream-key is an icon in its representation as a key which unlocks the door to the house; it is an index both to the dreamer's desire to open the door (it's in her hand, its why people lock doors--to keep someone out), as well as to the fear that what is behind the door might be harmful (in its glassy aspect, but also, that that is the other reason why people lock doors--to keep someone or thing in); it is a symbol (technically, for all these reasons and more) the moment we start unpacking the image of the key in all its various modes (as we did in lecture yesterday). If this seems a little confusing, it is. The problem is that all meaning systems (I think) are basically conventional, so that any iconic or indexical relation is overdetermined by the symbolism of conventional usage (ie, a key can only unlock a door if we know that that's what that funny shaped little thing does).

Clues are something completely different, yet fundamentally the same. A clue is a sign: it represents something to the reader (we readers, the investigative "readers" of the text) which means something other than itself, and does so by reference to a ground (a system of organizing meaning-making). The problem is that a clue can only be a clue if we know what it is that we are looking for. So, any sign _can_be_ a clue, but we only _know_ if it is a clue in retrospect. How do we get from one to the other? How do we know what to look for in order to make sense of what we see or read?

Peirce had an idea about this, which he called "abduction." "Abduction...is the creative act of making up explanatory hypotheses." (Davis 22) In the Maltese Falcon, for example, Spade "makes up" two separate narratives about the murder of Miles Archer, neither of which can be completely accepted as strictly factual, and both of which can be accepted as narratively plausible (ie, they tell the story in a way which makes sense of the events). So, we are left kind of high and dry if we want to critique the story that's told to us, unless we can back off of the thing and talk about the construction of narrative as such, how a successful narrative can be made to work. And this has to do with the mastery of discourse: the only way Spade can succeed (can get off the hook) is to successfully locate himself in that position (which due to certain accidents of birth and culture, he can do far easier than, say, Harriet Vane, or Easy Rawlins).

This was also the problem on which The Long Goodbye dwelt for something like 400 pages: the mystery here is not who killed so and so, but how the narrator can work story-telling itself to make some sense of a strange and uncanny landscape (the cartography -- political and geographical -- of LA, the relations between men, and between men and women, the relation of sign-systems like language and semiotics to narrativity as such). The point I'm taking a long time to get around to is that even as the "mystery" has its narrative uses (it can allegorize certain human dynamics, for example), there are always other stories being told, and other stories that can be told. Our job as readers is to locate the one, and our job as writers is to develop the other.

Hope you have a beautiful long weekend, s.

Notes:

Works Cited:

• Raymond Chandler, The Long Goodbye (New York: Vintage, 1992 [1953])
  
• William H. Davis, Peirce's Epistemology (The Hague: Martinus Nijhoff, 1972)
  
• Teresa de Lauretis, Alice Doesn't (Bloomington: University of Indiana
• Press, 1984)
  
• Dashiell Hammett, The Maltese Falcon (New York: Vintage, 1992 [1929])
  
• The Glass Key (New York: Vintage, 1989 [1931])
  
• Earl Jackson, Jr., Strategies of Deviance (Bloomington: University of
• Indiana Press, 1995) Theories of Representation and Difference. Teresa de Lauretis, General Editor
  
• Walter Mosley, Devil in a Blue Dress (New York: Simon & Schuster, 1990)
  
• Charles Sanders Peirce, Collected Works [CW] (Bloomington: University of
• Indiana Press, 1994-6)
  
• Dorothy L. Sayers,Gaudy Night (New York: Harper, 1995 [1936])
  
• Kaja Silverman, The Subject of Semiotics [SS] (New York: Oxford University
• Press, 1983)